dataset
stringclasses
45 values
id
stringlengths
17
64
messages
listlengths
2
2
science.scientific_lay_summarisation_elife_single_doc_summ
science.scientific_lay_summarisation_elife_single_doc_summ.1583
[ { "content": "You will be presented with the abstract, introduction, and discussion section from a biomedical research article. Your task is to create a summary that a layperson can understand, capturing the essence of the research article. Keep essential scientific terms, but ensure that the language remains clear and the concepts are explained in an uncomplicated manner.\n\nTitle: Microbial consortia at steady supply\nArticle:\nAbstract:\nMetagenomics has revealed hundreds of species in almost all microbiota. In a few well-studied cases, microbial communities have been observed to coordinate their metabolic fluxes. In principle, microbes can divide tasks to reap the benefits of specialization, as in human economies. However, the benefits and stability of an economy of microbial specialists are far from obvious. Here, we physically model the population dynamics of microbes that compete for steadily supplied resources. Importantly, we explicitly model the metabolic fluxes yielding cellular biomass production under the constraint of a limited enzyme budget. We find that population dynamics generally leads to the coexistence of different metabolic types. We establish that these microbial consortia act as cartels, whereby population dynamics pins down resource concentrations at values for which no other strategy can invade. Finally, we propose that at steady supply, cartels of competing strategies automatically yield maximum biomass, thereby achieving a collective optimum.\nIntroduction:\nIn this section, we present a model for the population dynamics of cell types metabolically competing for external resources( see Figure 1). Importantly, biomass production is governed by a physical model that respects flux conservation. 10. 7554/eLife. 22644. 003Figure 1. Model for metabolically competing cell types.( A) The rate of biomass production g⁢( c1,…, cp) is a function of the internal building-block concentrations.( B) Biologically relevant growth-rate functions g⁢( c1,…, cp) are increasing with respect to ci with diminishing returns.( C) Different cell types, i. e. metabolic strategies, are defined as specific distributions of enzymes for import αi and conversion κj⁢i, subject to a finite budget.( D) Cell types( e. g. σ1 and σ2) compete for external building blocks that are steadily and homogeneously supplied in volume V. DOI: http://dx. doi. org/10. 7554/eLife. 22644. 003\nDiscussion:\nFor simplicity, we assumed linear metabolic fluxes and uniform enzymatic rates, production costs, and building-block stoichiometries. However, the emergence of optimal cartels does not rely on these assumptions. Even allowing for fluxes that are nonlinear( e. g. Michaelis-Menten) with respect to building-block concentrations, microbes must utilize their enzymes in the linear regime to be metabolically optimal: Because resources are depleted by competitive growth between metabolic classes, fluxes mediated by saturated enzymes do not limit growth. Cells can improve their growth rate by reallocating their enzyme budget from saturated enzymes to the unsaturated enzymes mediating growth-limiting linear fluxes. Moreover, independent of rates, production costs, and stoichiometries, optimal metabolic types must consist of non-overlapping trees of conversions. Indeed, the optimality of such metabolic networks, obtained from transport-network theory, only requires the linearity of metabolic fluxes with respect to enzyme concentrations. As a result, optimal metabolic types, as well as cartels, can still be enumerated. Interestingly, we discovered that distinct cartels can arise for very similar external building-block availabilities, and cartels can even merge under special conditions. In an extended model that includes fluctuations, e. g. in enzyme expression( Wang and Zhang, 2011; Kiviet et al., 2014), we expect ‘ghosts’ of these neighboring cartels associated with similar resource availabilities to persist against the background of the dominant cartel. As our primary concern is the emergence of a division of labor, we consider only relatively large populations of cells for which we can neglect stochastic population fluctuations. What relevance might our results have for real metabolic networks? Microbes regulate metabolic processes via complex networks with, e. g., multistep reaction chains and metabolic branch points( Almaas et al., 2004). However, there is evidence of optimal partitioning of enzymes in these real networks: microbes produce components of multiprotein complexes in precise proportion to their stoichiometry, whereas they produce components of functional modules differentially according to their hierarchical role( Li et al., 2014). Recent experimental studies have revealed that optimal metabolic flux partitioning is an operating principle for resource allocation in the proteome economy of the cell( Hui et al., 2015; Hermsen et al., 2015). Provided optimality considerations apply to real metabolic networks, the approach we have taken can provide insight into flux partitioning and division of labor in microbial communities. For instance, we expect that for a group of interconvertible resources that are collectively growth limiting, the expressed metabolic network should have the topological properties discussed above— no ‘futile cycles’ and no ‘convergent pathways’. Such predictions are not at odds with the existence of well-known metabolic cycles such as the TCA cycle and the GOGAT cycle because these cycles are not futile but rather are energy yielding or assimilatory, respectively. Our predictions apply directly to irreversible conversion processes, e. g via chains of reactions with committed steps, as well as to reversible chains of reactions, for which the only cycles in optimal metabolic networks are two-cycles due to reversibility. The overall acyclic nature of anabolic fluxes can be tested experimentally by measuring reaction fluxes in metabolic networks, e. g. using isotope tracers and mass spectrometry. Abiotic and biotic processes controlling resource turnover in nutrient reservoirs, such as the ocean or soil sediments, operate on many different temporal and spatial scales( Braswell et al., 1997; Whitman et al., 1998). In our framework, steady but spatially inhomogeneous supply of diffusive building blocks should lead to the tiling of space by locally dominant cartels. Because of our model cells’ ability to shape their environment, we expect sharp transitions between neighboring tiles, consisting of cartels that differ by a single metabolic class. We expect spatial tiling to emerge in real microbial communities growing in inhomogeneous conditions, e. g. in a gradostat with spatially structured nutrient supply( Lovitt and Wimpenny, 1981). In such spatial communities, the detection of well-delimited patches of resource availabilities, with specific nutrient ratios, would be evidence of spatial tiling by microbial cartels. The spatial structure of microbial communities may also reflect the extracellular division of labor. Extracellular division of labor involves metabolic pathways with obligatory external reactions, i. e. with enzymes that are public goods. In a homogeneous environment, ‘cheating’ cell types that do not produce the public good are always at an advantage and their introduction causes the collapse of the entire population. In our framework, we expect producer cartels to spatially segregate from neighboring non-producer cartels( Drescher et al., 2014). Temporally varying supply can also be addressed within our framework. For supply fluctuations on long timescales ≫1/δ( the lifetime of a cell), the population dynamics within cartels keeps resource levels fixed, whereas fluctuations on short timescales ≪1/δ are self-averaging. In practice, slow supply fluctuations can arise due to seasonal biogeochemical cycles( Schoener, 2011), while fast supply fluctuations can arise from the transient biomass release upon cell death( Yoshida et al., 2003). The effect of supply fluctuations occurring on timescales ∼1/δ, which includes day-night cycles, is more complex. Transport-network theory predicts that fluctuating resource conditions select for networks with metabolic cycles, whose structures depend on the statistics of the driving fluctuations( Katifori et al., 2010; Corson, 2010). Characterizing the benefit of cycles in such networks may well reveal new optimization principles that underlie the microbial metabolic diversity. Microbes also adjust to fluctuating conditions by switching their metabolic type via gene regulation instead of relying on population dynamics. Within our framework, to consistently implement the optimal mix of metabolic strategies, the role of sensing and regulation is then primarily to determine the relevant ‘supply sector’ by assessing the relative abundance of various resources. Thus, in principle, division of labor within a single species can lead to cartels with distinct metabolic strategies associated with distinct phenotypic states. However, the persistence of cartels requires the coexistence of all cartel strategies, which within a single species could be facilitated by cell-to-cell communication( quorum sensing). We therefore anticipate that extension of our analysis to fluctuating supply conditions may provide insight into the design principles underlying regulation and signaling in microbial communities.", "role": "user" }, { "content": "Microbes are found in virtually every environment on Earth . Like other organisms , microbes grow by using enzymes to convert nutrients into proteins , DNA and other molecules that make up their cells . Together , these chemical transformations define the “metabolism” of a microbe . In any given environment , there is almost always a diverse variety of microbes living together . Different microbes in these communities will use different combinations of enzymes to exploit the available nutrients , and members of well-studied communities have been found to work together to make the most of the nutrient source . This is remarkable because one might expect competition between microbes to select for a single “best” microbe , rather than diverse communities . The economic concept of “division of labor” suggests that if microbes divide chemical tasks between each other , they will use the available resources more efficiently . The concept provides a possible explanation for metabolic diversity amongst microbes , yet it remains to be shown whether microbial communities actually benefit from a division of labor . Here , Taillefumier et al . used mathematical models to reveal that even in a uniform environment , metabolic competition generally leads to the steady coexistence of distinct microbes , collectively called a “consortium” . In a consortium , distinct microbes organize themselves to create a community-level metabolism that best exploits the nutrients present . The models showed that while growing , a consortium depletes the available pool of nutrients to such low levels that only members of the consortium can survive . The findings suggest that the benefit of metabolic diversity stems from the ability of a consortium to automatically deplete nutrients to levels at which no other microbes can invade . Taillefumier et al . propose that consortia that arise naturally under conditions where there is a steady supply of nutrients produce the maximum mass of microbes . Future experiments that analyze the impact of fluctuating nutrient supply may help us to understand the benefit of metabolic diversity in real-world microbial communities .", "role": "assistant" } ]
science.genia_ner
science.genia_ner.113
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"RNA\", \"CellType\", \"CellLine\", \"Protein\", \"DNA\"].\n\nPlease return the output as a JSON object of the format: {\"DNA\" : [\"TAL1\", ...], \"RNA\": [\"mRNAs\", ...], \"CellLine\": [\"T cells\", ...], \"CellType\": [\"monocytes\", ...], \"Protein\": [\"IL-2\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nInhibition of human immunodeficiency virus type 1 replication by a Tat-activated, transduced interferon gene: targeted expression to human immunodeficiency virus type 1-infected cells. We have examined the feasibility of using interferon (IFN) gene transfer as a novel approach to anti-human immunodeficiency virus type 1 (HIV-1) therapy in this study. To limit expression of a transduced HIV-1 long terminal repeat (LTR) -IFNA2 (the new approved nomenclature for IFN genes is used throughout this article) hybrid gene to the HIV-1-infected cells, HIV-1 LTR was modified. Deletion of the NF-kappa B elements of the HIV-1 LTR significantly inhibited Tat-mediated transactivation in T-cell lines, as well as in a monocyte line, U937. Replacement of the NF-kappa B elements in the HIV-1 LTR by a DNA fragment derived from the 5'-flanking region of IFN-stimulated gene 15 (ISG15), containing the IFN-stimulated response element, partially restored Tat-mediated activation of LTR in T cells as well as in monocytes. Insertion of this chimeric promoter (ISG15 LTR) upstream of the human IFNA2 gene directed high levels of IFN synthesis in Tat-expressing cells, while this promoter was not responsive to tumor necrosis factor alpha-mediated activation. ISG15-LTR-IFN hybrid gene inserted into the retrovirus vector was transduced into Jurkat and U937 cells. Selected transfected clones produced low levels of IFN A (IFNA) constitutively, and their abilities to express interleukin-2 and interleukin-2 receptor upon stimulation with phytohemagglutinin and phorbol myristate acetate were retained. Enhancement of IFNA synthesis observed upon HIV-1 infection resulted in significant inhibition of HIV-1 replication for a period of at least 30 days. Virus isolated from IFNA-producing cells was able to replicate in the U937 cells but did not replicate efficiently in U937 cells transduced with the IFNA gene. These results suggest that targeting IFN synthesis to HIV-1-infected cells is an attainable goal and that autocrine IFN synthesis results in a long-lasting and permanent suppression of HIV-1 replication.", "role": "user" }, { "content": "{\"CellLine\": [\"T - cell lines\", \"U937\", \"T cells\", \"transfected clones\", \"Tat - expressing cells\", \"IFNA - producing cells\", \"monocyte line\", \"Jurkat\", \"HIV - 1 - infected cells\", \"U937 cells\"], \"CellType\": [\"the HIV - 1 - infected cells\", \"human immunodeficiency virus type 1 - infected cells\", \"T cells\", \"monocytes\", \"HIV - 1 - infected cells\"], \"DNA\": [\"human IFNA2 gene\", \"the HIV - 1 LTR\", \"ISG15\", \"HIV - 1 LTR\", \"DNA fragment\", \"ISG15 - LTR - IFN hybrid gene\", \"5\\u0027 - flanking region\", \"Tat - activated , transduced interferon gene\", \"IFN - stimulated response element\", \"IFNA gene\", \"ISG15 LTR\", \"LTR\", \"IFN - stimulated gene 15\", \"IFN genes\", \"hybrid gene\", \"HIV - 1 long terminal repeat (LTR) - IFNA2\", \"chimeric promoter\", \"retrovirus vector\", \"NF - kappa B elements\"], \"Protein\": [\"IFN A\", \"phytohemagglutinin\", \"IFN\", \"interleukin - 2 receptor\", \"tumor necrosis factor alpha\", \"IFNA\", \"Tat\", \"NF - kappa B\", \"interleukin - 2\", \"interferon\"], \"RNA\": []}", "role": "assistant" } ]
science.acl_arc_intent_classification
science.acl_arc_intent_classification.1236
[ { "content": "You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories:\n- Background: provides context or foundational information related to the topic.\n- Extends: builds upon the cited work.\n- Uses: applies the methods or findings of the cited work.\n- Motivation: cites the work as inspiration or rationale for the research.\n- CompareOrContrast: compares or contrasts the cited work with others.\n- FutureWork: cites the work as a direction for future research.\n\nYour answer should be a single word from the following list of options: [\"Background\", \"Extends\", \"Uses\", \"Motivation\", \"CompareOrContrast\", \"FutureWork\"]. Do not include any other text in your response.\n\nSection Title:\nintroduction\n\nContext before the citation:\n• language learning (Green 1979; Mori and Moeser 1983; Morgan, Meier, and Newport 1989) • monolingual grammar induction (Juola 1998) • grammar optimization (Juola 1994) • insights into universal grammar (Juola 1998) • machine translation (Juola 1994, 1997; Veale and Way 1997; Gough, Way, and Hearne 2002) \u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000 The marker hypothesis has been used for a number of different language-related tasks, including \u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000 The marker hypothesis is arguably universal in presuming that concepts and structures like these have similar morphological or structural marking in all languages.\n\nCitation Sentence:\n• language learning ( Green 1979 ; Mori and Moeser 1983 ; Morgan , Meier , and Newport 1989 ) • monolingual grammar induction ( Juola 1998 ) • grammar optimization ( Juola 1994 ) • insights into universal grammar ( Juola 1998 ) • machine translation ( Juola 1994 , 1997 ; Veale and Way 1997 ; Gough , Way , and Hearne 2002 )\n\nContext after the citation:\nWith respect to translation, a potential problem in using the marker hypothesis is that some languages do not have marker words such as articles, for instance. Green’s (1979) work showed that artificial languages, both with and without specific marker words, may be learned more accurately and quickly if such psycholinguistic cues exist. The \u0000\u0000 research of Mori and Moeser (1983) showed a similar effect due to case marking on pseudowords in such artificial languages, and Morgan, Meier, and Newport (1989) demonstrated that languages that do not permit pronouns as substitutes for phrases also provide evidence in favor of the marker hypothesis.", "role": "user" }, { "content": "Background", "role": "assistant" } ]
science.drug_combo_extraction_re
science.drug_combo_extraction_re.534
[ { "content": "You will be shown a paragraph from a biomedical research article. Your task is to extract all drug combinations appearing in a specific sentence within the paragraph. Each drug combination should be assigned to one of the following three categories:\n\n- POS: Indicates a positive interaction between the drugs in the combination; for instance, drugs that have an additive or synergistic effect.\n- NEG: Indicates a negative interaction between the drugs in the combination; for instance, combining these drugs might result in adverse side effects, or might reduce their efficacy.\n- COMB: Indicates that the drugs are given in combination, but the type of interaction between the drugs cannot be determined based on the provided context.\n\nPlease format your answer as a JSON array. Each entry in the array should itself be an array formatted as [<drug_1>, <drug_2>, ..., <drug_n>, <category>] where <category> is one of the three categories listed above and <drug_1>, ..., <drug_n> are drug names copied verbatim from the input sentence. Different combinations may involve different numbers of drugs; in other words, the `n` in <drug_n> may vary.\n\nAs an example, the following output describes two extracted drug combinations; one between two drugs, and one between three drugs:\n\n[[\"artesunate\", \"artemether\", \"COMB\"], [\"Nifedipine\", \"Labetalol\", \"Prazosin\", \"POS\"]]\n\nThe paragraph and specific sentence are below. You should only extract relationships expressed in the specific sentence; the paragraph is provided for context.\n\nIn your output, return only the JSON array and no other text. If no drug combinations appear, return an empty array `[]`.\n\nparagraph: Evaluation of irinotecan in combination with 5-fluorouracil or etoposide in xenograft models of colon adenocarcinoma and rhabdomyosarcoma. irinotecan [7-ethyl-10-(4-[1-piperidino]-1-piperidino)-carbonyloxy-camptothec in] administered i.v. in two courses, each course consisting of administration every day for 5 days [(dx5)2] on days 1-5 and 8-12, has demonstrated significant activity against advanced human tumor xenografts derived from colon adenocarcinomas and several childhood cancers. To build on this therapy, we have evaluated the combination of irinotecan given on this schedule with 5-fluorouracil given on days 1, 7, and 14 with or without leucovorin [(dx5)3 i.v.] against colon tumors, or combined with etoposide administered (dx5)2 i.v. either 2 h before or 2 h after irinotecan for treatment of colon tumors and rhabdomyosarcomas. A combination of 5-fluorouracil at 75% and irinotecan at 50% of their respective maximum tolerated doses when administered as single agents on this schedule gave acceptable toxicity. Against colon adenocarcinoma xenografts, 5-fluorouracil did not enhance the response rate compared with that obtained with the optimum dose of irinotecan given as a single agent. Against GC3/TK- xenografts, which lack thymidine kinase and cannot salvage thymidine to circumvent the inhibition of thymidylate synthase, the addition of leucovorin to the combination increased the complete response rate from 10 to >90%, whereas the response rates for the optimal doses of irinotecan or 5-fluorouracil, as single agents, were 30 and <10%, respectively. etoposide d x 5 i.v. for two or three courses or (d x 5)3 p.o. did not cause objective regression of any colon tumors. In contrast, three of five rhabdomyosarcoma lines demonstrated a high frequency of partial regressions or complete regressions when treated (d x 5)1 i.v. Repetitive courses [e.g., (d x 5)2 or (d x 5)3] i.v. or p.o. or by 4-h infusion d x3 i.v. were either equally effective or less effective. Irinotecan and etoposide were combined using the ( d x 5)2 i.v . schedule for both drugs , in which irinotecan was given 2 h before or 2 h after the administration of etoposide . Each drug could be combined at only 38% of its respective maximum tolerated dose when administered as a single agent, indicating greater than additive toxicity. Toxicity was similar irrespective of the sequence of administration and was manifested by loss of weight (73% of the initial weight, nadir day 7), myelosuppression, and prolonged thrombocytopenia. The responses of colon carcinomas to the combination given in either sequence were similar to that achieved with irinotecan given alone at the same dose as used in the combination. Similarly, when etoposide was given before irinotecan, the responses of rhabdomyosarcomas were similar to those for irinotecan. However, in experiments in which etoposide was administered 2 h after each dose of irinotecan, there was significant antagonism of the antitumor activity of irinotecan.\n\nsentence: Irinotecan and etoposide were combined using the ( d x 5)2 i.v . schedule for both drugs , in which irinotecan was given 2 h before or 2 h after the administration of etoposide .", "role": "user" }, { "content": "[[\"Irinotecan\", \"etoposide\", \"NEG\"], [\"irinotecan\", \"etoposide\", \"NEG\"]]", "role": "assistant" } ]
science.scientific_papers_summarization_single_doc_arxiv
science.scientific_papers_summarization_single_doc_arxiv.25749
[ { "content": "Write an abstract for the research paper below. Your response should only include the abstract and no additional text.\n\nArticle text:\nthe evolution of late - type stars along the asymptotic giant branch ( agb ) is accompanied by significant mass loss to the circumstellar environment with mass - loss rates of up to @xmath3m@xmath4/year ( see e.g. , jura & kleinmann 1990 , kemball & diamond 1997 ) .\nagb stars with masses of the order of 1m@xmath4 become unstable to large - amplitude radial pulsations and become mira variable stars .\nthe nature of the mass - loss process of agb stars , and especially its connection with the pulsation mechanism in the case of mira variable stars , is not well understood .\njoint multi - wavelength studies of the stellar surface and the circumstellar environment at different distances from the stellar surface promise to lead to a better understanding of the nature of the mass - loss process , and thus of the evolution of agb stars and of the chemical enrichment process of galaxies .\nfigure 1 shows a scheme of the multi - wavelength view of a mira star and its circumstellar environment .\nthe conditions on the stellar surface can in principle be studied by means of optical / near - infrared long - baseline interferometry .\nthis technique provides information regarding the stellar diameter , effective temperature , and center - to - limb intensity variations . with multiple epochs ,\nthe variability of these parameters as a function of time / stellar pulsation cycle can be studied as well .\ninterferometry at a more advanced stage can likely probe horizontal temperature inhomogeneities of the stellar surfaces as well ( e.g. wittkowski et al . 2002 ) .\nrecent results on mira stars based on near - infrared interferometry are described by e.g. thompson et al .\n( 2002a / b ) .\nthe eso very large telescope interferometer ( vlti ) provides excellent capabilities to observe agb stars at near - infrared and mid - infrared wavelengths .\nit is currently in the phase of commissioning .\nabout 20 mira stars have already been observed with the near - infrared commissioning instrument vinci ( richichi & wittkowski 2003 ) . the structure and dynamics of the circumstellar atmosphere of mira stars at typically 23 stellar radii can be probed by mapping the sio maser emission towards these stars using very long baseline interferometry ( vlbi ) at radio wavelengths ( e.g. , kemball & diamond 1997 , boboltz et al .\n1997 , boboltz & marvel 2000 , boboltz 2003 ) .\nthe dust formation zone at distances of typically 410 stellar radii can be probed by mid - infrared interferometry ( see , e.g. danchi et al .\n1994 , townes 2003 ) .\nthe distance of the sio maser emission relative to the dust zone has been studied by greenhill et al .\n( 1995 ) . here\n, we report on first preliminary results of a recently started project to combine observations of the sio maser emission towards mira stars using the vla / vlba and observations of the near - infrared stellar disk using the vlti , focusing on the mira star sori .\nsori is a relatively cool mira star with spectral type m6.5e m9.5e and relatively large visual amplitude of @xmath5@xmath0 713 .\nmerchan benitez & jurado vargas ( 2000 ) report on a marked decrease of the stellar variability period , which might be caused by a recent helium - shell flash .\nthe variability period is currently @xmath0440 days .\nthe distance to sori was estimated by van belle ( 1996 ) to 422@xmath637pc ( based on wyatt & cahn 1993 and young 1995 ) .\nthere is one previous diameter measurement of sori by van belle et al .\n( 1996 ) , who derived a uniform disk ( ud ) diameter of 10.54@xmath60.68pc at stellar pulsation phase 0.56 .\nsio and oh maser emission towards this star has been reported .\nan interferometric map of this emission has so far not been obtained .\nwe have chosen the object sori for our study because it is known to show a fairly strong ( 10 - 20 jy ) sio maser emission , it has a size and k magnitude which is feasible for observations with the vlti and the 40 cm test siderostats , and it was available for both facilities , vlti and vlba .\n@xmath7-band interferometric data on sori were obtained on 14 days between january and march 2003 , corresponding to stellar phases @xmath00.77 0.92 , using the vlti with the commissioning instrument vinci , the 40 cm test siderostats , and a baseline of 24 m . these observations have been coordinated with vlba observations of the sio maser emission towards our source , which were obtained on dec . 29 , 2002 , at stellar phase @xmath00.7 . these observations were carried out for two maser transitions , namely @xmath8 , 42.8ghz , and @xmath9 , 43.1ghz .\nfigure 2 shows two examples of our preliminary vlti visibility curves for sori , for dates january 25 and march 17 .\nthey show that a uniform disk model fits our data .\nthere is no obvious indication of deviations from a uniform disk model at these spatial frequencies which might in principle be caused by contributions from a circumstellar envelope .\nfigure 3 shows our obtained near - infrared @xmath7-band ud diameters as a function of date and stellar variability phase . indicated\nare also the ud diameter measurement by van belle ( 1996 ) , and the date of our coordinated vlba measurements .\nour ud diameter values suggest a decrease of the ud diameter over our time period , which is consistent with the stellar phase .\nthe dotted and solid lines indicate best fits of a linear function and a sine function , respectively .\nfigure 5 shows our obtained vlba maps of the sio maser emissions towards sori .\nboth transitions show a clumpy distribution within a ring - like structure . indicated\nare also our obtained near - infrared @xmath7-band ud diameter at the phase of the vlba map and an approximate mean distance of the sio maser emission spots to the center of the star .\nthe relative positions of the near - infrared diameter and the sio maser emission are arbitrary .\nthe emission from both sio maser transmissions appear to be at a distance of @xmath02 stellar radii ( @xmath10@xmath010.7mas@xmath04.5au ; @xmath11@xmath020mas@xmath08.4au ) .\nthe emissions from the two sio maser transmissions ( 42.8ghz and 43.1ghz ) appear to have a separation of @xmath120.4au , which is consistent with predictions by gray & humphreys ( 2000 ) .\nfuture work on this project aims at obtaining near - infrared visibility values at more stellar phases and spatial frequencies to confirm the ud diameter change with stellar variability phase and to constrain the intensity profile , respectively .\nmulti - epoch vlba maps would allow us to correlate the near - infrared diameter change with changes of the distribution of the sio maser emission .\nmid - infrared interferometric measurements could in addition be used to probe the distribution and distance of the dust formation shell .\nboboltz , diamond , kemball , 1997 , apj 487 ,\nl147 boboltz & marvel , 2000 , apj 545 , l149 boboltz , 2003 , this volume danchi et al . , 1994 , aj 107 , 1469 greenhill et al . , 1995 , apj 449 , 365 hofmann , scholz , wood , a&a 339 , 846 jura & kleinmann , 1990 , apjs 73 , 769 kemball & diamond , 1997 , apj 481 , l111 merchn bentez p. , jurado vargas m. , 2002 , a&a 386 , 244 thompson , creech - eakman , akeson , 2002a , apj .\n570 , 373 - 378 thompson , creech - eakman , van belle , 2002b , apj 577 , 447 townes , 2003 , in reviews of modern astronomy 16 , e. schilicke ed .\nrichichi , wittkowski , 2002 , ap&ss , in press van belle , dyck , benson , lacasse , 1996 , aj 112 , 2147 wittkowski , schller , hubrig , posselt , von der lhe , 2002 , an 323 , 241 wittkowski , aufdenberg , kervella , 2003 , submitted to a&a wyatt , cahn , 1983 , apj 275 , 225 young , 1995 , apj 445 , 872", "role": "user" }, { "content": "we present preliminary results on a recently started project to perform coordinated observations of mira stars using near - infrared and radio long - baseline interferometry . \n we concentrate on recent observations of the mira star sori . \n observations with the eso very large telescope interferometer ( vlti ) were performed to measure the near - infrared diameter of the stellar surface . \n concurrent vlba observations of sio maser emission towards this mira star were performed to probe the structure and dynamics of the circumstellar atmosphere . \n our near - infrared measurements suggest a diameter change consistent with the photometric period . \n the sio maser emission appears to be at a distance of @xmath02 stellar radii ( @xmath110.7mas , @xmath220mas , both at stellar phase @xmath00.7 ) , and to show a clumpy distribution within a ring - like structure .", "role": "assistant" } ]
science.chemprot_ner
science.chemprot_ner.269
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\", \"Protein\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\" : [\"Dexamethasone\", ...], \"Protein\" : [\"BRCA-1\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nCASK is a new intracellular modulator of P2X3 receptors.\nATP-gated P2X3 receptors of sensory ganglion neurons are important transducers of painful stimuli and are modulated by extracellular algogenic substances, via changes in the receptor phosphorylation state. The present study investigated the role of calcium/calmodulin-dependent serine protein kinase CASK in interacting and controlling P2X3 receptor expression and function in mouse trigeminal ganglia. Most ganglion neurons in situ or in culture co-expressed P2X3 and CASK. CASK was immunoprecipitated with P2X3 receptors from trigeminal ganglia and from P2X3/CASK-cotransfected HEK cells. Recombinant P2X3/CASK expression in HEK cells increased serine phosphorylation of P2X3 receptors, typically associated with receptor upregulation. CASK deletion mutants also enhanced P2X3 subunit expression. After silencing CASK, cell surface P2X3 receptor expression was decreased, which is consistent with depressed P2X3 currents. The reduction of P2X3 expression levels was reversed by the proteasomal inhibitor MG-132. Moreover, neuronal CASK/P2X3 interaction was upregulated by NGF signaling and downregulated by P2X3 agonist-induced desensitization. These data suggest a novel interaction between CASK and P2X3 receptors with positive outcome for receptor stability and function. As CASK-mediated control of P2X3 receptors was dependent on the receptor activation state, CASK represents an intracellular gateway to regulate purinergic nociceptive signaling. This article is protected by copyright. All rights reserved.", "role": "user" }, { "content": "{\"Chemical\": [\"MG-132\", \"calcium\", \"ATP\", \"serine\"], \"Protein\": [\"calcium/calmodulin-dependent serine protein kinase\", \"P2X3\", \"NGF\", \"P2X3 receptors\", \"P2X3 receptor\", \"ATP-gated P2X3 receptors\", \"CASK\"]}", "role": "assistant" } ]
science.chemsum_single_document_summarization
science.chemsum_single_document_summarization.2300
[ { "content": "You will be shown the title of a chemistry paper, together with sections excerpted from the body of the paper. The paper sections may include HTML tags like `<p>` and `</p>`; please ignore these. Your task is to write an abstract for the paper. Your response should include the abstract and no other text.\n\nTitle:\nPhosphinate-containing rhodol and fluorescein scaffolds for the development of bioprobes\\xe2\\x80\\xa0\n\nPaper sections:\n\n<p>Small molecule-based fluorescent probes represent attractive alternatives to fluorescent proteins for microscopy studies.1–3 Nonetheless, there is a need for the further development of red-shifted, photostable probes in order to enable the further development of robust bioprobes for imaging experiments. Towards this goal, recent efforts have focused on the introduction of heteroatoms into the bridging position of xanthene scaffolds.4–18 Our lab's approach to this problem involves replacement of the bridging oxygen atom in the rhodamine scaffold with a phosphinate functionality (Fig. 1).19 The resulting Nebraska Red (NR) fluorophores retain the brightness and photostability of the rhodamine scaffold, while displaying >110 nm shifts in excitation and emission. Furthermore, the unique chemistry of the phosphinate functional group allows for the construction of new enzymatic probes,20 self-reporting small molecule delivery platforms,19 and provides inspiration for the development of new approaches for obtaining ratiometric probes.21 Building upon this, we sought to further increase the chemical functionality of NR dyes, through the synthesis and evaluation of rhodol- and fluorescein-based NR derivatives (Fig. 1). Importantly the phenol group present in these rhodol and fluorescein dyes provides a new chemical handle that can be leveraged to construct sensitive, red-shifted enzyme-activated probes for cellular imaging studies.22–25</p><p>Towards this goal, we first investigated synthetic routes capable of efficiently yielding rhodol and fluorescein NR derivatives. When compared to the synthesis of classical rhodamine dyes, insertion of phosphorous at the bridging position decreases the yield of the resulting fluorophore.9,19,26 Additional synthetic challenges arise in the case of fluorescein (due to the increased electron withdrawing capacity of the phenol group)10 and rhodol (due to the need to construct an asymmetric structure).27,28 During our investigation of different synthetic routes, the Yamaguchi group reported the direct conversion of phosphine oxide-containing rhodamines to rhodol and fluorescein via a hydrolytic deamination reaction under basic conditions (pH > 10).29 As an initial examination of this route in the context of NR dyes, we incubated NR700, a phosphinate ester-based tetramethylrhodamine derivative, with 0.1 M NaOH. After 3 h the products of the reaction were purified by HPLC, yielding the hydrolysis product (NR666) and the corresponding rhodol product (NR630, Fig. S1 and Scheme S1, ESI†). Upon incubation of NR700 with a 1 M solution of NaOH, instead of conversion to fluorescein we observed a rapid decrease in the color of the solution. This is likely due to attack at the C-9 position of the fluorophore as observed previously (Scheme S1, ESI†).29 In support of this hypothesis, we observed a restoration of the solution's color upon addition of 2 M HCl. Nonetheless, we were very encouraged by the potential of this route to afford rhodol and fluorescein NR derivatives.</p><p>In order to protect the C-9 position from nucleophilic attack, we adopted previous strategies for introduction of bulky groups into xanthenes,29,30 incorporating dimethyl or dimethoxy groups at the C-2′ and C-6′ of our NR dyes (Fig. 1). Our previously reported route to NR rhodamine analogues, relied on the condensation of 3-bromo-N,N-dimethylaniline with benzaldehydes19,21 or the use of a xanthone precursor.10,12,19,31 However, as mentioned above, these routes result in relatively low yields of the target fluorophores. Building upon a previously described route for phosphine oxide fluorophores, we found that yields of NR rhodamines could be significantly increased by first treating 3-bromo-N,N-dimethylaniline with sec-butyllithium and exposing the resulting lithiated reagent to ethyl dichlorophosphite.26 Subsequent oxidation with a 50% H2O2 solution yielded the phosphinate-bridged intermediate (2, Scheme 1a). A subsequent Vilsmeier–Haack reaction was used to generate the key intermediate 3. Reaction of 3 with the appropriate lithiated 2,6-disubstituted phenyl reagents (dimethyl or dimethoxy) generated the corresponding secondary alcohol, which was directly treated with 2 M HCl under reflux overnight. Finally, p-chloranil was used to generate the NR rhodamine analogues containing bulky substituents proximal to the C-9 position. Compared to our previously reported routes,19 the newly developed intermediate 3 is accessible in relatively higher yields, leading to a ≥4-fold increase in the yield of NR rhodamines.</p><p>With the NR rhodamine analogues in-hand, we reassessed the ability to directly generate rhodol and fluorescein derivatives upon treatment with NaOH. Upon exposure to 0.1 M NaOH, the color of the solution quickly changed from green to purple, indicating the generation of NR rhodol (Scheme 1b). Monitoring by HPLC indicated that the reaction was complete after 4 h. As expected, the generation of NR fluorescein required a longer reaction time (3 days) and a higher concentration of base (1 M).29 After purification by HPLC, the structures of all six NR derivatives were confirmed by NMR and mass spectrometry (see ESI†).</p><p>The photophysical properties of the NR rhodamine, rhodol, and fluorescein derivatives were assessed in PBS (10 mM, pH = 7.4) (Fig. 2a–c). The NR rhodamine derivatives showed the longest absorption and emission wavelengths, similar to our previously reported analogues.19 However, the NR rhodol and NR fluorescein derivatives displayed 35 and 70 nm blue-shifts, respectively relative to NR rhodamine. Comparison to the spectral properties of Tokyo Green,32 containing oxygen at the bridging position, indicated that introduction of the phosphinate in NR fluorescein resulted in a ≥109 nm red-shift. Among the NR family members, the dimethoxy analogues were slightly red-shifted compared to their dimethyl counterparts, while the dimethyl NR analogues displayed relatively higher brightness (ε × ϕ) with the fluorescein dimethyl derivative (NR600) displaying the overall the highest brightness. These NR dyes displayed blue-shifted excitation and emission maxima as well as increased brightness compared to previously reported phosphine oxide-based fluorophores (Table S1, ESI†).10,29</p><p>Next, we investigated the equilibrium between protonated and deprotonated forms of the rhodol and fluorescein derivatives. Absorption of the NR rhodol and NR fluorescein dyes at varying pHs was recorded (Fig. S2, ESI†). From this data, the NR rhodol derivatives exhibited a pKa of ~5.2 while the pKas for the NR fluorescein derivatives was ~6.7. Last, the photostability of NR family members was compared to Cy 5.5 using continuous irradiation for 1 h (Fig. 2d). Gratifyingly, all six NR dyes showed virtually no evidence of photobleaching, compared to a 50% decrease in Cy 5.5 fluorescence under the same conditions. Thus, these new NR derivatives provide a robust, tuneable palette of fluorescent reagents with emissions ranging from the far-red to near-infrared (NIR) region (Fig. 2e).</p><p>To highlight to utility of these new NR family members, we sought to develop a sensitive bioprobe for imaging enzymatic activity. For this application, we chose the dimethyl phosphinate-based fluorescein NR600, due to its relative brightness. Importantly, previous work has clearly demonstrated the utility of acetoxymethyl (AM) ether groups for masking phenol functionalities within fluorophores.23–25,33 The resulting AM-protected probes display low background fluorescence, high chemical stability, and are readily cleaved by cellular esterases producing a turn-on fluorescent probe for esterase activity. Based on our previous results, we also anticipated that the negatively charged phosphinate of NR600 would prevent the fluorophore from entering cells.19 Thus, we envisioned a diAM-protected version of NR600, termed diAM-NR600 (Fig. 3a). To examine the chemical stability of diAM-NR600, an in vitro assay was conducted by monitoring the fluorescence generated by diAM-NR600 over a period of 1 h (Fig. 3b). In the presence of pig liver esterase (PLE) we observed a clear, 13.8-fold increase in fluorescence that stabilized after 10 min. However, in the presence of PBS alone virtually no increase in fluorescence was observed. Moreover, exposure to BSA or cell culture media containing 10% FBS produced relatively low background fluorescent signal (2.2- and 2.9-fold respectively). These data clearly indicate that diAM-NR600 can act as an esterase substrate, forming NR600 as a fluorescent product.</p><p>To interrogate the ability of diAM-NR600 to act as a no wash probe for cellular esterase activity, HeLa cells were incubated with the esterase probe and Cell Tracker Green CMFDA, in order to verify localization. Cells were then imaged, without washing, using confocal microscopy (Fig. 4a). Indeed, the formation of NR600 in the cytosol could clearly be observed within cells (Fig. S3a, ESI†), yielding a 10-fold increase in signal over background (Fig. 4b). Similar signal to noise was obtained in HeLa cells that were washed prior to imaging (Fig. S3b, ESI†). In addition, formation of NR600 was also observed in the cytosol of NIH-3T3 and RAW 264.7 cells, indicating the generality of diAM-NR600 (Fig. S4 and S5, ESI†). Decreased nuclear localization of NR600 was consistently observed relative to Cell Tracker Green CMFDA (a fluorescein derivative), indicating a potential difference in the nuclear accumulation of these dyes. Toxicity assays also clearly demonstrated that diAM-NR600 was not toxic at concentrations up to 10 mM (Fig. S6, ESI†). These experiments verify the ability of diAM-NR600 to act as a far-red emitting, no-wash probe for cellular esterase activity.</p><p>In summary, we have broadened the color palette of our NR dyes to include rhodol and fluorescein derivatives. The resulting NR rhodol and fluorescein dyes display significantly red-shifted fluorescence relative to their oxygen analogues and display increased photostability compared to commonly used labels such as Cy 5.5. The new chemical functionality afforded by these scaffolds enabled the development of a red-shifted, no-wash probe for cellular esterase activity. Current efforts in our lab are focused on lowering the pKa of the NR fluorescein series to further increase the brightness at physiological pH. These new NR dye scaffolds also provide the starting point for the development of additional probes for enzymatic activity,34 reagents for advanced imaging experiments,35–37 and self-reporting delivery reagents.19</p><p>We gratefully acknowledge the Morrison Microscopy Core Research Facility and Prof. Christian Elowsky for assistance with confocal fluorescence microscopy as well as Prof. Edward Harris for use of cell culture equipment. We also thank the Research Instrumentation/NMR facility and the Nebraska Center for Mass Spectrometry for assistance with characterization of new compounds. This work was funded by the NIH (R35GM119751), NSF (CHE-1757957), and the University of Nebraska-Lincoln. The content of this work is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.</p><p>Electronic supplementary information (ESI) available: Experimental details, synthesis and characterization of new compounds, and supplementary figures. See DOI: 10.1039/c9cc02492a</p><p>Conflicts of interest</p><p>The authors declare no conflicts of interest.</p><p>Structures of xanthene-based Nebraska Red family members.</p><p>Photophysical properties and pKas of NR family members. All experiments were conducted in PBS buffer (10 mM, pH = 7.4 containing 1% DMSO). (a) Normalized absorption spectra. (b) Normalized emission spectra. (c) Summary of optical parameters including previously reported values for Tokyo Magenta5 and Tokyo Green.32 (d) Photostability in comparison to Cy 5.5. (e) Images corresponding to each dye solution. From top to bottom: white-light image, fluorescent image under a UV lamp (254 nm), and a NIR image with a 680 nm filter. From left to right: NR675, NR669, NR638, NR632, NR604, and NR600 in PBS buffer as well as NR604 and NR600 in buffer at pH = 4.0.</p><p>(a) Synthesis of the esterase sensor diAM-NR600. (b) In vitro stability of diAM-NR600 (5 μM) in PBS (10 mM, pH = 7.4) alone or in the presence of BSA (0.2 units) or PLE (0.2 unit). The fluorescence of diAM-NR600 in cell culture media (DMEM with 10% FBS) is also shown. Fluorescence emission at 619 nm was monitored using excitation at 600 nm.</p><p>(a) Confocal fluorescence microscopy imaging of HeLa cells incubated with 0.6 mM Cell Tracker Green CMFDA for 20 min followed by 0.6 μM diAM-NR600 for 30 min (Scale bar: 20 mm). (b) Quantification of fluorescence intensity from panel a (yellow line in red channel).</p><p>(a) Synthesis of phosphinate-based NR rhodamine analogues. (b) Direct transformation of NR rhodamine to NR rhodol or NR fluorescein.</p>", "role": "user" }, { "content": "A series of phosphinate-containing rhodol and fluorescein dyes are disclosed. These new fluorophores increase the color palette of phosphinate-based xanthenes in the far-red spectral region. The new chemical functionality of these scaffolds is leveraged to produce a sensitive, no-wash imaging probe for cellular esterase activity. The reported phosphinate-containing dyes provide platforms for the further development of imaging probes and self-reporting delivery vehicles.", "role": "assistant" } ]
science.qasa_abstractive_qa
science.qasa_abstractive_qa.1358
[ { "content": "You will be shown an excerpt from a computer science research paper, followed by a question about the paper. Please answer the question given the context. Do not include any text in your response other than the answer.\n\nContext: Multimodal data has become extremely important for robotics, due bothto the advent of new sensors such as the Kinect and the application ofrobots to more challenging tasks which require multiple modalities ofinformation to perform well. However, it can be very difficult todesign featureswhich do a good job of integrating many modalities. Whileour work focuses on color, depth, and surface normals as input modes,our structured multimodal regularization algorithm might also be applied toothers.This approach could improve performance while allowing roboticists to focus onother engineering challenges. Multimodal Deep Learning: Recent works in deep learning have extended these methods to handlemultiple modalities of input data, such as audio and video [43],text and image data [61], and even RGB-D data[59, 3].However, all of these approaches have fallen intotwo camps - either learning completely separate low-level features foreach modality [43, 61], or simply concatenating the modalities [59, 3].The former approaches have proven effective fordata where the basic modalities differ significantly, such as the aforementionedcase of text and images, while the latter is more effective in cases where themodalities are more similar, such as RGB-D data.\n\nQuestion: What does it mean to be multimodal in the context of \"multimodal inputs\"?", "role": "user" }, { "content": "It means that the system is able to handle multiple modalities of input data, such as audio and video, text and image data, and even RGB-D data; challenging tasks which require multiple modalities of information to perform well.", "role": "assistant" } ]
science.scitldr_aic
science.scitldr_aic.755
[ { "content": "Below are the Abstract, Introduction, and Conclusion sections of a computer science research paper. Please summarize the main contribution of the work in a single sentence. Your response should include the summary sentence and no additional text.\n\nPaper text:\n\nThe high-quality node embeddings learned from the Graph Neural Networks (GNNs) have been applied to a wide range of node-based applications and some of them have achieved state-of-the-art (SOTA) performance.\nHowever, when applying node embeddings learned from GNNs to generate graph embeddings, the scalar node representation may not suffice to preserve the node/graph properties efficiently, resulting in sub-optimal graph embeddings.\n\n\nInspired by the Capsule Neural Network (CapsNet), we propose the Capsule Graph Neural Network (CapsGNN), which adopts the concept of capsules to address the weakness in existing GNN-based graph embeddings algorithms.\nBy extracting node features in the form of capsules, routing mechanism can be utilized to capture important information at the graph level.\nAs a result, our model generates multiple embeddings for each graph to capture graph properties from different aspects.\nThe attention module incorporated in CapsGNN is used to tackle graphs with various sizes which also enables the model to focus on critical parts of the graphs.\n\n\nOur extensive evaluations with 10 graph-structured datasets demonstrate that CapsGNN has a powerful mechanism that operates to capture macroscopic properties of the whole graph by data-driven.\nIt outperforms other SOTA techniques on several graph classification tasks, by virtue of the new instrument.\nGNN is a general type of deep-learning architectures that can be directly applied to structured data.\nThese architectures are mainly generalized from other well-established deep-learning models like CNN BID9 and RNN BID12 .\nIn this paper, we mainly focus on Convolution-based Graph Neural Networks which attract increasing interest recently.\nConvolution operation can be embedded into Graph Neural Networks from spectral or spatial perspective.\nBID1 defines the convolution operation in the Fourier domain which needs to calculate the eigendecomposition of the graph Laplacian.\nThis method is computationally expensive and the filters they defined are non-spatially localized.\nLater, BID4 introduces Chebyshev expansion of the graph Laplacian to avoid computing eigenvectors and BID8 proposes to do convolution within 1-step neighbor nodes to reduce the complexity.\nFrom the spatial perspective, BID3 and propose to define a node receptive-field and do convolution within this field during which the information of each node as well as their neighbor nodes is gathered and new representation of each node is generated through an activation function.\nBoth of these two perspectives perform well in node representation learning and a number of variants BID20 are developed based on the convolution idea and some of them have proven to achieve SOTA in various tasks.The success of GNN in node representation learning has inspired many deep-learning-based approaches to leverage on node embeddings extracted from GNN to generate graph embeddings for graph-based applications.\nHowever, during this procedure, the learned representation of each node will be considered as multiple individual scalar features instead of one vector.\nFor example, applies element-wise max-pooling to nodes embeddings when generating graph embeddings, BID22 generates graph embeddings by computing the element-wise covariance of all nodes.\nThese operations indicate that the authors capture node features in the form of scalar when they generate graph embeddings which may not suffice to preserve the node/graph properties efficiently.To build high-quality graph embeddings, it is important to not only detect the presence of different structures around each node but also preserve their detailed properties such as position, direction, connection, etc.\nHowever, encoding these properties information in the form of scalar means activating elements in a vector one-by-one which is exponentially less efficient than encoding them with distributed representations.\nThis has been identified discussed in BID16 .\nInspired by CapsNet, we propose to extend scalar to vector during the procedure of applying GNN to graph representation learning.\nCompared with scalar-based neural network, vector-based neural network preserves the information of node/graph properties more efficiently.\nThe technique for extracting features in the form of vectors is proposed in BID5 and improved in BID16 and BID6 .\nThis technique is mainly devised for image processing.\nIn their work, the extracted vector is referred to as capsule (a group of neurons in neural network), so we follow the same notation in our work.\nIntroducing capsules allows us to use routing mechanism to generate high-level features which we believe is a more efficient way for features encoding.\nCompared with max-pooling in CNN in which all information will be dropped except for the most active one, routing preserves all the information from low-level capsules and routes them to the closest high-level capsules.\nBesides, this allows to model each graph with multiple embeddings and each embedding reflects different properties of the graph.\nThis is more representative than only one embedding used in other scalar-based approaches.In this paper, we propose Capsule Graph Neural Network (CapsGNN), a novel deep learning architecture, which is inspired by CapsNet and uses node features extracted from GNN to generate high-quality graph embeddings.\nIn this architecture, each graph is represented as multiple embeddings and each embedding reflects the graph properties from different aspects.\nMore specifically, basic node features are extracted in the form of capsules through GNN and routing mechanism is applied to generate high-level graph capsules as well as class capsules.\nIn the procedure of generating graph capsules, an Attention Module can be applied to tackle graphs in various sizes.\nIt also assigns different weights to each capsule of each node so that this model focuses on critical parts of the graph.\nWe validate the performance of generated graph embeddings on classification task over 5 biological datasets and 5 social datasets.\nCapsGNN achieves SOTA performance on 6 out of 10 benchmark datasets and comparable results on the rest.\nT-SNE BID11 ) is used to visualize the learned graph embeddings and the results show that different graph capsules indeed capture different information of the graphs.\nWe have proposed CapsGNN, a novel framework that fuses capsules theory into GNN for more efficient graph representation learning.\nInspired by CapsNet, the concepts of capsules are introduced in this architecture to extract features in the form of vectors on the basis of nodes features extracted from GNN.\nAs a result, one graph is represented as multiple embeddings and each embedding captures different aspects of the graph properties.\nThe generated graph and class capsules can preserve not only the classification-related information but also other information with respect to graph properties which might be useful in the follow-up work and we leave this to be explored in the future.\nWe believe this is a novel, efficient and powerful data-driven method to represent high-dimensional data such as graphs.\nOur model has successfully achieved better or comparable performance when compared with other SOTA algorithms on 6 out of 10 graph classification tasks especially on social datasets.\nCompared with similar scalar-based architectures, CapsGNN is more efficient in encoding features and this would be very beneficial for processing large datasets.", "role": "user" }, { "content": "Inspired by CapsNet, we propose a novel architecture for graph embeddings on the basis of node features extracted from GNN.", "role": "assistant" } ]
science.pico_ner
science.pico_ner.277
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Participant\", \"Outcome\", \"Intervention\"].\n\nPlease return the output as a JSON object of the format: {\"Participant\": [\"patients with COPD\", ...], \"Intervention\": [\"Atenolol\", ...], \"Outcome\": [\"blood pressure\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nA randomized controlled trial assessing the health economics of realtime teledermatology compared with conventional care : an urban versus rural perspective . A randomized controlled trial was carried out to measure the cost-effectiveness of realtime teledermatology compared with conventional outpatient dermatology care for patients from urban and rural areas . One urban and one rural health centre were linked to a regional hospital in Northern Ireland by ISDN at 128 kbit/s . Over two years , 274 patients required a hospital outpatient dermatology referral -- 126 patients ( 46 % ) were randomized to a telemedicine consultation and 148 ( 54 % ) to a conventional hospital outpatient consultation . Of those seen by telemedicine , 61 % were registered with an urban practice , compared with 71 % of those seen conventionally . The clinical outcomes of the two types of consultation were similar -- almost half the patients were managed after a single consultation with the dermatologist . The observed marginal cost per patient of the initial realtime teledermatology consultation was 52.85 Pounds for those in urban areas and 59.93 Pounds per patient for those from rural areas . The observed marginal cost of the initial conventional consultation was 47.13 Pounds for urban patients and 48.77 Pounds for rural patients . The total observed costs of teledermatology were higher than the costs of conventional care in both urban and rural areas , mainly because of the fixed equipment costs . Sensitivity analysis using a real-world scenario showed that in urban areas the average costs of the telemedicine and conventional consultations were about equal , while in rural areas the average cost of the telemedicine consultation was less than that of the conventional consultation .", "role": "user" }, { "content": "{\"Intervention\": [\"telemedicine consultation\", \"conventional outpatient dermatology care\", \"conventional consultation\", \"conventional consultations\", \"conventionally\", \"realtime teledermatology\", \"telemedicine\", \"teledermatology\", \"conventional care\", \"realtime teledermatology consultation\", \"conventional hospital outpatient consultation\"], \"Outcome\": [\"health economics of realtime teledermatology\", \"cost-effectiveness of realtime teledermatology\", \"total observed costs\", \"observed marginal cost per patient\", \"consultation\", \"observed marginal cost\", \"average costs\", \"clinical outcomes\"], \"Participant\": [\"274\", \"126\"]}", "role": "assistant" } ]
science.mslr2022_ms2_multidoc_summarization
science.mslr2022_ms2_multidoc_summarization.2097
[ { "content": "You will be presented with a collection of medical randomized controlled trial (RCT) reports addressing a shared clinical question -- for instance, the efficacy of a medical intervention on a disease or condition. Given the titles and abstracts of these reports, your task is to generate the conclusion of the systematic review article synthesizing the key findings. Your conclusion should be a few sentences long. In your response, include only the conclusion and no other text. The RCT reports are below.\n\n\n\nTitle 1:\nCancer incidence in the agricultural health study.\nAbstract 1:\nOBJECTIVES This large , prospect i ve cohort study of private applicators , commercial applicators , and spouses of farmer applicators was undertaken to ascertain the etiology of cancers elevated in agriculture . METHODS The participants were matched to cancer registry files in Iowa and North Carolina . Incident cases were identified from enrollment through 31 December 2002 . St and ardized incidence ratios ( SIR ) were used to compare the cancer incidence of the participants with that of the total population in the two states . RESULTS The overall cancer incidence among farmers [ SIR 0.88 , 95 % confidence interval ( 95 % CI ) 0.84 - 0.91 ] and their spouses ( SIR 0.84 , 95 % CI 0.80 - 0.90 ) were significantly lower than expected , particularly for respiratory and urinary cancers . Commercial pesticide applicators had an overall cancer incidence comparable with the expected ( SIR 1.01 , 95 % CI 0.84 - 1.20 ) . Smoking prevalence was significantly lower than the national average . Prostate cancer was elevated among private applicators ( SIR 1.24 , 95 % CI 1.18 - 1.33 ) and commercial applicators ( SIR 1.37 , 0.98 - 1.86 ) . Excess ovarian cancer was observed for female applicators ( SIR 2.97 , 95 % CI 1.28 - 5.85 ) , but not for female spouses ( SIR 0.55 , 95 % CI 0.38 - 0.78 ) . Female spouses had a significant excess of melanoma ( SIR 1.64 , 95 % CI 1.24 - 2.09 ) , which was not observed among pesticide applicators . CONCLUSIONS Low overall cancer incidence rates seem to be a result of low overall smoking prevalence and other lifestyle factors , while excess cancer of the prostate and ovaries among applicators may be occupationally related . The excess risk of melanoma observed among spouses was unexpected\n\n\nTitle 2:\nPhase II trial of intravenous endotoxin in patients with colorectal and non-small cell lung cancer.\nAbstract 2:\nWe report the immunological and clinical results of a phase II trial with intravenously administered highly purified endotoxin ( Salmonella abortus equi ) in patients with advanced cancer . 15 patients with non-small cell lung cancer and 27 with colorectal cancer were entered into the study . 37 evaluable patients received at least four injections of endotoxin ( 4 ng/kg body weight ) and 1600 mg ibuprofen orally in 2-week intervals . Transient renal ( WHO grade 0 - 1 ) and hepatic ( WHO grade 0 - 4 ) toxicities occurred in several patients . Constitutional side-effects such as fever , chills and hypotension could not be prevented completely by pretreatment with ibuprofen . 3 patients in the colorectal cancer group demonstrated objective responses ( 1 complete remission ( CR ) , 2 partial remission ( PR ) ) . The complete remission has been maintained for more than 3 years , while the partial remissions were stable for 7 and 8 months , respectively . Only marginal antitumour effects were seen in the lung cancer group . Tolerance of the macrophage system to the stimulatory effect of endotoxin , as measured by human necrosis factor alpha ( TNF-alpha ) release into serum , built up after the first administration and remained at a steady-state level after each subsequent injection . In constrast , rising CD4:CD8 ratio and release of tumour necrosis factor beta ( TNF-beta ) indicated the continuing activation of the lymphocyte system by repetitive injections of endotoxin\n\n\nTitle 3:\nNatural History and Risk Factors of Early Respiratory Responses to Exposure to Cotton Dust in Newly Exposed Workers\nAbstract 3:\nObjective : A prospect i ve study of newly exposed cotton workers was performed to investigate the natural history of respiratory symptoms and lung function changes . Methods : A total of 157 workers naive to cotton dust exposure were investigated by question naire , spirometry , and skin tests . They were examined before employment ( baseline ) and at the end of the first week , and the first , third , sixth , and 12th month after starting work . Acute airway response was defined as either a cross-first-shift or a cross-week fall in forced expiratory volume in one second ( FEV1 ) . The longitudinal change of lung function over the year was also calculated . Five hundred seventy-two personal dust sampling and 191 endotoxin measurements were performed to assess the exposure . Results : Forty percent of workers reported work-related symptoms in the first week of the study . Smoking , endotoxin , and dust concentrations were risk factors for all work-related symptoms . Acute airway responses were witnessed after immediate exposure . Female status was the only factor found to be predictive of acute airway response . The mean longitudinal fall in FEV1 at 1 year was 65.5 mL ( st and ard error = 37.2 ) . Age , early respiratory symptoms , and early fall in cross-week FEV1 were found to predict the 12-month fall in FEV1 . Cross-first-shift and cross-week falls in FEV1 reduced in magnitude during the course of the study . Conclusions : This study of workers naive to cotton dust exposure has demonstrated that respiratory symptoms and acute airway responses develop early following first exposure , and a tolerance effect develops in those workers with the continued exposure . Current smoking and increasing exposure predicts the development of work-related lower respiratory tract symptoms , while early symptoms and acute airway changes across the working week predict the longitudinal loss of lung function at 1 year\n\n\nTitle 4:\nExposure–response relationship between endotoxin exposure and lung function impairment in cotton textile workers\nAbstract 4:\nObjectives Preventive workplace regulations are so far not based on an ubiquitously accepted threshold for airborne endotoxin in the bioaerosol . Methods In a cross-sectional study , 150 employees of a cotton spinning mill underwent lung function testing . Furthermore , in a r and om subset of 75 textile workers cross-shift lung function test and methacholine challenges were performed . Airborne current endotoxin exposure was classified as “ low ” , “ medium ” , and “ high ” ( ≤100 , > 100–≤450 , and > 450 Endotoxin Units (EU)/m3 , respectively ) based on endotoxin activity . Results The exposure – response relationship between current endotoxin exposure and prevalence of an obstructive ventilation pattern was significant ( test for linear trend : P = 0.019 ) ; the adjusted odds ratio for high endotoxin exposure was 11.22 ( 95 % confidence interval 1.03–121.17 ) . Within individuals , FEV1/FVC% was significantly reduced after the shift ( paired t test : P = 0.009 ) but not related to current endotoxin exposure . Twelve workers showed bronchial hyperresponsiveness ( 8.1 % before and 12.2 % after the work shift ; Fisher ’s exact test : P = 0.021 ) . Conclusion The study among German cotton textile workers suggests an exposure-dependent effect of current endotoxin exposure on lung function impairment with significant effects above 450 EU/m3\n\n\nTitle 5:\nA 20-year follow-up study on chronic respiratory effects of exposure to cotton dust\nAbstract 5:\nIn order to evaluate chronic effects of long-term exposure to cotton dust on respiratory health , and the role of dust and endotoxin , longitudinal changes in lung function and respiratory symptoms were observed prospect ively from 1981 to 2001 in 447 cotton textile workers , along with 472 silk textile controls . The results from five surveys conducted over the 20-yr period are reported , including st and ardised question naires , pre- and post-shift spirometric measurements , work-area inhalable dust sample collection s and airborne Gram-bacterial endotoxin analysis . Cotton workers had more persistent respiratory symptoms and greater annual declines in forced expiratory volume in one second ( FEV1 ) and forced vital capacity as compared with silk workers . After exposure cessation , in the final 5-yr period , the rate of FEV1 decline tended to slow in nonsmoking males , but not in nonsmoking females . Workers who reported byssinotic symptoms more persistently suffered greater declines in FEV1 . Chronic loss in lung function was more strongly associated with exposure to endotoxin than to dust . In conclusion , the current study suggests that long-term exposure to cotton dust , in which airborne endotoxin appears to play an important role , results in substantial adverse chronic respiratory effects\n\n\nTitle 6:\nCancer among farmers in central Italy.\nAbstract 6:\nThis case-referent study evaluated cancer risks among farmers in central Italy . Cancer cases ( N = 1674 , 17 sites ) were selected from all deceased men aged 35 - 80 years ; a r and om sample of 480 decedents formed the reference series . Farmers had a decreased risk of lung and bladder cancer and melanoma and nonsignificant excess risks for stomach , rectal , kidney , and nonmelanoma skin cancer . Stomach and kidney cancer were significantly increased among the farmers with > 10 years ' experience , and stomach , rectal , and pancreatic cancer were increased among licensed pesticide users with > 10 years ' experience . Possible relationships emerged between specific crops and cancer : fruit and colon and bladder cancer , wheat and prostate cancer , olives and kidney cancer , and potato and kidney cancer . The results regarding stomach , pancreatic , lung , bladder , and prostate cancer and melanoma congrue with earlier results . The kidney cancer excess , the association of colon and bladder cancer with orchard farming , and the excess of rectal cancer among licensed farmers are new and unexpected findings\n\n\nTitle 7:\nCancer among women textile workers in Shanghai, China: overall incidence patterns, 1989-1998.\nAbstract 7:\nBACKGROUND Cancer incidence in women textile workers has not been adequately studied . The aim of this study was to examine site-specific cancer incidence patterns in a cohort of 267,400 women textile workers in Shanghai , China . METHODS Women employed by the Shanghai Textile Industry Bureau ( STIB ) were followed for cancer incidence from 1989 to 1998 . Age-adjusted st and ardized incidence ratios ( SIRs ) and 95 % confidence intervals ( CI ) were computed based on Shanghai Cancer Registry ( SCR ) rates . RESULTS There was a decrease in cancer incidence for the cohort compared with urban Shanghai women ( SIR = 0.91 , 95 % CI = 0.89 - 0.93 ) . There were small increased risks of other endocrine tumors ( SIR = 1.31 , 95 % CI = 1.02 - 1.65 ) . There were decreased risks for esophageal ( SIR = 0.54 , 95 % CI = 0.44 - 0.66 ) , stomach ( SIR = 0.79 , 95 % CI = 0.73 - 0.85 ) , rectal ( SIR = 0.88 , 95 % CI = 0.78 - 0.98 ) , lung ( SIR = 0.80 , 95 % CI = 0.74 - 0.86 ) , cervical ( SIR = 0.37 , 95 % CI = 0.28 - 0.50 ) , ovarian ( SIR = 0.85 , 95 % CI = 0.75 - 0.96 ) , and bladder cancers ( SIR = 0.63 , 95 % CI = 0.46 - 0.85 ) . CONCLUSIONS Women employed in the textile industry had a lower than expected cancer experience compared with urban Shanghai women . Further research on this cohort will examine associations between site-specific cancers and occupational exposures to dusts and chemicals", "role": "user" }, { "content": "Conclusion Despite several limitations , this meta- analysis based on high- quality studies adds weight to the hypothesis that occupational exposure to endotoxin in cotton textile production and agriculture is protective against lung cancer", "role": "assistant" } ]
science.drug_combo_extraction_re
science.drug_combo_extraction_re.1009
[ { "content": "You will be shown a paragraph from a biomedical research article. Your task is to extract all drug combinations appearing in a specific sentence within the paragraph. Each drug combination should be assigned to one of the following three categories:\n\n- POS: Indicates a positive interaction between the drugs in the combination; for instance, drugs that have an additive or synergistic effect.\n- NEG: Indicates a negative interaction between the drugs in the combination; for instance, combining these drugs might result in adverse side effects, or might reduce their efficacy.\n- COMB: Indicates that the drugs are given in combination, but the type of interaction between the drugs cannot be determined based on the provided context.\n\nPlease format your answer as a JSON array. Each entry in the array should itself be an array formatted as [<drug_1>, <drug_2>, ..., <drug_n>, <category>] where <category> is one of the three categories listed above and <drug_1>, ..., <drug_n> are drug names copied verbatim from the input sentence. Different combinations may involve different numbers of drugs; in other words, the `n` in <drug_n> may vary.\n\nAs an example, the following output describes two extracted drug combinations; one between two drugs, and one between three drugs:\n\n[[\"artesunate\", \"artemether\", \"COMB\"], [\"Nifedipine\", \"Labetalol\", \"Prazosin\", \"POS\"]]\n\nThe paragraph and specific sentence are below. You should only extract relationships expressed in the specific sentence; the paragraph is provided for context.\n\nIn your output, return only the JSON array and no other text. If no drug combinations appear, return an empty array `[]`.\n\nparagraph: Modification of local anesthetic-induced antinociception by fentanyl in rats. In clinical practice, using the lowest doses of drugs for anesthesia or analgesia is the main goal. Opioid combinations with local anesthetics can be preferable for achieving adequate anesthesia or analgesia. The primary purpose of this study was to examine possible thermal antinociceptive effects of the opioid -fentanyl and the amide local anesthetics levobupivacaine and lidocaine when locally administered alone or in combination. The paw withdrawal latencies to noxious thermal stimuli in rats were measured to assess the antinociceptive actions of drugs after subcutaneous intraplantar injection into the hind paw. All drugs examined in this study produced dose- and time-dependent increases in the paw withdrawal latencies. Fentanyl is approximately 125 and 500 times more potent than levobupivacaine and lidocaine, respectively. At the same dose, the antinociceptive potency of levobupivacaine was 3.6-fold higher than that of lidocaine. Co-injection of the lowest doses of levobupivacaine and lidocaine dramatically increased the paw withdrawal latency . However, in the presence of fentanyl, the effects of levobupivacaine and lidocaine were different. Although co-injection of levobupivacaine with fentanyl both enhanced and prolonged antinociceptive action, the lidocaine-fentanyl combination did not significantly change the paw withdrawal latency. These results suggest that intraplantar co-administration of fentanyl with levobupivacaine, but not lidocaine, may provide more effective antinociception without increasing the dose requirements.\n\nsentence: Co-injection of the lowest doses of levobupivacaine and lidocaine dramatically increased the paw withdrawal latency .", "role": "user" }, { "content": "[[\"levobupivacaine\", \"lidocaine\", \"NEG\"]]", "role": "assistant" } ]
science.scireviewgen_multidoc_summarization
science.scireviewgen_multidoc_summarization.33633
[ { "content": "You are presented with multiple paper abstracts, a literature review title, and a specific chapter title. Your task is to write a focused literature review chapter that aligns with the given chapter title. The review should synthesize the content from the provided abstracts, draw connections between different works, highlight major findings, and identify any gaps in the existing research.\nLiterature Review Title: A Survey of Scrutinizing Delaminated Composites via Various Categories of Sensing Apparatus \nChapter Title: Camera-Based Interferometry \n1. Abstract of Cited Paper (BIB001): The paper describes the simultaneous use of digital shearographic and infrared thermographic techniques when testing aerospace structures made of honeycomb and composite materials. Non-destructive testing of the structures was accomplished using a proprietary portable shearographic system, produced at the Mechanical Engineering Department of the University of Cape Town, and a relatively inexpensive newcomer to the infrared thermography field, known as IRISYS Universal Thermal Imager. Simply, both systems were employed to image the component simultaneously and record the effect of it cooling down after being heated with the hot air from a hair drier or a heating lamp. The techniques proved effective in identifying defects such as the crushed honeycomb core in a section of the wing of an unmanned aerial vehicle (UAV) and the delaminations at different depths in the main rotor blade of an Oryx helicopter. The images indicating the defects as produced by the shearographic system are visually more dramatic, particularly when enhanced by phase stepping and colour techniques. On the other hand, the thermographic images, besides indicating the position of the flaw, gave information regarding the temperature of the region of the defect relative to the cooler surrounding material. In addition, there was the indication that during the cooling down, the infrared system continued to indicate the presence of the flaws/defects for a longer period of time than the shearographic system. It is suggested that in similar tests, the temperature data be coupled to the out of plane displacement gradients with a view to providing quantitative thermal stress results. \n2. Abstract of Cited Paper (BIB002): Delamination is considered one of the most serious problems associated with the machining of fiber-reinforced composite materials. The evaluation of the drilling-induced delamination damage in these materials is rather difficult, particularly for the carbon fiber-based composites. This is attributed to their color that makes the current visual inspection techniques extremely difficult. This paper presents a non-contact nondestructive technique for measuring and determining the extent of delamination in composite materials. The approach developed consists of utilizing a shadow moire laser based imaging technique. This technique allows the assessment of the degree of delamination resulting from the machining processes necessary for final set-up and assembly. Mechanical drilling has been implemented on carbon fiber reinforced epoxy (CFRE) composites. The effects of cutting variables (feed and speed) on the delamination size in mechanical drilling of CFRE composites have been investigated. The discussed work should improve reliability and economy in fabrication and assembly of carbon fiber-based composites. Moreover, the results show that the proposed technique is easily adaptable to typical manufacturing environments and possesses promising advantages with respect to accuracy and simplicity. \n3. Abstract of Cited Paper (BIB003): Abstract A novel procedure for the identification of delaminations in composite laminated panels from surface displacement measurements is presented in this paper. The damage identification procedure is formulated as an inverse problem where experimental data from digital speckle pattern interferometry (DSPI) out-of-plane displacement measurements, are used to quantify the planar location, size, orientation and depth of the flaw. The inverse problem is then reduced to an optimization problem where the optimum crack parameters are found by minimizing the objective function through the use of a real-coded low-population genetic algorithm. DSPI measurements of ten delaminated T700/LTM-45EL carbon/epoxy laminate panels are used to validate the methodology presented in this paper. \n4. Abstract of Cited Paper (BIB004): Abstract The location of delamination in composite laminates based on their vibration characteristics is presented in this paper. Composite materials are widely used as structural materials for aerospace engineering, because of its excellent mechanical properties such as light weight, high stiffness and anti-corrosion characteristics. This paper uses pulsed electronic speckle pattern interferometry (Pulsed-ESPI) to perform non-destructive evaluation of carbon fiber reinforced plastic (CFRP) laminate plates containing various sizes of artificial defects located at the centre of the specimen, based on vibration characteristic using double-pulsed Ruby laser with a pulse duration of 30 ns. The pulsed separation can be adjusted in between 2 and 800 μs. The main advantage of this technique over conventional ESPI is the ability to carry out measurements even under harsh environmental conditions. From the results, it seems to be effective for the detection of defects in various kinds of composite materials. In this paper, the basic principles of the technique are described briefly. \n5. Abstract of Cited Paper (BIB005): We report on a wide-field optical monitoring method for revealing local delaminations in sandwich-type composite plates at video-rate by holographic vibrometry. Non-contact measurements of low frequency flexural waves is performed with time-averaged heterodyne holography. It enables narrowband imaging of local out-of-plane nanometric vibration amplitudes under sinusoidal excitation and reveals delamination defects, which cause local resonances of flexural waves. The size of the defect can be estimated from the first resonance frequency of the flexural wave and the mechanical parameters of the observed layer of the composite plate.\n\nLiterature Review:", "role": "user" }, { "content": "Moiré interferometry: The capability of fringe patterns of shadow moiré formed through master grating interacting with its shadow on the specimen's surface has been investigated to detect drilling-caused delamination as a result of high strains caused around machining holes in composite laminates BIB003 . Delamination size decided as the size and number multiplication of pixels in the delamination area was assessed by out-of-plane surface displacement associated with the fringe's phase change in CFRP laminates illuminated by a laser source. Speckle pattern interferometry (SPI): Speckle interferometer as a strain measurement technique incorporating laser beams derives surface displacements. Upon the SPI application to delamination detection, data of digital SPI detecting delamination-induced bulging effect appeared through a sequence of the out-of-plane displacements of the cross-ply CFRP laminate's surface placed in a vacuum has quantitatively been utilized for delamination detection BIB004 . Data of largely deformed areas located within bulge borders processed through genetic algorithm along with FE data predicted a thorough identification of delamination. Moreover, electronic SPI displacements measured at in-plane and out-of-plane directions have successfully been applied to inspect composite laminates containing damage mechanisms of splitting, microcrack, and delamination under laser illumination BIB001 . Holography: Holography as a powerful interferometric technique to measure surface displacement and vibration modes even under noisy environments is a complicated technique in terms of measurement interpretation. In this regard, the shape and location of delamination in CFRP laminate have been estimated through out-of-plane vibrational mode patterns at specific frequencies for each delamination shape measured using pulsed laser holography, emphasizing on the technique's merit, that is the reduction of the environmental noises BIB005 . Additionally, amplitudes of the local out-of-plane vibration modes at a specific frequency at a healthy area and delaminated area derived from heterodyne holography were applied for delamination inspection, thereby verifying delamination location with data measured by SLDV and thermography BIB006 . Shearography: Shearography (speckle pattern shearing interferometry) developed to eliminate limitations of the holography technique and particularly used for inspecting delamination has simplified the optical operation and measurement interpretation. Eliminating the reference beam, it directly renders the first derivative of the out-of-plane displacement difference of the specimen's surface under applied stress with respect to the unstressed state. With regard to the shearography application to delamination inspection in CFRP laminates, perturbed speckle patterns revealed delamination shape and size while assessing the delamination depth through embedding both resonant frequencies derived from shearography and theoretical ones as a function of delamination depth (within the idealized delaminated area) into an optimization process . In testing composite structures via shearography synchronized with thermography, delamination was dramatically localized through shearography based fringe patterns, particularly when incorporated with phase shifted technique using filtering and color routines BIB002 . Delamination of a lower depth shows more fringes compared with deeper ones, the fringes number, thus can be an indicator of delamination depth.", "role": "assistant" } ]
science.chemsum_single_document_summarization
science.chemsum_single_document_summarization.74679
[ { "content": "You will be shown the title of a chemistry paper, together with sections excerpted from the body of the paper. The paper sections may include HTML tags like `<p>` and `</p>`; please ignore these. Your task is to write an abstract for the paper. Your response should include the abstract and no other text.\n\nTitle:\nGenome-scale CRISPR pooled screens\n\nPaper sections:\nIntroduction\n<p>Pooled genetic screens, where each cell receives a different genetic perturbation prior to a phenotype-based selection, are a powerful technique for rapidly identifying specific genome elements that influence a selected phenotype. A single successful screen may provide a treasure trove of new genotype-phenotype interactions that can be used as a launching point for follow-up studies on the most promising gene targets. The key idea behind pooled screens is that phenotypic selection results in an enrichment (or depletion) of genetic perturbations relevant to the phenotype. An example of a selected phenotype can be resistance to a drug or expression of a particular cell surface receptor. In this way, pooled screens can search over a space of thousands of perturbations (e.g. genome-scale), providing a substantially simpler technique than arrayed-format screens where each perturbation is delivered in a separate well.</p><p>Traditionally, to generate the pool of perturbations, these screens have relied on DNA mutagenesis (induced through chemical mutagens, mobile genetic element insertion or radiation) or manipulations at the transcript-level, such as RNA interference (RNAi). However, pooled screens using DNA mutagenesis can require painstaking work to map the mutation sites. Pooled screening at the transcript level with RNAi also presents challenges, such as incomplete knock-down and large off-target effects [1,2].</p><p>Recently, the easy programmability of microbial CRISPR systems has created a new opportunity for performing pooled screens at the DNA-level in a targeted fashion [3,4]. In their native context, CRISPR nucleases function as a prokaryotic immune system, cleaving foreign DNA from phage or plasmids before they can damage the host cell [5–7]. The nuclease is guided to a specific DNA sequence using a short, single-stranded RNA that is complementary to the target DNA. Genome editing applications using S. pyogenes Cas9 [8,9], the most commonly used CRISPR system, replace the endogenous prokaryotic RNA components with a synthetic single guide RNA (sgRNA) that has a 20-bp sequence complementary to the target site [10]. After the Cas9-sgRNA ribonucleoprotein complex recognizes the target sequence in the genome, it creates a double-strand break (DSB). DSB repair mechanisms, such as non-homologous end-joining (NHEJ), can delete or add a few bases during the repair process. When the NHEJ-mediated repair occurs in a coding region, this can introduce a frameshift mutation where the net result is gene loss-of-function. As reviewed elsewhere, the Cas9 system results in high efficiency genome modification in diverse genetic model (and non-model) organisms [11].</p><p>Using this RNA-guided system to produce targeted loss-of-function mutations, it is possible to build large, genome-scale libraries of CRISPR reagents for high-throughput pooled screens. The combination of DNA-level manipulation with easy targeting/programmability presents a new avenue for understanding the function of genome (and epigenome) elements in a wide variety of functional assays.</p>\n\nTechnological underpinnings of CRISPR screens\n<p>Within months of the initial use of Cas9 for mammalian genome editing [8,9], several groups (including our own) took advantage of existing pooled synthesis and cloning techniques to construct large Cas9-sgRNA lentiviral libraries using similar methods to those previously used for large RNAi lentiviral libraries [3,4,12]. Typically, libraries are designed with multiple sgRNAs targeting each gene. Consistent changes in multiple sgRNAs can be used to increase confidence in a particular candidate gene. Briefly, libraries are synthesized as DNA and cloned into plasmids to produce lentivirus. After designing sgRNA guide sequences to target different genes, the oligonucleotides are synthesized in a pool of typically 104–105 different guides. Several different commercial platforms exist for producing pooled oligonucleotides (e.g. CustomArray, Twist Bioscience, Agilent Technologies) and synthesis fidelity and length is constantly improving. The pooled oligos are cloned into the sgRNA cassette of the backbone and the resulting lentivirus expresses the Cas9 protein and target-specific sgRNA (Figure 1a); alternatively, for higher viral titer, the small sgRNA can be delivered on a separate virus from the large Cas9 protein [13].</p><p>Lentivirus is produced from the cloned plasmid pool and then used to transduce mammalian cells for screening. Because lentiviruses integrate into the genome, the viral integrant serves as a tag for readout of which sgRNA construct is delivered to a particular cell. Viruses are introduced at a low multiplicity-of-infection (MOI) in order to ensure that cells receive only a single sgRNA from the library pool. The frequency of observing any particular tag (sgRNA) before and after phenotypic selection is the key parameter measured during a pooled screen. This frequency can be computed by taking a representative sample of the population, PCR-amplifying the lentiviral cassettes and their sgRNAs, and then counting the frequency of each sgRNA after next-generation sequencing (Figure 1b). Ideally, the initial distribution of the library is as uniform as possible, so that, after selection, any depletion or enrichment of specific sgRNAs is readily identifiable. In reality, biases can be introduced at many different stages, such as library synthesis, cloning, viral transduction, cell sampling and PCR amplification, which makes it important to carefully control each stage of library handling before sequencing.</p><p>After selection, changes in sgRNA representation (enrichment or depletion) indicate potential candidates for further study (Figure 1b). A key concept in screen analysis is that the strongest gene candidates are those which have multiple sgRNAs simultaneously enriched/depleted. In enrichment screens, cells carrying specific sgRNAs are selected for. Representative examples of enrichment are drug resistance or toxin screens, where a gene knock-out promotes cell survival after exposure to a drug/toxin, or FACS-based selection of cells with a fluorescent gene reporter to find mutations that modulate gene expression. By contrast, depletion screens can identify essential genes, where loss-of-function is incompatible with continued cell survival, and synthetic lethal interactions, where typically nonessential genes become essential when cells carry a particular (usually oncogenic) mutation in a different gene. Depletion analysis in a drug or toxin screen can also be used to find gene targets that enhance sensitivity to the drug or toxin.</p>\n\nApplications of pooled CRISPR screens\n<p>Over the past two years, pooled CRISPR screens have been deployed in diverse phenotypic assays: cancer drug resistance [3,4,14], bacterial toxin resistance [12], West Nile virus-induced cell death [15], mitochondrial metabolism [16], identifying essential and synthetic lethal genes in cancer cell lines [17–19], identifying genetic drivers of metastasis in an in vivo screen [20], and understanding gene networks in immune cells [21,22]. In addition, two catalytically inactive versions of Cas9 with arrays of transcriptional activation domains (in conjunction with sgRNA libraries targeting promoter regions) have facilitated genome-scale gain-of-function screens (Figure 2) [23,24]. To illustrate the utility of pooled CRISPR screens, I focus here on two examples of loss-of-function screens: an enrichment screen and a set of related depletion screens.</p><p>One of the first pooled screens was a positive selection screen for drug resistance in a human BRAF mutant (V600E) melanoma cell line using a genome-scale library targeting ~18,000 genes with ~65,000 sgRNAs [3]. This BRAF gain-of-function mutation is found in more than 50% of malignant melanomas and vemurafenib, a FDA-approved BRAF inhibitor, was shown to induce apoptosis preferentially in cells with the mutant form of BRAF [25]. In the clinic, vemurafenib results in reduction of solid tumors but, within weeks, resistance to the drug develops, which is followed by tumor re-growth and a poor prognosis. Identifying mechanisms of resistance can help inform the design of new treatment strategies to combine targeted BRAF inhibitors with other drugs. In the BRAF mutant melanoma cells, the genome-scale CRISPR screen identified both of the previously established loss-of-function mechanisms of vemurafenib resistance and several novel gene targets [26,27]. Importantly, the consistency of different reagents targeting the same gene was higher in the CRISPR screen than in the same screen using a genome-scale lentiviral RNAi library [27]. For example, the 10 most highly enriched genes in the RNAi screen had, on average, only 20% of the reagents targeting each gene enriched, indicating little agreement between different RNAi reagents designed to target the same gene. The CRISPR screen had, on average, 80% of reagents targeting each gene enriched for the 10 most highly enriched genes. This large difference in consistency suggests that the false-positive rate is lower using CRISPR reagents (further confirmed in a recent detailed comparison between screening technologies [28]) and thus gives us greater confidence in the identified gene candidates. As expected, individual loss-of-function mutations in several candidates from the CRISPR screen were shown to confer vemurafenib resistance in follow-up experiments [3], whereas only one gene from the RNAi screen could be validated [27]. A practical advantage of the lower false-positive rate is that genome-wide CRISPR screens can be conducted with smaller libraries (and thus using fewer cells) than a comparable genome-wide RNAi screen.</p><p>Two recent papers used depletion screens in multiple cell lines to estimate the percentage of genes in the human genome that are essential [17,18]. By measuring sgRNAs that are consistently depleted across multiple cell lines, they estimate that about 1,700 genes (~10% of all protein coding genes) are essential, which is about 5-fold higher than previous estimates from RNAi-based screens [29,30]. One reason for this increase is that RNAi knock-down is incomplete, whereas CRISPR is able to modify all copies of a gene when delivered using lentivirus and expressed constitutively [3,17]. This advantage is also relevant when comparing CRISPR screens to other DNA-based screening technologies: Wang et al. used both CRISPR knock-out and a gene-trap technique using retroviral insertions in the near-haploid KBM7 cell line [17]. The gene-trap screen was unable to detect any essential genes on chromosome 8, which is the only diploid chromosome in KBM7 cells. Since gene-trap screens rely on random insertion in the genome, complete knock-out is only possible in haploid cells, which only have one copy of each gene. This issue is especially relevant in cancer cell lines where multi-copy gene amplifications are more common. Finally, by comparing essential genes across different cell lines, both papers identified core essential genes and a much smaller set of context-dependent essential genes that were essential only in particular lines [17,18].</p>\n\nNew frontiers in CRISPR screens\n<p>Since their initial development, pooled CRISPR screens have delivered new insights in basic biology and in applied, clinically relevant domains, yet there are still many opportunities for further development of this technology. Although in vivo screens have established a first step toward understanding cell-cell interactions [20], most CRISPR screens have been performed with a single cell type in isolation. New screens could replace drug selection with a cell-based selective pressure (e.g. T cell-mediated targeting of cancer cells) or combine library-transduced cells with other disease-relevant cell types (e.g. cancer cells cultured with endothelial or immune cells). It also is possible to deliver different pooled libraries to different populations of cells, capitalizing on recent cell type-specific gene expression data from resources like the Genotype-Tissue Expression (GTEx) project and the Allen Brain Atlas [31,32]. For example, in a co-culture system of neurons and glia, each cell type could separately be transduced with a specific library (targeting either neuron- or glia-relevant genes) before being cultured together. Along similar lines, combinatorial delivery of two or more sgRNAs in a single construct would allow targeting of multiple genes within one cell (Figure 2). However, it will be necessary to use smaller, targeted libraries to avoid a combinatorial explosion: A genome-scale CRISPR library with 105 distinct sgRNAs would require (at 103 cells/construct) an impractical 10 trillion cells to screen all possible pairs of sgRNAs but a library with 102–103 distinct sgRNAs is feasible given current screen handling and cell culture capacities.</p><p>The success in repurposing CRISPR using transcriptional activation domains for gain-of-function screens suggests several other kinds of screens might be possible by attaching different epigenetic effector domains to a catalytically inactive CRISPR nuclease. These could act to modulate chromatin states or alter other epigenetic elements at specific genomic locations [33,34]. Another consideration is where in the genome to target, either with nuclease or other effector domains. Presently, there has been little work on using pooled CRISPR screening approaches to manipulate non-coding genome elements, such as non-coding RNAs (e.g. miRNAs, lincRNAs), enhancers, repressors, promoters, or structural elements (e.g. CTCF binding sites) [13,20,35]. Delivery of multiple guides with a nuclease may enable large deletions, which would be useful for assessing the importance of larger non-coding elements. Both dual sgRNA-based deletions and single sgRNA scanning mutagenesis will be helpful for locating and understanding functional elements in the non-coding genome. From a technological standpoint, most CRISPR screens have focused on loss-of-function because of the high efficiency of NHEJ-mediated DSB repair after nuclease cutting, but it would be more desirable to precisely knock-in specific human genetic variants (e.g. SNPs, CNVs) at predefined locations. Although doable at a single genomic locus (c.f. CRISPR-based pooled saturating mutagenesis of 3 residues in the gene BRCA1 [36]), this type of precision editing is not yet possible at multiple genomic loci — however, given the rapid pace of genome engineering technology development, this may change soon.</p><p>Compared with RNAi and other pooled screening technologies, CRISPR pooled screens have demonstrated higher consistency between different reagents targeting the same genetic element and, because of this, they have already found broad applicability and produced new biological insights. Given the many promising future directions, CRISPR pooled screens have enormous potential to improve our understanding of the genome and to identify the causal role of genome elements in both normal and disease states.</p>", "role": "user" }, { "content": "Genome editing technologies such as Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) systems have ushered in a new era of targeted DNA manipulation. The easy programmability of CRISPR using short oligonucleotides enables rapid synthesis of large-scale libraries for functional genetic screens. Here we present fundamental concepts and methods for pooled CRISPR screens and review biological results from recent genome-scale loss-of-function and gain-of-function screens. We also discuss new frontiers in pooled screens, including novel effector domains for functional screens and applications in the noncoding genome.", "role": "assistant" } ]
science.bioasq_factoid_qa
science.bioasq_factoid_qa.1261
[ { "content": "You will be shown biomedical passages and your task is to answer the provided question. Provide a concise and accurate response and do not include any additional text in the response other than the answer.\n\nQuestion: What are TAMs in cancer therapy?\n\nContext: tumor-associated macrophages (TAMs)\ntumour-associated macrophages (TAMs)\ntumor-associated macrophages (TAMs)\ntumor-associated macrophages (TAMs)\nHowever, endocrine therapy resistance mechanisms mediated through interactions between breast cancer cells and tumor-associated macrophages (TAMs) are still unclear.\nntional cancer therapy also strengthens cancer-related inflammation by inducing massive tumor cell death that activate surrounding immune-infiltrating cells such as tumor-associated macrophages (TAMs). Mac\nAlthough M2-like tumor-associated macrophages (TAMs) have been considered as a vital therapeutic target in cancer therapy due to their role in promoting tumor progression and metastasis, very few compounds have been identified to inhibit M2-like polarization of TAMs. H\nEliciting effective antitumor immune responses in patients who fail checkpoint inhibitor therapy is a critical challenge in cancer immunotherapy, and in such patients, tumor-associated myeloid cells and macrophages (TAMs) are promising therapeutic targets. We\nAs CSCs depend on their specific niche, including tumor-associated macrophages (TAM), elucidating the network between CSCs and TAMs may help to effectively inhibit the progression and ADT resistance of prostate cancer.Experimental Design: The underlying intracellular mechanism that sustains the stem-like characteristics of CSCs in prostate cancer was assessed via RNA sequencing, co-immunoprecipitation, chromatin immunoprecipitation, and other assays.\nTumor-associated macrophages (TAMs) is a promising therapeutic target for cancer immunotherapy, while TAMs targeting therapy using nano-sized drug delivery system (NDDS) is a great challenge.\nTumor-associated macrophages (TAMs) play a crucial part in cancer evolution.\nTumor-associated macrophages (TAMs) of M2 phenotype have mediated the immunosuppression in a tumor microenvironment, facilitating the escape of tumor cells from immunosurveillance.\nIn this review, we outline the importance of tumor-associated macrophages (TAM), a major component of the TME, in the response of tumors to cancer therapy.\nIn particular, tumor-associated macrophages (TAMs) are the most abundant leucocyte subset in many cancers and play a major role in the creation of a protective niche for tumor cells.\nThese cells, known as tumor-associated macrophages (TAMs), play complex but pivotal roles in the outcome of photodynamic therapy (PDT) of malignant lesions.\nTumor-associated macrophages (TAMs) represent the most abundant innate immune cells in tumors.\nTumor-associated macrophages (TAMs) are among the most abundant immune cells in the TME.", "role": "user" }, { "content": "Tumor Associated Macrophages", "role": "assistant" } ]
science.chemdner_ner
science.chemdner_ner.2571
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\": [\"Arsenic\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nAcetylcholinesterase inhibition reveals endogenous nicotinic modulation of glutamate inputs to CA1 stratum radiatum interneurons in hippocampal slices. The involvement of brain nicotinic acetylcholine receptors (nAChRs) in the neurotoxicological effects of soman, a potent acetylcholinesterase (AChE) inhibitor and a chemical warfare agent, is not clear. This is partly due to a poor understanding of the role of AChE in brain nAChR-mediated functions. To test the hypothesis that AChE inhibition builds sufficient acetylcholine (ACh) in the brain and facilitates nAChR-dependent glutamate transmission, we used whole-cell patch-clamp technique to record spontaneous glutamate excitatory postsynaptic currents (EPSCs) from CA1 stratum radiatum interneurons (SRI) in hippocampal slices. First, the frequency, amplitude and kinetics of EPSCs recorded from slices of control guinea pigs were compared to those recorded from slices of guinea pigs after a single injection of the irreversible AChE inhibitor soman (25.2μg/kg, s.c.). Second, EPSCs were recorded from rat hippocampal slices before and after their superfusion with the reversible AChE inhibitor donepezil (100nM). The frequency of EPSCs was significantly higher in slices taken from guinea pigs 24h but not 7 days after the soman injection than in slices from control animals. In 52% of the rat hippocampal slices tested, bath application of donepezil increased the frequency of EPSCs. Further, exposure to donepezil increased both burst-like and large-amplitude EPSCs, and increased the proportion of short (20-100ms) inter-event intervals. Donepezil's effects were suppressed significantly in presence of 10μM mecamylamine or 10nM methyllycaconitine. These results support the concept that AChE inhibition is able to recruit nAChR-dependent glutamate transmission in the hippocampus and such a mechanism can contribute to the acute neurotoxicological actions of soman.", "role": "user" }, { "content": "{\"Chemical\": [\"glutamate\", \"donepezil\", \"acetylcholine\", \"methyllycaconitine\", \"Donepezil\", \"mecamylamine\"]}", "role": "assistant" } ]
science.scicite_classification
science.scicite_classification.6356
[ { "content": "You will be presented with a citation sentence from a research paper. Classify the intent behind this citation by choosing from one of the following categories:\n\n- Method: Cites the paper for its methodology or procedure.\n- Background: Cites the paper to provide background information.\n- Result: Cites the paper for its findings or results.\n\nYour answer should be a single word from the following list of options: [\"Method\", \"Background\", \"Result\"]. Do not include any other text in your response.\n\nCitation sentence:\n\nMore information about the samples and recruitment can be found in Jacob et al. (2003) and Duncan et al. (2008).", "role": "user" }, { "content": "Method", "role": "assistant" } ]
science.pubmedqa_qa
science.pubmedqa_qa.29
[ { "content": "You will be shown an abstract from a research paper and a corresponding research question. Determine the answer to the research question based on the context provided. Your answer should be one of: \"YES\", \"NO\", or \"MAYBE\".\n\nAbstract:\n(OBJECTIVE) To assess pain-related attentional biases among individuals with episodic migraine.\n(BACKGROUND) Prior studies have examined whether chronic pain patients selectively attend to pain-related stimuli in the environment, but these studies have produced largely mixed findings and focused primarily on patients with chronic musculoskeletal pain. Limited research has implicated attentional biases among chronic headache patients, but no studies have been conducted among episodic migraineurs, who comprise the overwhelming majority of the migraine population.\n(METHODS) This was a case-control, experimental study. Three hundred and eight participants (mean age = 19.2 years [standard deviation = 3.3]; 69.5% female; 36.4% minority), consisting of 84 episodic migraineurs, diagnosed in accordance with International Classification of Headache Disorders (2(nd) edition) criteria using a structured diagnostic interview, and 224 non-migraine controls completed a computerized dot probe task to assess attentional bias toward headache-related pictorial stimuli. The task consisted of 192 trials and utilized 2 emotional-neutral stimulus pairing conditions (headache-neutral and happy-neutral).\n(RESULTS) No within-group differences for reaction time latencies to headache vs happy conditions were found among those with episodic migraine or among the non-migraine controls. Migraine status was unrelated to attentional bias indices for both headache (F [1,306] = 0.56, P = .45) and happy facial stimuli (F [1,306] = 0.37, P = .54), indicating a lack of between-group differences. Lack of within- and between-group differences was confirmed with repeated measures analysis of variance.\n\nQuestion:\nDo episodic migraineurs selectively attend to headache-related visual stimuli?\n\nAnswer:", "role": "user" }, { "content": "NO", "role": "assistant" } ]
science.cdr_ner
science.cdr_ner.235
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\", \"Disease\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\": [\"Naloxone\", ...], \"Disease\": [\"Hypotensive\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nEstradiol reduces seizure-induced hippocampal injury in ovariectomized female but not in male rats. Estrogens protect ovariectomized rats from hippocampal injury induced by kainic acid-induced status epilepticus (SE). We compared the effects of 17beta-estradiol in adult male and ovariectomized female rats subjected to lithium-pilocarpine-induced SE. Rats received subcutaneous injections of 17beta-estradiol (2 microg/rat) or oil once daily for four consecutive days. SE was induced 20 h following the second injection and terminated 3 h later. The extent of silver-stained CA3 and CA1 hippocampal neurons was evaluated 2 days after SE. 17beta-Estradiol did not alter the onset of first clonus in ovariectomized rats but accelerated it in males. 17beta-Estradiol reduced the argyrophilic neurons in the CA1 and CA3-C sectors of ovariectomized rats. In males, estradiol increased the total damage score. These findings suggest that the effects of estradiol on seizure threshold and damage may be altered by sex-related differences in the hormonal environment.", "role": "user" }, { "content": "{\"Chemical\": [\"Estradiol\", \"silver\", \"lithium\", \"estradiol\", \"pilocarpine\", \"17beta-estradiol\", \"17beta-Estradiol\", \"kainic acid\"], \"Disease\": [\"SE\", \"seizure\", \"hippocampal injury\", \"status epilepticus\"]}", "role": "assistant" } ]
science.covid_deepset_qa
science.covid_deepset_qa.513
[ { "content": "You will be shown a paragraph from a biomedical research paper, followed by a question about the paper. Please extract text from the paragraph to answer the question. Do not include any text in your response other than the answer.\n\nParagraph: Furthermore, we tested the proposed model using two different datasets of weekly influenza confirmed cases in two countries, namely the USA and China. The outcomes also showed good performances. Text: A large family of viruses, called coronaviruses, are severe pathogens for human beings, which infect respiratory, hepatic, gastrointestinal, and neurologic diseases. They are distributed among humans, birds, livestock, mice, bats, and other wild animals [1] [2] [3] . The outbreaks of two previous coronaviruses, SARS-CoV and MERS-CoV in 2003 and 2012, respectively, have approved the transmission from animal to animal, and human to human [4] . In December 2019, the World Health Organization (WHO) received notifications from China for many cases of respiratory illness that were linked to some people who had visited a seafood market in Wuhan [5] . Currently, Wuhan city suffers from the spreading of a novel coronavirus, called COVID-19 (previously, it was called 2019-nCoV). In [6] , the authors concluded that COVID-19 likely originated in bats, because it is more similar to two bat-derived coronavirus strains. However, the source of the COVID-19 is not confirmed yet, and it communities, Hong Kong and Toronto, were 1.2 and 1.32, respectively. Ong et al. [20] proposed a monitoring and forecasting model for influenza A (H1N1-2009). Furthermore, Nah et al. [21] proposed a probability-based model to predict the spread of the MERS. The Adaptive Neuro-Fuzzy Inference System (ANFIS) [22] is widely applied in time series prediction and forecasting problems, and it showed good performance in many existing applications. It offers flexibility in determining nonlinearity in the time series data, as well as combining the properties of both artificial neural networks (ANN) and fuzzy logic systems. It has been applied in various forecasting applications, for example, in [23] , a stock price forecasting model was proposed using ANFIS and empirical mode decomposition. Chen et al. [24] proposed a TAIEX time series forecasting model based on a hybrid of ANFIS and ordered weighted averaging (OWA). In [25] , another time series forecasting method was presented for electricity prices based on ANFIS. Svalina et al. [26] proposed an ANFIS based forecasting model for close price indices for a stock market for five days. Ekici and Aksoy [27] presented an ANFIS based building energy consumption forecasting model. More so, ANFIS is also applied to forecast electricity loads [28] . Kumar et al. [29] proposed an ANFIS based model to forecast return products. Ho and Tsai [30] applied ANFIS to forecast product development performance. However, estimating ANFIS parameters is a challenge that needs to be improved. Therefore, in previous studies, some individual swarm intelligence (SI) methods have been applied to the ANFIS parameters to enhance time series forecasting because these parameters have a significant effect on the performance of ANFIS. The SI methods include the particle swarm optimization (PSO) [31, 32] , social-spider optimization [33] , sine-cosine algorithm (SCA) [34] , and multi-verse optimizer (MVO) [35] . For example, in [34] SCA algorithm was applied to improve the ANFIS model to forecast oil consumption in three countries, namely, Canada, Germany, and Japan. In the same context, in [35] , The MVO algorithm was used to enhance the ANFIS model to forecast oil consumption in two countries. In addition, in [36] the PSO was used with ANFIS to predict biochar yield. However, individual SI algorithms may stock at local optima. Therefore, one solution is to apply hybrid SI algorithms to avoid this problem. In [37] , a hybrid of two SI algorithms, namely GA and SSA, was presented to improve the ANFIS model. The proposed new model called GA-SSA-ANFIS was applied to forecast crude oil prices for long-term time series data. However, the previously mentioned methods suffer from some limitations that can affect the performance of the forecasting output such as slow convergence and the ability to balance between exploration and exploitation phases can influence the quality of the final output. This motivated us to propose an alternative forecasting method dependent on the hybridization concept. This concept avoids the limitations of traditional SI techniques by combining the strengths of different techniques, and this produces new SI techniques that are better than traditional ones. In the current study, we propose an improved ANFIS model based on a modified flower pollination algorithm (FPA) using the salp swarm algorithm (SSA). The FPA is an optimization algorithm proposed by Yang [38] , which was inspired by the flow pollination process of the flowering plants. The FPA was employed in various optimization applications, for example to estimate solar PV parameter [39, 40] , solving sudoku puzzles [41] , feature selection [42] , antenna design [43] , and other applications [44] [45] [46] [47] . Moreover, SSA is also an optimization algorithm proposed by Mirjalili et al. [48] inspired by the behavior of salp chains. In recent years, the SSA was utilized to solve different optimization problems, such as feature selection [49, 50] , data classification [51] , image segmentation [52] , and others [53, 54] . The proposed method called FPASSA is a hybrid of FPA and SSA, in which the SSA is applied as a local search method for FPA. The proposed FPASSA starts by receiving the historical COVID-19 dataset. Then a set of solutions is generated where each of them represents the value for the parameters of the ANFIS model. Then the quality of each solution is calculated using the fitness value, and the solution that has the best fitness value is chosen to represent the best solution. Then the probability of each solution is computed. Then the current solution will be updated, either using global or local strategy in FPA. However, in the case of local strategy, the operators of SSA or FPA will be used according to the probability of the fitness value for each solution. The process of updating the solutions is repeated until reaching the stop condition, and the best parameter configurations are used to forecast the number of confirmed cases of COVID-19. The main contribution points of the current study are as follows:\n\n1. We propose an efficient forecasting model to forecast the confirmed cases of the COVID-19 in China for the upcoming ten days based on previously confirmed cases. An improved ANFIS model is proposed using a modified FPA algorithm, using SSA. We compare the proposed model with the original ANFIS and existing modified ANFIS models, such as PSO, GA, ABC, and FPA. The rest of this study is organized as follows. The preliminaries of ANFIS, FPA, and SSA are described in Section 2. Section 3 presents the proposed FPASSA, and Section 4 presents the experimental setup and results. We conclude this study in Section 5. The principles of the ANFIS are given in this section. The ANFIS model links the fuzzy logic and neural networks [22] . It generates a mapping between the input and output by applying IF-THEN rules (it is also called Takagi-Sugeno inference model). Figure 1 illustrates the ANFIS model where, y and x define the inputs to Layer 1 whereas, O 1i is its output of node i that is computed as follows:\n\nwhere µ denotes the generalized Gaussian membership functions. A i and B i define the membership values of µ. α i and ρ i denote the premise parameters set. The output of Layer 2 (it is also known as the firing strength of a rule) is calculated as follows:\n\nMeanwhile, the output of Layer 3 (it is also known as the normalized firing strength) is calculated as follows:\n\nThe output of Layer 4 (it is also known as an adaptive node) is calculated as follows:\n\nwhere r i , q i , and p i define the consequent parameters of the node i. Layer 5 contains only one node; its output is computed as: \n\nFlower Pollination Algorithm is an optimization method proposed by Yang [38] . It simulates the transfer of flowers' pollen by pollinators in nature. This algorithm utilizes the two types of pollination (i.e., self-pollination and cross-pollination). In self-pollination, the pollination occurs with no pollinators, whereas, in cross-pollination, the pollens are moved between different plants. In more detail, the self-pollination can be represented as a local pollination while the cross-pollination can be called global pollination. The global pollination or cross-pollination can be mathematically formed as follows:\n\nwhere x t i defines the pollen i at iteration t. L denotes the pollination's strength or the step size. F * is the target position or best solution. In some cases, insects can fly with different distance steps for a long space; therefore, Levy fly distribution is applied to simulate this movement. where λ = 1.5. Γ(λ) denotes the gamma function. This distribution is available for large steps s > 0. The self-pollination or local pollination can be mathematically formed as follows:\n\nwhere x t i and x k i represent pollens from different flower in the same plant. in the range [0,1] The process of pollination can be done using cross-pollination or self-pollination. Therefore, the random variable p, in the range [0, 1], is used to determine this process. SSA is an optimization technique introduced by [48] . It simulates the Salps' behavior in nature. This behavior is called salp chain. The mathematical model of SSA begins by splinting its population into a leader group and followers group. The leader is the front salp, whereas, the followers are the other salps. The search space is determined in n-dimensions with n variables. Equation (10) works to update the salps' positions. where x 1 j denotes the leader's position in j-th dimension. F j is the target position. ub j and lb j represent the max and min bounds, respectively. c 2 and c 3 denote random numbers in [0, 1]. c 1 is an important parameter; it balances between the exploration and exploitation phases. It is computed as follows:\n\nwhere the current loop number is t and the max loop' number is t max . Then, the followers' position is updated as follows:\n\nwhere x i j defines the i-th position of the follower in j-th dimension. i > 1. This section explains the proposed FPASSA-ANFIS method. It is a time series method for forecasting the confirmed cases of the COVID-19, as given in Figure 2 . The FPASSA-ANFIS utilizes the improved FPA to train the ANFIS model by optimizing its parameters. The FPASSA-ANFIS contains five layers as the classic ANFIS model. Layer 1 contains the input variables (the historical COVID-19 confirmed cases). Whereas Layer 5 produces the forecasted values. In the learning phase, the FPASSA is used to select the best weights between Layer 4 and Layer 5. The FPASSA-ANFIS starts by formatting the input data in a time series form. In our case, the autocorrelation function (ACF) was considered. ACF is one of the methods applied to find patterns in the data; it presents information about the correlation between points separated by various time lags. Therefore, in this paper, the variables with ACF greater than 0.2 are considered i.e., 5-lags. Besides, the training data contains 75% of the dataset, whereas the testing data contains 25% of them. The number of clusters is defined by the fuzzy c-mean (FCM) method to construct the ANFIS model. The parameters of the ANFIS model are prepared by the FPASSA algorithm. In the training phase, the calculation error (as in Equation (13)) between the real data and the predicted data is used to evaluate the parameters' quality. where T is the real data, and P is the predicted data. N s is the sample length. The smaller values of the objective function indicate good ANFIS's parameter. On the other hand, the updating phase of the followers' positions in the SSA algorithm is applied to improve the global pollination phase in the FPA algorithm. In this improvement, there is a random variable (r) used to switch between both phases. If r > 0.5, then the operators of the SSA is used; otherwise, the operators of the FPA are used. In general, The FPASSA starts by constructing the population (X); afterward, the objective function is calculated for each solution. The solution with the lowest error value is saved to the next iteration. This sequence is repeated until meeting the stop condition, which in this paper, is the maximum number of iterations. Then the best solution is passed to train the parameters of the ANFIS model. After finishing the training phase, the testing phase is started with the best solution to compute the final output. The performance of the proposed method is evaluated by comparing the real data with the predicted data using the performance measures. Finally, the FPASSA produces a foretasted value for confirmed cases of COVID-19 in China in the next day. The steps of the proposed FPASSA are presented in Algorithm 1. Input: Historical COVID-19 dataset, size of population N, total number of iterations t max . Divide the data into training and testing sets. Using Fuzzy c-mean method to determine the number of membership functions. Constructing the ANFIS network. Set the initial value for N solutions (X). Return the best solution that represents the best configuration for ANFIS. Apply the testing set to the best ANFIS model. Forecasting the COVID-19 for the next ten days. This section presents the description of the used dataset, the performance measures, the parameter setting for all methods, the experiment results, and discussions. The main dataset of this study is COVID-19 dataset. It was collected from the WHO website (https: //www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports/). It contains the daily confirmed cases in China from 21 January 2020 to 18 February 2020, as shown in Table 1 . We used 75% from the dataset to train the model while the rest is used to test it. Moreover, we evaluated the performance of the proposed method using two datasets of weekly influenza confirmed cases. The first one is called DS1; it was collected from the Centers for Disease Control and Prevention (CDC) (https://www.cdc.gov/flu/weekly/). It starts from week number 40 in 2015 and continues until week number 6 in 2020. Whereas, the second one is called DS2. It was collected from the WHO website (https://www.who.int/influenza). It contains the data of weekly influenza confirmed cases in China from week number 1 in 2016 to week number 8 in 2020. The quality of the proposed method is evaluated using a set of performance metrics as follows:\n\n• Root Mean Square Error (RMSE):\n\nwhere Yp and Y are the predicted and original values, respectively. • Mean Absolute Error (MAE):\n\n• Mean Absolute Percentage Error (MAPE):\n\n• Root Mean Squared Relative Error (RMSRE):\n\nN s represents the sample size of the data. • Coefficient of Determination (R 2 ):\n\nwhere Y represents the average of Y. The lowest value of RMSE, MAE, MAPE, and RMSRE refers to the best method. The higher value of R 2 indicates better correlation for the method. This paper aims to assess the ability of the FPASSA to forecast the COVID-19 by comparing its performance with other methods, namely the ANFIS and the trained ANFIS models using PSO, GA, ABC, FPA, and FPASSA. The parameters' setting for these models is listed in Table 2 . The common parameters, such as population size, are set to 25 and 100 iterations are applied. Besides, each algorithm is performed for 30 independent runs to fair comparisons.\n\nQuestion: What does ANFIS offer?", "role": "user" }, { "content": "flexibility in determining nonlinearity in the time series data, as well as combining the properties of both artificial neural networks (ANN) and fuzzy logic systems.", "role": "assistant" } ]
science.bioasq_general_qa
science.bioasq_general_qa.1027
[ { "content": "You will be shown a question, followed by excerpts from biomedical research papers. Please answer the question based on the provided context. Do not include any text in your response other than the answer.\n\nQuestion: What is CIS43LS?\n\nContext: . On August 26, 2021, the findings from a phase 1 trial on a new monoclonal antibody to PfCSP, CIS43LS, showed that a single dose of the CIS43LS monoclonal antibody resulted in protection against malaria.\nBased on these data, CIS43LS has advanced to phase I clinical trials, and AAV delivery provides a potential next-generation approach for malaria prevention.\nMETHODS: We conducted a two-part, phase 1 clinical trial to assess the safety and pharmacokinetics of CIS43LS, an antimalarial monoclonal antibody with an extended half-life, and its efficacy against infection with Plasmodium falciparum.\nCONCLUSIONS: Among adults who had never had malaria infection or vaccination, administration of the long-acting monoclonal antibody CIS43LS prevented malaria after controlled infection.\nion or vaccination, administration of the long-acting monoclonal antibody CIS43LS prevented malaria after controlled infection. (Funded by the National Institute of Allergy and Infectio\ne findings from a phase 1 trial on a new monoclonal antibody to PfCSP, CIS43LS, showed that a single dose of the CIS43LS monoclonal antibody resulted in protection against malaria. These new findings have\nby malaria.METHODS: We conducted a two-part, phase 1 clinical trial to assess the safety and pharmacokinetics of CIS43LS, an antimalarial monoclonal antibody with an extended half-life, and its efficacy against infection with Pl\nBACKGROUND: Additional interventions are needed to reduce the morbidity and mortality caused by malaria.METHODS: We conducted a two-part, phase 1 clinical trial to assess the safety and pharmacokinetics of CIS43LS, an antimalarial monoclonal antibody with an extended half-life, and its efficacy against infection with P\nOn August 26, 2021, the findings from a phase 1 trial on a new monoclonal antibody to PfCSP, CIS43LS, showed that a single dose of the CIS43LS monoclonal antibody resulted in protection against malaria.\nlled human malaria infection. Two participants who received 40 mg per kilogram of CIS43LS and underwent controlled human malaria infection approximately 36 weeks later had no parasitemia, with serum concentrations of CIS43LS of 46 and 57 μg per milliliter at the time of controlled human malaria infection.CONCLUSIONS: Among adults who had never had malaria infection or vaccination, administration of the long-acting monoclonal antibody CIS43LS pr", "role": "user" }, { "content": "CIS43LS is an antimalarial monoclonal antibody with an extended half-life against infection with Plasmodium falciparum.", "role": "assistant" } ]
science.bioasq_yesno_qa
science.bioasq_yesno_qa.562
[ { "content": "You have to answer a biomedical question in binary format, i.e., only yes and no are acceptable answers. A list of paragraphs is provided as context to help you answer the question.\n\nQuestion: Can glyburide reduce cerebral edema?\n\nContext: Preclinical studies have shown that a continuous infusion of glyburide blocks edema formation and improves outcome. We hypothesize that treatment with RP-1127 (Glyburide for Injection) reduces formation of brain edema in patients after large anterior circulation infarction.\nCONCLUSIONS: GAMES-RP was designed to provide critical information regarding glyburide for injection in patients with large hemispheric stroke and will inform the design of future studies.\nGlyburide is associated with attenuated vasogenic edema in stroke patients.\nGlyburide is reported to prevent brain swelling in preclinical rodent models of ischemic stroke through inhibition of a non-selective channel composed of sulfonylurea receptor 1 and transient receptor potential cation channel subfamily M member 4.\nRESULTS: We report that IV glyburide was associated with T2 fluid-attenuated inversion recovery signal intensity ratio on brain MRI, diminished the lesional water diffusivity between days 1 and 2 (pseudo-normalization), and reduced blood MMP-9 level.CONCLUSIONS: Several surrogate markers of vasogenic edema appear to be reduced in the setting of IV glyburide treatment in human stroke.\nPilot study of intravenous glyburide in patients with a large ischemic stroke.\nBACKGROUND AND PURPOSE: Preclinical and retrospective clinical data indicate that glyburide, a selective inhibitor of sulfonylurea receptor 1-transient receptor potential melastatin 4, is effective in preventing edema and improving outcome after focal ischemia.\nPreclinical data suggest that glyburide, an inhibitor of SUR1-TRPM4, is effective in preventing edema.\nGlyburide in Treating Malignant Cerebral Edema.\nGlyburide in Treating Malignant Cerebral Edema. Blocking Sulfonyl Urea One (SUR1) Receptors.\nThe sulfonylurea receptor 1-regulated NC(Ca-ATP) channel is upregulated in rodent models of stroke with block of the channel by the sulfonylurea, glibenclamide (glyburide), significantly reducing mortality, cerebral edema, and infarct volume.\nGlyburide is a safe, inexpensive, and efficacious alternative to dexamethasone for the treatment of cerebral metastasis-related vasogenic edema.\nIn this focused review, we explore preclinical data linking Sur1 channel formation to development of edema and reference evidence suggesting that the antidiabetic sulfonylurea drug glyburide (a Sur1 inhibitor) is an inexpensive and well-tolerated agent that can be clinically tested to reduce or prevent malignancy and/or treatment-associated edema.\nPotential of glyburide to reduce intracerebral edema in brain metastases.\nGlyburide Advantage in Malignant Edema and Stroke (GAMES-RP) Trial: Rationale and Design.\nPreclinical data suggest that glyburide, an inhibitor of SUR1-TRPM4, is effective in preventing edema.\nInhibition of sulfonylurea receptor 1 (SUR1) by glyburide has been shown to decrease edema after subarachnoid hemorrhage.\nGlyburide is a safe, inexpensive, and efficacious alternative to dexamethasone for the treatment of cerebral metastasis-related vasogenic edema\nRESULTS: We report that IV glyburide was associated with T2 fluid-attenuated inversion recovery signal intensity ratio on brain MRI, diminished the lesional water diffusivity between days 1 and 2 (pseudo-normalization), and reduced blood MMP-9 level.\nPreclinical data suggest that glyburide, an inhibitor of SUR1-TRPM4, is effective in preventing edema.\nExploratory analysis of glyburide as a novel therapy for preventing brain swelling.\nInhibition of sulfonylurea receptor 1 (SUR1) by glyburide has been shown to decrease edema after subarachnoid hemorrhage.\nIn this focused review, we explore preclinical data linking Sur1 channel formation to development of edema and reference evidence suggesting that the antidiabetic sulfonylurea drug glyburide (a Sur1 inhibitor) is an inexpensive and well-tolerated agent that can be clinically tested to reduce or prevent malignancy and/or treatment-associated edema.\nGlyburide is a safe, inexpensive, and efficacious alternative to dexamethasone for the treatment of cerebral metastasis-related vasogenic edema.\nGlyburide is reported to prevent brain swelling in preclinical rodent models of ischemic stroke through inhibition of a non-selective channel composed of sulfonylurea receptor 1 and transient receptor potential cation channel subfamily M member 4.\nSeveral surrogate markers of vasogenic edema appear to be reduced in the setting of IV glyburide treatment in human stroke.\nWe hypothesize that treatment with RP-1127 (Glyburide for Injection) reduces formation of brain edema in patients after large anterior circulation infarction.\nGlyburide is a safe, inexpensive, and efficacious alternative to dexamethasone for the treatment of cerebral metastasis-related vasogenic edema..\nGlyburide is associated with attenuated vasogenic edema in stroke patients.", "role": "user" }, { "content": "yes", "role": "assistant" } ]
science.chia_ner
science.chia_ner.1627
[ { "content": "You will be shown inclusion or exclusion criteria for a clinical trial. Given these criteria, your task is to extract all unique entities of the following types: [\"Person\", \"Condition\", \"Value\", \"ReferencePoint\", \"Temporal\", \"Qualifier\", \"Observation\", \"Visit\", \"Measurement\", \"Mood\", \"Device\", \"Scope\", \"Negation\", \"Drug\", \"Procedure\", \"Multiplier\"].\n\nPlease return the output as a JSON object of the format: {\"Condition\": [\"breast cancer\", ...], \"Device\": [\"ports\", ...], \"Drug\": [\"Atenolol\", ...], \"Measurement\": [\"blood pressure\", ...], \"Mood\": [\"suspected\", ...], \"Multiplier\": [\"more than two\", ...], \"Negation\": [\"no\", ...], \"Observation\": [\"history\", ...], \"Person\": [\"male\", ...], \"Procedure\": [\"surgery\", ...], \"Qualifier\": [\"severe\", ...], \"ReferencePoint\": [\"enrollment\", ...], \"Scope\": [\"infection\", ...], \"Temporal\": [\"one day\", ...], \"Value\": [\"50%\", ...], \"Visit\": [\"ICU\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nAge 10 to 65 years\nTemperature less than 100.4 F\nNormal neurologic exam and normal mental status", "role": "user" }, { "content": "{\"Condition\": [], \"Device\": [], \"Drug\": [], \"Measurement\": [\"Temperature\", \"mental status\"], \"Mood\": [], \"Multiplier\": [], \"Negation\": [], \"Observation\": [], \"Person\": [\"Age\"], \"Procedure\": [\"neurologic exam\"], \"Qualifier\": [], \"ReferencePoint\": [], \"Scope\": [], \"Temporal\": [], \"Value\": [\"10 to 65 years\", \"Normal\", \"normal\", \"less than 100.4 F\"], \"Visit\": []}", "role": "assistant" } ]
science.bc7_litcovid_topic_classification
science.bc7_litcovid_topic_classification.18422
[ { "content": "You will be presented with a paragraph from a COVID-19 literature. Your task is to classify the topics of this piece of text by choosing from the following categories:\n\n- Epidemic Forecasting: The content relates to predictions about the spread of COVID-19.\n- Treatment: The paragraph discusses therapeutic methods or medication for COVID-19.\n- Prevention: The content covers measures to prevent the spread of COVID-19.\n- Mechanism: The paragraph explains the biological mechanisms of the virus.\n- Case Report: The content is a report of a COVID-19 case.\n- Transmission: The paragraph discusses how COVID-19 is transmitted.\n- Diagnosis: The content is about diagnosing COVID-19 in individuals.\n\nYour classification should be provided as an element within a JSON array. The array should be formatted with topics enclosed in double quotes and separated by commas, following the JSON format, for example: [\"Topic1\", \"Topic2\", ...]. Include all applicable categories within the array. \nYour output should be a valid JSON array, and no additional text should be included in your response.\n\nArticle:\n\nBased on the rapid spread of the CoViD-2019, a lockdown was declared in the whole Northern Italy by the Government. The application of increasingly rigorous containment measures allowed to reduce the impact of the CoViD-2019 pandemic on the Italian National Health System but at the same time these restriction measures gave also the opportunity to assess the effect of anthropogenic activities on air pollutants in an unprecedented way. This paper aims to study the impact of the partial and total lockdown (PL and TL, respectively) on air quality in the Metropolitan City of Milan. As results, the severe limitation of people movements following the PL and the subsequent TL determined a significant reduction of pollutants concentration mainly due to vehicular traffic (PM10, PM2.5, BC, benzene, CO, and NOx). The lockdown led to an appreciable drop in SO2 only in the city of Milan while it remained unchanged in the adjacent areas. Despite the significant decrease in NO2 in the TL, the O3 exhibited a significant increase, probably, due to the minor NO concentration. In Milan and SaA the increase was more accentuated, probably, due to the higher average concentrations of benzene in Milan than the adjacent areas that might have promoted the formation of O3 in a more significant way.", "role": "user" }, { "content": "[\"Prevention\"]", "role": "assistant" } ]
science.scientific_papers_summarization_single_doc_pubmed
science.scientific_papers_summarization_single_doc_pubmed.100803
[ { "content": "You will be presented with the full text of a biomedical research paper. Please write an abstract for this paper. Your response should only include the abstract and no additional text.\n\nPaper text:\nthe neoplastic process is asymptomatic until the tumor starts exerting pressure on the surrounding organs , impairing their function .\nthe primary focus may develop in the uterus , while the most common locations of secondary tumors include the retroperitoneal space [ 2 , 3 ] .\nmetastases in the area of the heart are not typical of leiomyosarcoma , but such cases have also been reported [ 411 ] .\nthe patient was an obese woman , aged 61 , with a long history of hypertension , treated with continuous positive airway pressure ( cpap ) for obstructive sleep apnea .\nher medical history included incidents of angioedema during attempts at antihypertensive treatment with angiotensin - converting enzyme ( ace ) inhibitors . from 2004 ,\nthe patient underwent numerous surgical procedures ( extirpation of the uterus , removal of a tumor in the right lumbar area , removal of a tumor in the right thigh , removal of a tumor in the left arm ) .\nthe patient was referred to the cardiology clinic after a cardiac tumor was revealed in follow - up echocardiography . the tumor with jagged edges ( 35 35 mm ) filled the apical portion of the left ventricle .\ncardiac diagnostics was conducted due to the gradually deteriorating exercise tolerance , even though the patient 's fatigability could have been attributed to her age , body type , inactive lifestyle , and long - term hypertension with left ventricular hypertrophy leading to diastolic heart failure .\nregular cardiac function in physical examination ( 76/min ) , quiet systolic murmur in the apex and erb 's point , no auscultatory changes over the lung fields , trace of peripheral swelling , and normal blood pressure with the currently applied treatment were noted . after a cardiological consultation ,\ndue to the fact that the echocardiographic image clearly corresponded to a metastatic course of leiomyosarcoma , other methods of diagnostic imaging were not utilized ( fig .\nthe patient was operated on with cardiopulmonary bypass , under systemic hypothermia ( 28c ) , using crystalloid cardioplegia .\nthe left ventricle was opened on the anterior wall between the posterior interventricular artery and the diagonal branch .\na large tumor ( 3 3 cm ) was removed from the wall of the left ventricle with the coagulation of the cut edges .\naortic cross - clamping time during the procedure was 19 minutes , while cardiopulmonary bypass time was 45 minutes .\nthe early postoperative period at the intensive care unit was extended due to low cardiac output syndrome ( ef 26% ) , requiring the inclusion of catecholamines , and bradyarrhythmia , necessitating temporary external cardiac pacing .\nthe patient was moved to the cardiology clinic on the 15 postoperative day ( fig . 2 and 3 ) .\n100/min , blood pressure 135/85 mmhg , crackles at the base of both lungs , traces of lower leg swelling , lack of postcardiotomy syndrome in echocardiography .\nsigns of heart failure requiring the implementation of an appropriate treatment were dominant during subsequent hospitalization . due to the presence of angioedema in the patient 's medical history ,\nace - i ( angiotensin - converting - enzyme inhibitor ) treatment was not applied .\na follow - up echo examination revealed a slight improvement in left ventricular ejection fraction ( lvef 30% ) .\nthe tumor is marked with an arrow echocardiographic image taken on the 15 day after tumor resection , apical four - chamber view echocardiographic image taken on the 15 day after tumor resection , tissue doppler imaging .\nhigh e / e ratio indicates increased left ventricular filling pressure the applied pharmacological treatment stabilized the patient ( nyha iii ) ; a six - minute walk test with exercise tolerance at the level of 280 m was conducted before discharge . due to the radical nature of the tumor resection , no supplementary methods of treatment were implemented .\nthe period of outpatient treatment , including cardiac rehabilitation , was uneventful ; the condition of the patient was stable , with a gradual improvement of physical performance ( nyha iii nyha ii ) .\nafter 3 months , a follow - up echocardiographic examination showed a further improvement of left ventricular ejection fraction ( lvef 38% ) .\ncurrently , the long - term follow - up exceeds 12 months ; no new neoplastic foci were found in echo and computed tomography ; left ventricular ejection fraction remains stable ( lvef 40% ) .\nthe good clinical condition of the patient enables her to conduct her daily activities at a level sufficient for her needs .\nthe most common benign cardiac tumors include myxoma ( approximately half of all cases ) , papillary fibroelastoma , and lipoma ( 16% for each type ) .\nsarcomas constitute the dominant group of malignant tumors ( 95% ) , the most common being angiosarcoma ( 28% ) .\nit is important to emphasize that metastatic heart tumors are over 20 times more frequent than primary tumors and are found in 10 - 12% of patients with neoplastic diseases .\nheart metastasis occurs most frequently in the course of lung , breast , and kidney cancers as well as in the case of lymphomas and melanoma .\nthe prevalence of leiomyosarcoma in the adult population is estimated at 5% of all cardiac tumors and 1% of all neoplasms .\nleiomyosarcoma is most commonly localized in the retroperitoneal space ( 19% ) , but it may also develop in the uterus , stomach , intestines , skin , and vascular walls .\nleiomyosarcoma of the heart is usually located in the left atrium , but it may also occur in the left ventricle , as in the case presented above .\nit is assumed that some of them may develop from the smooth muscles of pulmonary veins . due to their oligosymptomatic course ,\ncardiac tumors are underdiagnosed ; diagnosis is established intraoperatively or during an autopsy . the development of medical imaging methods enabled earlier diagnosis of cardiac tumors .\ndue to the slow and asymptomatic growth and lack of specific biochemical markers , the main role in diagnostics is played by imaging methods , mainly ultrasound . in the case of cardiac tumors ,\nan echocardiographic diagnosis of a cardiac tumor requires differentiation ; apart from tumors , it is necessary to consider thrombi or bacterial vegetation . the final diagnosis is frequently reached through correct differentiation based on the patient 's disease history as well as tumor location and structure .\nother methods of medical imaging used in the diagnostics of cardiac tumors include transesophageal echocardiography and computed tomography .\nthe results of imaging examinations enable the assessment of the anatomical condition in the area of the tumor , which allows for the establishment of the extent of myocardial , pericardial , and extracardiac involvement .\npossible clinical symptoms : tumors in cardiac cavities usually cause heart failure with its typical symptomatology , while cardiac dysrhythmia may be expected in the case of intramural tumors .\nthe clinical manifestation of cardiac tumors is dependent on the influence of tumor location and size on the hemodynamics of the heart .\nmoreover , left heart tumors may cause embolisms in the systemic circulation , in the form of ischemic stroke , infarctions of abdominal organs , and peripheral embolisms in extremities . in advanced forms of the disease ,\ntumor growth is frequently accompanied by the presence of bloody exudate in the pericardial sac , which may result in the development of life - threatening tamponade .\nthe general operative indications for cardiac tumors result from the hemodynamic disorders associated with tumor growth , increased risk of embolism , and cardiac dysrhythmia .\nthe main factors determining the results of cardiac surgical treatment include the general condition of the patient , the stage of the underlying neoplastic process ( infiltration by continuity ) , and the occurrence of clinically significant distant metastases ( crucial for prognosis ) .\ntherefore , surgical treatment is rarely applied , and long - term surgical treatment results are poor . in the case of the present patient ,\nthe ailments were associated with heart failure in nyha class ii ; however , as a result of the irregular , jagged surface of the tumor revealed in echocardiography , this form of the disease was characterized by increased risk of embolic complications .\nthe condition of the patient , anatomy of the tumor , and lack of metastases to other organs at the current stage of the disease enabled the qualification of the patient for surgical treatment . in the case of malignant tumors , a surgical procedure is not a fully effective method of treatment , but isolated cases of lasting recovery after radical tumor resection have been described in the literature .\nthe factors determining unfavorable prognosis include incomplete tumor resection and high grade of tumor differentiation ( g3 ) leading to the rapid spread of metastases , mainly to the lungs and the brain .\nthe effectiveness of adjuvant radio- and chemotherapy in the treatment of leiomyosarcoma has not yet been explicitly established , while the applied therapeutic schemes were transferred from the treatment of soft tissue sarcomas . in the present case ,\nthe occurrence of low cardiac output syndrome , treated with a catecholamine infusion , and symptomatic bradyarrhythmia , requiring temporary external cardiac pacing , constituted significant problems in the early postoperative period .\nthe complications were associated with the extent of the procedure , the necessary coagulation of the tumor resection line with subsequent necrosis of the adjacent tissue , and postoperative myocardial stunning .\nthe applied pharmacological treatment and cardiac rehabilitation enabled the reduction of heart failure symptoms to nyha class ii , with a gradual improvement of left ventricular ejection fraction . a standard management\nafter cardiac tumor resection is yet to be established ; the prognosis is not certain due to the risk of local recurrence , occurring in half of the patients undergoing surgery , as well as the appearance of metastases , observed in 20 - 40% of cases . in cases of primary cardiac tumors and incomplete tumor resection , the average time of survival\nis estimated at 6 to 12 months , while , in the case of radical tumor resection , various reports estimate it at 24 - 45 months ; radical procedures may consist in autotransplantation with the reconstruction of the removed elements of the heart .\nthe majority of reports concerning cardiac surgery and adjuvant chemotherapy , however , estimate the average survival time at approximately 18 months .\nsarcomas are characterized by the recurrence of hyperplasia , which necessitates regular clinical monitoring with diagnostic imaging methods ( watchful waiting strategy ) . in the case of leiomyosarcoma , it is recommended to carefully consider the possibility of local recurrence and metastasis to the brain and lungs .\ncurrently , the long - term follow - up of the present patient exceeds 12 months .\nthe good clinical condition of the patient enables daily activity at a level sufficient for her needs .", "role": "user" }, { "content": "we present a case report of a 60-year - old woman with a long history of leiomyosarcoma in different locations . \n she was admitted to the clinic due to a left ventricular tumor diagnosed in echo examination . \n the patient was qualified for radical tumor resection . \n the early postoperative period was complicated due to low cardiac output syndrome and bradyarrhythmia requiring temporary cardiac pacing . \n optimized pharmacological therapy resulted in a gradual reduction of symptoms and a clinical improvement of congestive heart failure ( nyha iii nyha ii ) . due to the radical nature of the surgery , \n the patient was not referred for supplementary treatment . \n the follow - up currently exceeds 12 months no new metastases have been found . \n this case provides an example of how to diagnose and treat heart tumors .", "role": "assistant" } ]
science.bc7_litcovid_topic_classification
science.bc7_litcovid_topic_classification.18820
[ { "content": "You will be presented with a paragraph from a COVID-19 literature. Your task is to classify the topics of this piece of text by choosing from the following categories:\n\n- Epidemic Forecasting: The content relates to predictions about the spread of COVID-19.\n- Treatment: The paragraph discusses therapeutic methods or medication for COVID-19.\n- Prevention: The content covers measures to prevent the spread of COVID-19.\n- Mechanism: The paragraph explains the biological mechanisms of the virus.\n- Case Report: The content is a report of a COVID-19 case.\n- Transmission: The paragraph discusses how COVID-19 is transmitted.\n- Diagnosis: The content is about diagnosing COVID-19 in individuals.\n\nYour classification should be provided as an element within a JSON array. The array should be formatted with topics enclosed in double quotes and separated by commas, following the JSON format, for example: [\"Topic1\", \"Topic2\", ...]. Include all applicable categories within the array. \nYour output should be a valid JSON array, and no additional text should be included in your response.\n\nArticle:\n\nCOVID-19 coagulopathy is a hypercoagulable state which predisposes to venous, arterial and small vessel thrombosis. We describe a patient with COVID-19 who developed an acute superior mesenteric vein thrombosis with acute intestinal obstruction despite adequate anticoagulation.", "role": "user" }, { "content": "[\"Case Report\"]", "role": "assistant" } ]
science.scierc_re
science.scierc_re.218
[ { "content": "You will be given an abstract from a computer science research paper. Given the abstract, your task is to extract all unique relationships between entities of the following, and ONLY the following, types:\n\n- \"USED-FOR\": For instance, \"Our method models user proficiency\" -> [\"method\", \"user proficiency\", \"USED-FOR\"].\n- \"EVALUATE-FOR\": For instance, \"We evaluate our method using F1-measure\" -> [\"F1-measure\", \"method\", \"EVALUATE-FOR\"].\n- \"FEATURE-OF\": For instance, \"prior knowledge of the model\" -> [\"prior knowledge\", \"model\", \"FEATURE-OF\"].\n- \"HYPONYM-OF\": For instance, \"TUIT is a software library\" -> [\"TUIT\", \"software library\", \"HYPONYM-OF\"].\n- \"PART-OF\": For instance, \"We incorporate NLU module into the system\" -> [\"NLU module\", \"system\", \"PART-OF\"].\n- \"COMPARE\": For instance, \"Unlike the quantitative prior, the qualitative prior is often ignored\" -> [\"quantitative prior\", \"qualitative prior\", \"COMPARE\"].\n- \"CONJUNCTION\": For instance, \"obtained from human expert or knowledge base\" -> [\"human expert\", \"knowledge base\", \"CONJUNCTION\"].\n\nFormat your output as a JSON array. Each entry in the array should express a single relation, formatted as [\"<Entity_A>\", \"<Entity_B>\", \"<RELATION_A_B>\"]. An example output might look like:\n\n[[\"neural networks\", \"classification\", \"USED-FOR\"], [\"neuron\", \"neural network\", \"PART-OF\"]].\n\nIf no relations are found, please return an empty JSON array \"[]\".\n\nAbstract:\n\nAn efficient bit-vector-based CKY-style parser for context-free parsing is presented. The parser computes a compact parse forest representation of the complete set of possible analyses for large treebank grammars and long input sentences. The parser uses bit-vector operations to parallelise the basic parsing operations. The parser is particularly useful when all analyses are needed rather than just the most probable one.", "role": "user" }, { "content": "[[\"bit-vector operations\", \"bit-vector-based CKY-style parser\", \"USED-FOR\"], [\"parse forest representation\", \"large treebank grammars\", \"USED-FOR\"], [\"bit-vector-based CKY-style parser\", \"context-free parsing\", \"USED-FOR\"], [\"bit-vector-based CKY-style parser\", \"parse forest representation\", \"USED-FOR\"]]", "role": "assistant" } ]
science.acl_arc_intent_classification
science.acl_arc_intent_classification.1595
[ { "content": "You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories:\n- Background: provides context or foundational information related to the topic.\n- Extends: builds upon the cited work.\n- Uses: applies the methods or findings of the cited work.\n- Motivation: cites the work as inspiration or rationale for the research.\n- CompareOrContrast: compares or contrasts the cited work with others.\n- FutureWork: cites the work as a direction for future research.\n\nYour answer should be a single word from the following list of options: [\"Background\", \"Extends\", \"Uses\", \"Motivation\", \"CompareOrContrast\", \"FutureWork\"]. Do not include any other text in your response.\n\nSection Title:\nrelated work\n\nContext before the citation:\nThe language grounding problem has received significant attention in recent years, owed in part to the wide availability of data sets (e.g. Flickr, Von Ahn (2006)), computing power, improved computer vision models (Oliva and Torralba, 2001; Lowe, 2004; Farhadi et al., 2009; Parikh and Grauman, 2011) and neurological evidence of ties between the language, perceptual and motor systems in the brain (Pulverm¨uller et al., 2005; Tettamanti et al., 2005; Aziz-Zadeh et al., 2006).\n\nCitation Sentence:\nThe language grounding problem has received significant attention in recent years , owed in part to the wide availability of data sets ( e.g. Flickr , Von Ahn ( 2006 ) ) , computing power , improved computer vision models ( Oliva and Torralba , 2001 ; Lowe , 2004 ; Farhadi et al. , 2009 ; Parikh and Grauman , 2011 ) and neurological evidence of ties between the language , perceptual and motor systems in the brain ( Pulverm ¨ uller et al. , 2005 ; Tettamanti et al. , 2005 ; Aziz-Zadeh et al. , 2006 ) .\n\nContext after the citation:\nMany approaches to multimodal research have succeeded by abstracting away raw perceptual information and using high-level representations instead. Some works abstract perception via the usage of symbolic logic representations (Chen et al., 2010; Chen and Mooney, 2011; Matuszek et al., 2012; Artzi and Zettlemoyer, 2013), while others choose to employ concepts elicited from psycholinguistic and cognition studies. Within the latter category, the two most common representations have been association norms, where subjects are given a \u0000 1http://stephenroller.com/research/ emnlp13", "role": "user" }, { "content": "Background", "role": "assistant" } ]
science.qasa_abstractive_qa
science.qasa_abstractive_qa.231
[ { "content": "You will be shown an excerpt from a computer science research paper, followed by a question about the paper. Please answer the question given the context. Do not include any text in your response other than the answer.\n\nContext: However, recent advances of image classification focus on training feedforward convolutional neural networks using “very deep” structure [27, 33, 10]. The feedforward convolutional network mimics the bottom-up paths of human cortex. Various approaches have been proposed to further improve the discriminative ability of deep convolutional neural network. VGG [27], Inception [33] and residual learning [10] are proposed to train very deep neural networks. Stochastic depth [14], Batch Normalization [15] and Dropout [28] exploit regularization for convergence and avoiding overfitting and degradation. The design of soft attention structure in our Residual Attention Network is inspired by recent development of localization oriented task, i.e. segmentation [22, 25, 1] and human pose estimation [24]. These tasks motivate researchers to explore structure with fined-grained feature maps. The frameworks tend to cascade a bottom-up and a top-down structure. The bottom-up feedforward structure produces low resolution feature maps with strong semantic information. After that, a top-down network produces dense features to inference on each pixel. Skip connection [22] is employed between bottom and top feature maps and achieved state-of-the-art result on image segmentation. The recent stacked hourglass network [24] fuses information from multiple scales to predict human pose, and benefits from encoding both global and local information. Following previous attention mechanism idea in DBN [21], our mask branch contains fast feed-forward sweep and top-down feedback steps. The former operation quickly collects global information of the whole image, the latter operation combines global information with original feature maps. In convolutional neural network, the two steps unfold into bottom-up top-down fully convolutional structure. (3) Bottom-up top-down feedforward attention: Bottom-up top-down feedforward structure has been successfully applied to human pose estimation [24] and image segmentation [22, 25, 1]. We use such structure as part of Attention Module to add soft weights on features. This structure can mimic bottom-up fast feedforward process and top-down attention feedback in a single feedforward process which allows us to develop an end-to-end trainable network with top-down attention. The bottom-up top-down structure in our work differs from stacked hourglass network [24] in its intention of guiding feature learning.\n\nQuestion: What does \"bottom-up top-down feedforward structure\" means ?", "role": "user" }, { "content": "The bottom-up top-down feedforward structure is a combination of a bottom-up fast feedforward process that creates low resolution features maps to quickly collect global information, and a top-down attention feedback process that uses the global information along with the original feature maps to create features for inference.", "role": "assistant" } ]
science.scientific_lay_summarisation_plos_single_doc_summ
science.scientific_lay_summarisation_plos_single_doc_summ.9431
[ { "content": "You will be presented with the abstract, introduction, and discussion section from a biomedical research article. Your task is to create a summary that a layperson can understand, capturing the essence of the research article. Keep essential scientific terms, but ensure that the language remains clear and the concepts are explained in an uncomplicated manner.\n\nTitle: No Reliable Association between Runs of Homozygosity and Schizophrenia in a Well-Powered Replication Study\nArticle:\nAbstract:\nIt is well known that inbreeding increases the risk of recessive monogenic diseases, but it is less certain whether it contributes to the etiology of complex diseases such as schizophrenia. One way to estimate the effects of inbreeding is to examine the association between disease diagnosis and genome-wide autozygosity estimated using runs of homozygosity( ROH) in genome-wide single nucleotide polymorphism arrays. Using data for schizophrenia from the Psychiatric Genomics Consortium( n = 21, 868), Keller et al.( 2012) estimated that the odds of developing schizophrenia increased by approximately 17% for every additional percent of the genome that is autozygous( β = 16. 1, CI( β) =[6. 93, 25. 7], Z = 3. 44, p = 0. 0006). Here we describe replication results from 22 independent schizophrenia case-control datasets from the Psychiatric Genomics Consortium( n = 39, 830). Using the same ROH calling thresholds and procedures as Keller et al.( 2012), we were unable to replicate the significant association between ROH burden and schizophrenia in the independent PGC phase II data, although the effect was in the predicted direction, and the combined( original + replication) dataset yielded an attenuated but significant relationship between Froh and schizophrenia( β = 4. 86, CI( β) =[0. 90, 8. 83], Z = 2. 40, p = 0. 02). Since Keller et al.( 2012), several studies reported inconsistent association of ROH burden with complex traits, particularly in case-control data. These conflicting results might suggest that the effects of autozygosity are confounded by various factors, such as socioeconomic status, education, urbanicity, and religiosity, which may be associated with both real inbreeding and the outcome measures of interest.\nIntroduction:\nClose inbreeding( e. g., cousin-cousin mating) is known to decrease fitness in animals[1] and to increase risk for recessive Mendelian diseases in humans[2], a phenomenon known as inbreeding depression. Inbreeding depression is thought to occur due to evolutionary selection against genetic variants that decrease fitness—e. g., variants that increase risk of disorders[3]. Such fitness-reducing variants should not only be more rare, but also more recessive than expected under a neutral evolution model( i. e., show directional dominance). If so, individuals with a greater proportion of their genome in autozygous stretches( two homologous segments of a chromosome inherited from a common ancestor identical by descent[IBD]) should have higher rates of disorders. This is because autozygous regions reveal the full, harmful effects of any deleterious, recessive alleles that existed on the haplotype of the common ancestor. Whether inbreeding increases risk for complex disorders like schizophrenia is less clear. Previous studies have found that inbreeding is associated with higher rates of complex disorders[4–9]. However, sample sizes have typically been small and the possibility that confounding factors might explain the results has left the links inconclusive. Moreover, close inbreeding accounts for fewer than 1% of marriages in industrialized countries[10], and information on pedigrees going back many generations is difficult to collect reliably. For these reasons, investigators have recently begun looking at signatures of very distant inbreeding( e. g., common ancestry up to ~100 generations ago) using genome-wide single nucleotide polymorphism( SNP) data in an attempt to understand whether autozygosity increases the risk to schizophrenia and other complex diseases[11]. Autozygosity in SNP data is typically inferred from runs of homozygosity( ROHs): long, contiguous stretches( e. g., > 40) of homozygous SNPs. The proportion of the genome contained in such ROHs, Froh, can then be used to predict complex traits[12–19]. Keller et al.[11] showed that Froh is the optimal method for detecting inbreeding signals that are due to rare, recessive to partially recessive mutations, such as those thought to occur when traits are under directional selection[3]. The low variation in Froh means that large sample sizes( e. g., >12, 000) are required to uncover realistic effects of distant inbreeding on complex diseases in samples unselected for inbreeding[11]. In 2012, Keller et al.[20] used the original Psychiatric Genomics Consortium schizophrenia data( 17 case-control datasets, total n = 21, 831) to investigate whether Froh is associated with increased risk of schizophrenia. The authors estimated that the odds of developing schizophrenia increased by approximately 17% for every additional percent of the genome that is contained in autozygous regions( β = 16. 1, CI( β) =[6. 93, 25. 7], p = 6x10-4.) This was by far the largest study to that date examining the association between Froh and any psychiatric disorder, and the significant relationship between Froh and case-control status remained robust through secondary analyses of various covariate combinations, common vs. rare IBD haplotypes, and SNP thresholds used to define ROHs. These results are consistent with the hypothesis that autozygosity causally increases the risk of schizophrenia. Nevertheless, because various confounding factors may increase likelihood of distant inbreeding as well as the probability of having offspring with schizophrenia, these results do not imply a causal relationship. For example, parents higher on schizophrenia liability may pass their higher liability to offspring and mate with more genetically similar partners( e. g., due to decreased mobility, educational opportunities, etc.). The current study seeks to provide a well-powered, independent replication of Keller et al.( 2012)[20]. In light of the growing concern about publication bias[21, 22] and dearth of well-powered replications[23, 24], this follow-up analysis is a necessary step in validating the Froh—schizophrenia relationship. The present study used genome-wide SNP data from 22 independent schizophrenia case-control datasets( n = 39, 830) from the PGC[25] to further examine the relationship between Froh and schizophrenia. Our replication attempt is an important contribution to the growing body of literature examining autozygosity and psychiatric disorders, and should help verify whether autozygosity estimated from ROHs is robustly related to schizophrenia risk and, by extension, can help elucidate whether schizophrenia risk alleles are biased, on average, toward recessive effects.\nDiscussion:\nDespite exploring various homozygous SNP length thresholds, Mb thresholds, and combinations of covariates, the findings from this study do not lend much support to the original observation of a highly significant Froh-schizophrenia association[20], and provide only equivocal support, based on combining the original and replication data, for the hypothesis that autozygosity is a risk factor for schizophrenia. Perhaps the simplest explanation for this pattern of results is that the conclusions about distant inbreeding from the original data represent a type-I error or that the lack of replication in the current report was a type-II error. Despite the fact that the effect in the original study was highly significant( p = 6x10-4) and the statistical power in the replication study to detect the observed effect size in the original study was nearly 100%, it is possible that the estimated effects of the original analysis could have been over-estimated and/or those of the replication analysis under-estimated, due to sampling variability. There is some support for this interpretation, as there was not a significant difference in results between replication versus original datasets( interaction p = 0. 07). An alternative explanation for the overall pattern of results has to do with the potential influence of unmeasured confounding factors in both the original and replication analyses. Unlike genotype frequencies, which change very slowly and are unaffected by inbreeding, ROH levels can change substantially after even a single generation of inbreeding, making ROH analyses highly susceptible to confounding factors associated with both disease risk and the degree of inbreeding/outbreeding. For example, contrary to initial predictions, Abdellaoui et al.[28] identified a significant and negative( “protective”) relationship between Froh and risk for major depressive disorder( MDD) in the Dutch population. However, the authors found that religiosity was significantly associated with both higher autozygosity and lower MDD in this population. When religiosity was accounted for in their regression model, the original association between MDD and Froh disappeared. A similar effect was detected for educational attainment: highly educated individuals were more likely to migrate and mate with highly educated and more diverse partners, making highly educated spouse pairs share less ancestry and leading to their offspring having lower Froh[29]. Thus, assortative mating on variables such as education or religion could subtly influence observed Froh associations, potentially affecting results in ways that can be difficult to account for. For example, an observed Froh-schizophrenia relationship could be due to parents with a higher schizophrenia liability mating with less genetically diverse mates due to, e. g., fewer educational opportunities or lower migration rates. Thus, the causation may be reversed: schizophrenia liability in parents could cause not only higher schizophrenia risk, but also higher Froh, in offspring rather than Froh in offspring increasing their schizophrenia liability. Such reverse and third variable causation possibilities can only be tested if relevant socio-demographic variables in subjects and( optimally) their parents are collected. The possibility of unmeasured variables confounding Froh-disorder relationships seems particularly likely in analyses conducted on ascertained samples. Ascertainment of cases and controls not perfectly matched on socio-demographic factors that might affect degree of outbreeding( e. g., socioeconomic status, education level, age, religion, urbanicity) can mask any true Froh association and bias the observed association in either direction. Such a scenario might explain otherwise contradictory findings in previous ROH case-control analyses[18, 28, 30–36]. For example, following two studies showing that genome-wide autozygosity was significantly associated with schizophrenia risk, including the original Keller et al. study[13, 20], two newer studies failed to replicate this association[34, 35], although both replication sample sizes( n = 3, 400 and 11, 244 respectively) were substantially smaller than the current one( n = 39, 830).( It should be noted that the sample used in the latter study[36] overlapped with the samples in both the original Keller et al.[20] study and the current replication study). Even within the same study, Froh results in ascertained samples have been inconsistent. Using PGC MDD data, Power et al.[36] found a significant positive Froh-MDD relationship in data from three German sites but a significant negative Froh-MDD relationship in six non-German sites. A possible explanation for this and other such examples of heterogeneity across sites they observed is that cases and controls differed on socio-demographic factors that were associated with Froh, and the direction of this ascertainment bias was inconsistent across data collection sites. We believe that similar ascertainment biases could have affected results in the present study as well as in the original Keller et al.[20] report. Many of the PGC schizophrenia datasets used cases ascertained from hospitals, clinics, health surveys, and advertisements but controls from previous biomedical research volunteers, university students, blood donors, and population registries. While such differences in ascertainment between cases and controls are highly unlikely to lead to allele frequency differences, and thus are of little concern to genome-wide association studies, they could very easily lead to Froh differences due to differences in degree of inbreeding/outbreeding in the populations from which cases and controls were drawn. Controlling for ancestry principal components in this case would only help to the degree that degree of inbreeding/outbreeding is associated with ancestry. Unfortunately, none of the other variables that might statistically control for such biases due to differences in case/control ascertainment are currently available in the PGC data collection. The PGC collection of studies was designed for association analyses; it was not optimally designed for ancillary purposes, such as ROH analyses. It is important to recognize that even ascertainment biases that differ at random across sites would substantially inflate type-I error rates because the proper degrees of freedom for the test should be closer to the number of independent sites rather than the number of independent cases and controls. To demonstrate this, we permuted data under the null hypothesis of no relationship between Froh and schizophrenia in the 17 datasets from the original 2012 study by randomly flipping case or control status within each dataset for each permutation( e. g., cases and control statuses in a dataset either remained the same or were flipped to the opposite status). We then calculated the overall Froh ~ schizophrenia relationship with the same logistic regression model and using the same covariates as in the original analysis. Across 1, 000 permutations, 183 p-values were significant( p < 0. 05), implying a type-I error rate of 0. 18 and demonstrating how false conclusions about Froh relationships can be reached even when ascertainment biases are random across multiple sites.", "role": "user" }, { "content": "It is well known that mating between relatives increases the risk that a child will have a rare recessive genetic disease , but there has also been increasing interest and inconsistent findings on whether inbreeding is a risk factor for common , complex psychiatric disorders such as schizophrenia . The best powered study to date investigating this theory predicted that the odds of developing schizophrenia increase by approximately 17% for every additional percent of the genome that shows evidence of inbreeding . In this replication , we used genome-wide single nucleotide polymorphism data from 18 , 562 schizophrenia cases and 21 , 268 controls to quantify the degree to which they were inbred and to test the hypothesis that schizophrenia cases show higher mean levels of inbreeding . Contrary to the original study , we did not find evidence for distant inbreeding to play a role in schizophrenia risk . There are various confounding factors that could explain the discrepancy in results from the original study and our replication , and this should serve as a cautionary note–careful attention should be paid to issues like ascertainment when using the data from genome-wide case-control association studies for secondary analyses for which the data may not have originally been intended .", "role": "assistant" } ]
science.bioasq_general_qa
science.bioasq_general_qa.185
[ { "content": "You will be shown a question, followed by excerpts from biomedical research papers. Please answer the question based on the provided context. Do not include any text in your response other than the answer.\n\nQuestion: How is CBX1/M31 related to position-effect variegation?\n\nContext: First, forced expression of full-length SUV39H1 (412 amino acids) redistributes endogenous M31 (HP1beta) and induces abundant associations with inter- and metaphase chromatin\nTogether, our data reveal a dominant role(s) for the SET domain of SUV39H1 in the distribution of prominent heterochromatic proteins and suggest a possible link between a chromosomal SU(VAR) protein and histone H3\nUsing this motif, termed chromo box, we have cloned a mouse candidate modifier gene, M31, that also shows considerable sequence homology to Drosophila HP1\nHere we report evidence of at least four independently segregating loci in the mouse homologous to the M31 cDNA. One of these loci--Cbx-rs1--maps to the X Chromosome (Chr), 1 cM proximal to Amg and outside the X-inactivation center region\nAlthough there is significant heterochromatic overlap between SUV39H1 and M31 (HP1(beta)) during interphase, mitotic SUV39H1 displays a more restricted spatial and temporal association pattern with metaphase chromosomes than M31 (HP1(beta)), or the related HP1(&agr;) gene product\nwhich complex with the heterochromatin component M31.\nIn addition, Suv39h1/SUV39H1 proteins associate with M31, currently the only other characterized mammalian SU(VAR) homologue\nM31, a murine homolog of Drosophila HP1, is concentrated in the XY body during spermatogenesis\nThe HP1 class of chromobox genes are thought to encode proteins involved in the packaging of chromosomal DNA into repressive heterochromatin domains, as seen, for example, in position-effect variegation\nStudy of the distribution of a murine HP1-like chromodomain protein, M31, during spermatogenesis revealed spreading from the tip of the XY body in mid-stage pachytene spermatocytes to include the whole of the XY body in late-pachytene spermatocytes\nWe also demonstrate that the formation of the XY body during spermatogenic progression in neonatal mice coincides with the expression of a novel nuclear isoform of M31, M31(p21\nFurthermore, by overexpressing a mammalian homologue (M31) of Drosophila melanogaster heterochromatin protein 1 (HP1; refs 7,8) in transgenic mouse lines that exhibit PEV, it is possible to modify the proportion of cells that silence the transgene in a dose-dependent manner\nThus, we show M31 overexpression to have two contrasting effects which are dependent on chromosomal context: (i) it enhanced PEV in those lines with centromeric or pericentromeric transgene locations; and (ii) it suppressed PEV when the transgene was non-centromeric", "role": "user" }, { "content": "M31 is a heterochromatin component, that is concentrated in the XY body during spermatogenesis. M31 overexpression has two contrasting effects which are dependent on chromosomal context: (i) it enhanced PEV in those lines with centromeric or pericentromeric transgene locations; and (ii) it suppressed PEV when the transgene was non-centromeric.", "role": "assistant" } ]
science.scierc_re
science.scierc_re.274
[ { "content": "You will be given an abstract from a computer science research paper. Given the abstract, your task is to extract all unique relationships between entities of the following, and ONLY the following, types:\n\n- \"USED-FOR\": For instance, \"Our method models user proficiency\" -> [\"method\", \"user proficiency\", \"USED-FOR\"].\n- \"EVALUATE-FOR\": For instance, \"We evaluate our method using F1-measure\" -> [\"F1-measure\", \"method\", \"EVALUATE-FOR\"].\n- \"FEATURE-OF\": For instance, \"prior knowledge of the model\" -> [\"prior knowledge\", \"model\", \"FEATURE-OF\"].\n- \"HYPONYM-OF\": For instance, \"TUIT is a software library\" -> [\"TUIT\", \"software library\", \"HYPONYM-OF\"].\n- \"PART-OF\": For instance, \"We incorporate NLU module into the system\" -> [\"NLU module\", \"system\", \"PART-OF\"].\n- \"COMPARE\": For instance, \"Unlike the quantitative prior, the qualitative prior is often ignored\" -> [\"quantitative prior\", \"qualitative prior\", \"COMPARE\"].\n- \"CONJUNCTION\": For instance, \"obtained from human expert or knowledge base\" -> [\"human expert\", \"knowledge base\", \"CONJUNCTION\"].\n\nFormat your output as a JSON array. Each entry in the array should express a single relation, formatted as [\"<Entity_A>\", \"<Entity_B>\", \"<RELATION_A_B>\"]. An example output might look like:\n\n[[\"neural networks\", \"classification\", \"USED-FOR\"], [\"neuron\", \"neural network\", \"PART-OF\"]].\n\nIf no relations are found, please return an empty JSON array \"[]\".\n\nAbstract:\n\nThis paper investigates some computational problems associated with probabilistic translation models that have recently been adopted in the literature on machine translation. These models can be viewed as pairs of probabilistic context-free grammars working in a ` synchronous ' way. Two hardness results for the class NP are reported, along with an exponential time lower-bound for certain classes of algorithms that are currently used in the literature.", "role": "user" }, { "content": "[[\"probabilistic translation models\", \"machine translation\", \"USED-FOR\"], [\"computational problems\", \"probabilistic translation models\", \"FEATURE-OF\"], [\"probabilistic context-free grammars\", \"probabilistic translation models\", \"FEATURE-OF\"]]", "role": "assistant" } ]
science.mslr2022_cochrane_multidoc_summarization
science.mslr2022_cochrane_multidoc_summarization.2736
[ { "content": "You will be presented with a collection of medical randomized controlled trial (RCT) reports addressing a shared clinical question -- for instance, the efficacy of a medical intervention on a disease or condition. Given the titles and abstracts of these reports, your task is to generate the conclusion of the systematic review article synthesizing the key findings. Your conclusion should be a few sentences long. In your response, include only the conclusion and no other text. The RCT reports are below.\n\n\n\nTitle 1:\nA prospective randomized study comparing acupuncture with physiotherapy for low-back and pelvic pain in pregnancy.\nAbstract 1:\nThe aim of this study was to describe the effects of acupuncture in the treatment of low-back and pelvic pain during pregnancy and compare it with physiotherapy.\n Sixty pregnant women were allotted to acupuncture or physiotherapy. The women estimated the severity of their pain using a visual analog scale (VAS) from 0 to 10 and disability in performing twelve common daily activities using a disability-rating index (DRI) from 0 to 10.\n In the acupuncture group all 30 women completed the study (two exclusions), in the physiotherapy group only 18. Before treatment the two study groups were rather similar with respect to pain and disability. After treatment the mean morning VAS had declined from 3.4 to 0.9 (p<0.01) in the acupuncture group and from 3.7 to 2.3 (NS) in the physiotherapy group. The corresponding evening values had declined from 7.4 to 1.7 (p<0.01) and 6.6 to 4.5 (p<0.01), respectively. The mean VAS values were lower after acupuncture than after physiotherapy both in the morning (p=0.02) and in the evening (p<0.01). After treatment also the mean DRI values had decreased significantly in the acupuncture group for 11 of 12 activities and the values were significantly lower for all activities than in the physiotherapy group where no significant changes had taken place. Overall satisfaction was good in both groups. There were no serious adverse events in any of the patients.\n Acupuncture relieved pain and diminished disability in low-back pain during pregnancy better than physiotherapy.\n\n\nTitle 2:\nEffect of the \"sitting pelvic tilt exercise\" during the third trimester in primigravidas on back pain.\nAbstract 2:\nA single center, prospective, randomized, single blinded, controlled study comparing the effects and safety of \"sitting pelvic tilt exercise\" in relieving back pain during the third trimester in primigravida was carried out. The samples were composed of 67 primigravidas who attended the prenatal clinic at King Chulalongkorn Memorial Hospital. All subjects were selected by the random sampling technique and allocated into two groups for the experimental group and the control group; for 32 and 35 pregnants, respectively. The experimental group received the pelvic tilt exercise program for 8 weeks during the third trimester. Pain intensity was measured by visual analogue scale (VAS) at day 0 and day 56 in both groups. The result of the study revealed 1) The mean VAS of back pain in the experimental group was significantly lower at day 56 than at day 0 and lower than the control group at day 56 (p< 0.05) by unpaired t-test 2) There was no incidence of preterm labor, low birth weight or neonatal complication in the experimental group. In conclusion, the \"sitting pelvic tilt exercise\" during the third trimester in primigravidas could decrease back pain intensity without incidence of preterm labor, low birth weight or neonatal complication.\n\n\nTitle 3:\nThe effect of exercise on the intensity of low back pain in pregnant women.\nAbstract 3:\nTo investigate the effect of exercise during pregnancy on the intensity of low back pain and kinematics of spine.\n A prospective randomized study was designed. 107 women participated in an exercise program three times a week during second half of pregnancy for 12 weeks and 105 as control group. All filled a questionnaire between 17-22 weeks of gestation and 12 weeks later for assessment of their back pain intensity. Lordosis and flexibility of spine were measured by Flexible ruler and Side bending test, respectively, at the same times. Weight gain during pregnancy, Pregnancy length and neonatal weight were recorded.\n Low back pain intensity was increased in the control group. The exercise group showed significant reduction in the intensity of low back pain after exercise (p<0.0001). Flexibility of spine decreased more in the exercise group (p<0.0001). Weight gain during pregnancy, pregnancy length and neonatal weight were not different between the two groups.\n Exercise during second half of the pregnancy significantly reduced the intensity of low back pain, had no detectable effect on lordosis and had significant effect on flexibility of spine.\n\n\nTitle 4:\nTreatment of Pregnancy-Related Lumbar and Pelvic Girdle Pain by the Yoga Method: A Randomized Controlled Study.\nAbstract 4:\nAbstract Objective: Pregnancy-related lumbopelvic pain is a major problem for the majority of pregnant women. Complementary medicine has been used to alleviate pain, and yoga is one of the most commonly chosen alternative methods. The objective of this study was to assess the effectiveness of Hatha yoga in the reduction of lumbopelvic pain in pregnancy. Methods: A randomized controlled trial with 60 pregnant women (age range, 14-40 years) who reported lumbopelvic pain at 12 to 32 weeks of gestation was conducted from June 2009 to June 2011. Pregnant women who had twin pregnancies, had medical restrictions for exercise, used analgesics, and participated in physical therapy were excluded from the study. Pregnant women were divided into two groups: the yoga group, practicing exercises guided by this method, and the postural orientation group, performing standardized posture orientation according to instructions provided in a pamphlet. Treatment in each group lasted 10 weeks. A visual analog scale (VAS) was used to measure pain intensity. Lumbar pain and posterior pelvic pain provocation tests were used to confirm the presence of pain. Statistical analysis included the Mann-Whitney test, the McNemar test, a paired Wilcoxon test, and analysis of covariance. Results: The median pain score was lower in the yoga group (p<.0058) than the postural orientation group. Lumbar pain provocation tests showed a decreased response in relation to posterior pelvic pain provocation tests and a gradual reduction in pain intensity during 10 yoga sessions (p<.024). Conclusions: The yoga method was more effective at reducing lumbopelvic pain intensity compared with postural orientation.\n\n\nTitle 5:\nWater-gymnastics reduced the intensity of back/low back pain in pregnant women.\nAbstract 5:\nTo investigate if water-gymnastics during pregnancy may reduce the intensity of back/low back pain and the number of days on sick-leave.\n A prospective, randomized study. One hundred and twenty-nine women were randomized to participate in water-gymnastics once a week during the second half of pregnancy and 129 were randomized to a control group. The women in both groups filled in questionnaires in gestational weeks 18, 34 and within the first postpartum week. Every day from week 18 to labor they assessed the intensity of back/low back pain.\n Back pain intensity increased during pregnancy. No excess risk for the pregnancy associated with water-gymnastics was observed. The women participating in water-gymnastics recorded a lower intensity of back/low back pain. The total number of days on sick-leave because of back/low back pain was 982 in the water-gymnastics group (124 women) compared with 1484 in the control group (120 women). After weeks 32 33, seven women in the water-gymnastics group compared with 17 in the control group were on sickleave because of back/ low back pain (p=0.031).\n Intensity of back/low back pain increased with advancing pregnancy. There was no excess risk for urinary or vaginal infections associated with water-gymnastics. Water-gymnastics during the second half of pregnancy significantly reduced the intensity of back/ low back pain. Water-gymnastics decreased the number of women on sick-leave because of back/low back pain. Water-gymnastics during pregnancy can be recommended as a method to relieve back pain and may reduce the need for sick-leave.\n\n\nTitle 6:\nAcupuncture relieves pelvic and low-back pain in late pregnancy.\nAbstract 6:\nThe study was designed to evaluate the analgesic effect and possible adverse effects of acupuncture for pelvic and low-back pain during the last trimester of pregnancy.\n Following individual informed consent, 72 pregnant women reporting pelvic or low-back pain were randomized during pregnancy weeks 24-37 to an acupuncture group (n = 37) or to a control group (n = 35) at three maternity wards in southern Sweden. Traditional acupuncture points and local tender points (TP) were chosen according to individual pain patterns and stimulated once or twice a week until delivery or complete recovery in acupuncture patients. Control patients were given no sham stimulation. Throughout the study period each patient made weekly visual analog scale (VAS) evaluations of maximal and minimal pain intensity as well as three-point assessments of pain intensity during various activities.\n During the study period, VAS scorings of pain intensity decreased over time in 60% of patients in the acupuncture group and in 14% of those in the control group (p < 0.01). At the end of the study period, 43% of the acupuncture patients were less bothered than initially by pain during activity compared with 9% of control patients (p < 0.01). No serious adverse effects of acupuncture were found in the patients, and there were no adverse effects at all in the infants.\n Acupuncture relieves low-back and pelvic pain without serious adverse effects in late pregnancy.\n\n\nTitle 7:\nEffects of acupuncture and stabilising exercises as adjunct to standard treatment in pregnant women with pelvic girdle pain: randomised single blind controlled trial.\nAbstract 7:\nTo compare the efficacy of standard treatment, standard treatment plus acupuncture, and standard treatment plus stabilising exercises for pelvic girdle pain during pregnancy.\n Randomised single blind controlled trial. Settings East Hospital, Gothenburg, and 27 maternity care centres in Sweden.\n 386 pregnant women with pelvic girdle pain.\n Treatment for six weeks with standard treatment (n = 130), standard treatment plus acupuncture (n = 125), or standard treatment plus stabilising exercises (n = 131).\n Primary outcome measure was pain (visual analogue scale); secondary outcome measure was assessment of severity of pelvic girdle pain by an independent examiner before and after treatment.\n After treatment the stabilising exercise group had less pain than the standard group in the morning (median difference = 9, 95% confidence interval 1.7 to 12.8; P = 0.0312) and in the evening (13, 2.7 to 17.5; P = 0.0245). The acupuncture group, in turn, had less pain in the evening than the stabilising exercise group (-14, -18.1 to -3.3; P = 0.0130). Furthermore, the acupuncture group had less pain than the standard treatment group in the morning (12, 5.9 to 17.3; P < 0.001) and in the evening (27, 13.3 to 29.5; P < 0.001). Attenuation of pelvic girdle pain as assessed by the independent examiner was greatest in the acupuncture group.\n Acupuncture and stabilising exercises constitute efficient complements to standard treatment for the management of pelvic girdle pain during pregnancy. Acupuncture was superior to stabilising exercises in this study.\n\n\nTitle 8:\nEvaluation of a maternity cushion (Ozzlo pillow) for backache and insomnia in late pregnancy.\nAbstract 8:\nThe effect of a wedge-shaped pillow (Ozzlo pillow) was compared with a standard hospital pillow, used to support the abdomen of a pregnant woman while lying on her side, in preventing or alleviating backache and backache-related insomnia; 92 women at 36 weeks' gestation completed the study. Backache was found to be very common (87%), the onset of pain occurring before 29 weeks in 59%. Age, parity, previous backache and type of bed used did not correlate with the backache scores during the period of study. Lower scores for backache were recorded by women in the week they used the Ozzlo pillow compared with the week they used the standard pillow. Sleeping was deemed better by the patient with the Ozzlo pillow, though actual sleeping scores did not corroborate this. While significantly more felt the Ozzlo pillow was superior to a standard pillow for backache and sleeping, some found both methods helpful. The simple measure of supporting the abdomen with a pillow when in lateral recumbency is likely to benefit many women in late pregnancy. A wedge-shaped pillow of the Ozzlo type, conforming to the shape of the abdomen and supporting it more closely, may be of greater help than a standard cushion or pillow.", "role": "user" }, { "content": "All but one study had moderate to high potential for bias, so results must be viewed cautiously. Adding pregnancy-specific exercises, physiotherapy or acupuncture to usual prenatal care appears to relieve back or pelvic pain more than usual prenatal care alone, although the effects are small. We do not know if they actually prevent pain from starting in the first place. Water gymnastics appear to help women stay at work. Acupuncture shows better results compared to physiotherapy.", "role": "assistant" } ]
science.acl_arc_intent_classification
science.acl_arc_intent_classification.148
[ { "content": "You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories:\n- Background: provides context or foundational information related to the topic.\n- Extends: builds upon the cited work.\n- Uses: applies the methods or findings of the cited work.\n- Motivation: cites the work as inspiration or rationale for the research.\n- CompareOrContrast: compares or contrasts the cited work with others.\n- FutureWork: cites the work as a direction for future research.\n\nYour answer should be a single word from the following list of options: [\"Background\", \"Extends\", \"Uses\", \"Motivation\", \"CompareOrContrast\", \"FutureWork\"]. Do not include any other text in your response.\n\nSection Title:\nNone\n\nContext before the citation:\nNote that although our feature set was drawn primarily from our prior uncertainty detection experiments (Forbes-Riley and Litman, 2011a; Drummond and Litman, 2011), we have also experimented with other features, including state-of-theart acoustic-prosodic features used in the last Interspeech Challenges (Schuller et al., 2010; Schuller et al., 2009b) and made freely available in the openSMILE Toolkit (Florian et al., 2010). • Lexical and Dialogue Features dialogue name and turn number question name and question depth ITSPOKE-recognized lexical items in turn ITSPOKE-labeled turn (in)correctness incorrect runs • User Identifier Features: gender and pretest score \u0000\u0000\u0000\u0000 deviation running totals and averages for all features\n\nCitation Sentence:\nNote that although our feature set was drawn primarily from our prior uncertainty detection experiments ( Forbes-Riley and Litman , 2011a ; Drummond and Litman , 2011 ) , we have also experimented with other features , including state-of-theart acoustic-prosodic features used in the last Interspeech Challenges ( Schuller et al. , 2010 ; Schuller et al. , 2009b ) and made freely available in the openSMILE Toolkit ( Florian et al. , 2010 ) .\n\nContext after the citation:\nTo date, however, these features have only decreased the crossvalidation performance of our models.8 While some of our features are tutoring-specific, these have similar counterparts in other applications (i.e., answer (in)correctness corresponds to a more general notion of “response appropriateness” in other domains, while pretest score corresponds to the general notion of domain expertise). Moreover, all of our features are fully automatic and available in real-time, so that the model can be directly implemented and deployed. To that end, we now describe the results of our intrinsic and extrinsic evaluations of our DISE model, aimed at determining whether it is ready to be evaluated with real users. \u0000\u0000\u0000\u0000", "role": "user" }, { "content": "Extends", "role": "assistant" } ]
science.covid_deepset_qa
science.covid_deepset_qa.782
[ { "content": "You will be shown a paragraph from a biomedical research paper, followed by a question about the paper. Please extract text from the paragraph to answer the question. Do not include any text in your response other than the answer.\n\nParagraph: SARS to novel coronavirus – old lessons and new lessons\n\nhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC7026896/\n\nSHA: 5d254ed178c092d 3 6 9ce70ae965 59 acc471f9\n\nAuthors: McCloskey, Brian; Heymann, David L.\nDate: 2020-02-05\nDOI: 10.1017/s0950268820000254\nLicense: cc-by\n\nAbstract: The response to the novel coronavirus outbreak in China suggests that many of the lessons from the 200 SARS epidemic have been implemented and the response improved as a consequence. Nevertheless some questions remain and not all lessons have been successful. The national and international response demonstrates the complex link between public health, science and politics when an outbreak threatens to impact on global economies and reputations. The unprecedented measures implemented in China are a bold attempt to control the outbreak – we need to understand their effectiveness to balance costs and benefits for similar events in the future. Text: On 29 December 2019 clinicians in a hospital in Wuhan City, China noticed a clustering of cases of unusual pneumonia (with the first case identified at that time on 12 December) with an apparent link to a market that sells live fish, poultry and animals to the public. This event was reported to the World Health Organisation (WHO) on 1 December [1]. Within 4 weeks, by 26 January 2020, the causative organism had been identified as a novel coronavirus, the genome of the virus had been sequenced and published, reverse transcription polymerase chain reaction tests had been developed, the WHO R&D Blueprint had been activated to accelerate diagnostics, therapeutics and vaccine development and a candidate vaccine was ready for initial laboratory testing. Currently Chinese health authorities are building a 1000 bed hospital in Wuhan in 10 days. By 26 January also, almost 50 million people in Wuhan and neighbouring cities had effectively been placed in quarantine while the WHO had determined that the event should not yet be declared as a Public Health Emergency of International Concern (PHEIC) [2] and had recommended no specific travel restrictions. The WHO have emphasised the importance of exit screening at ports in countries showing transmission of the novel coronavirus and have provided guidance for countries implementing entry screening at airports while acknowledging that evidence for the effectiveness of entry screening is equivocal. This response is one of the swiftest, coordinated global responses to an emerging infectious disease the world has seen in modern times, but is it the appropriate response, will it be effective and is it sustainable? According to the situation report published by the WHO on 28 January 2020 [ ], a total of 2798 confirmed 2019-nCoV cases have been reported globally; of these, 2761 cases were from China, including Hong Kong (8 cases), Macau (5) and Taipei (4). Thirty-seven confirmed cases have been reported outside of China in eleven countries in Europe, North America, Australia and Asia; of these 7 exported cases, 6 had a travel history from China or an epidemiological link to a case from China. Of the confirmed cases in China, 461 have been reported as severely ill, with 80 deaths to date. This outbreak and the response to it illustrate some key issues about how global preparedness and response capacity for outbreaks have evolved over almost two decades since the severe acute respiratory syndrome (SARS) epidemic of 2002/ and what lessons have, or have not, been learned. It also raises questions about the impact these lessons have had on the way agencies and governments respond to these events and about the role of the WHO and the International Health Regulations (IHR). One of the critical lessons from the SARS experience was the absolute necessity to be able to coordinate the international resources that are available in an outbreak and to get them focussed on identifying priorities and solving problems. The WHO established the means to do this for SARS and it has since been further developed and integrated into global preparedness, especially after the West Africa Ebola epidemic. Organisations such as the Global Outbreak Alert and Response Network (GOARN), the Coalition for Epidemic Preparedness Innovations (CEPI), the Global Research Collaboration For Infectious Disease Preparedness (GloPID-R) and the Global Initiative on Sharing All Influenza Data (GISAID) have been supported by the WHO Research Blueprint and its Global Coordinating Mechanism to provide a forum where those with the expertise and capacity to contribute to managing new threats can come together both between and during outbreaks to develop innovative solutions to emerging problems. This global coordination has been active in the novel coronavirus outbreak. WHO's response system includes three virtual groups based on those developed for SARS to collate real time information to inform real time guidelines, and a first candidate vaccine is ready for laboratory testing within 4 weeks of the virus being identified. Another key factor in successfully preventing and managing emerging threats is the rapid and transparent sharing of information between countries and agencies. There was extensive criticism of China for its perceived failure to share information about the emerging SARS infection early enough in the outbreak to allow countries to prepare and respond. There were similar concerns about information sharing as Middle East Respiratory Syndrome (MERS) emerged and evolved in the Middle East in 2012, particularly in Saudi Arabia, and about the emergence of Ebola in West Africa in 2014. On this occasion information sharing seems to have been rapid and effective (while recognising that the information available in the early stages of an outbreak is always less than the global community would like). The WHO was notified of the original clustering within days and the full genomic sequence of the new virus was published less than 2 weeks after the cluster was first detected. The WHO has expressed its satisfaction with the actions of the Chinese authorities in sharing information with the WHO. Working with journalists and the media to help them understand the science and epidemiology, particularly in a fast moving event, will improve risk communication to the public and reduce inappropriate concerns and panic. While reporting of this outbreak shows signs of the efforts of epidemiologists, infectious disease experts, national and international public health agencies and others engaging with journalists, there are also signs that this is not yet achieving it's goal. For example, the public perception is that the increase in case numbers reported daily by the Chinese authorities represents a daily escalation in the epidemic while the reality is that these numbers are also the result of active, aggressive, case finding in China and some of these cases are 'old' cases newly recognised as being due to the novel coronavirus. Similarly the virus is usually described by the media as 'deadly' and although this is true in the sense that it has caused deaths, the nuances of uncertain case fatality rates in the early stages of an outbreak are not being communicated. The current estimated case fatality rate seems to be around % which is significant but not comparable to the 10% rate for SARS or 4% reported for MERS. These misperceptions are still driving public anxiety. To supplement formal reporting mechanisms between countries and with WHO (including the IHR), the use of informal mechanisms such as media and social media reports was advocated in the light of the SARS experience. There are now globally several systems that provide collated information from informal reporting including networks of experts and scanning of media and social media. These contribute to, and amplify, epidemic intelligence and are being integrated with national and international surveillance systems. The value, and the challenges, of this additional source of information has been evident in the current outbreak. The value comes from ensuring that early indications of cases beyond the initial outbreak city have been detected and can supplement the global risk assessment and monitoring of the evolution of the outbreak. The challenges lie in the volume and diversity of the information available and the relative lack of verification mechanisms, such that one of these systems (ProMed) has commented that it was becoming increasingly difficult to assimilate the information being supplied [4] and to make meaningful interpretations. Early in the outbreak it was reported that health workers had not been infected. This was reassuring because it is health workers who many times, and inadvertently, amplify transmission. Failure to wash hands between patients, for example, can result not only in autoinfection, but also in infection of patients hospitalised for other causes when they provide care. Autoinfection is not only a risk for the health worker, but also for their families and the communities in which they live, depending on the transmissibility and means of transmission. More recently infection, and at least one death, in health workers has been confirmed. Although not unexpected this does add to the epidemiological risk. A characteristic of the SARS outbreak was the variability of transmissibility between cases and the occurrence of 'superspreading events' where a case infected significantly more contacts than the average. This was also seen with MERS in the outbreak in the Republic of Korea (RoK). In this current novel coronavirus outbreak, such superspreading events have not been documented but the epidemiology is still not clear. Confirming whether or not this is happening must be an urgent task for the Chinese investigation. Modellers have suggested reproductive rates (R 0 ) of .8 (95% confidence interval, .6-4.0) [5] and 2.6 (1.5- .5) [6] ; R 0 for SARS was estimated at around in the absence of control measures [7] . The economic impact of major outbreaks can be substantial for the affected country. This was seen clearly in SARS, MERS in RoK and Ebola in West Africa. One analyst estimates that the current coronavirus outbreak's likely impact will range from a 0.8% cut to real GDP if the epidemic is controlled within months, to a 1.9% cost to GDP if the epidemic lasts 9 months [8] . This may increase substantially in the light of the extended restrictions on movement, and therefore trade and commerce, within China. The emergence of a significant respiratory illness linked to a novel coronavirus represents a test of the global capacity to detect and mange emerging disease threats. Its emergence in China adds an additional dimension in the light of previous experience with SARS. The timing of the outbreak immediately before the Chinese Lunar New Year with its attendant population movements adds extra risk and urgency to the response. The rapid sharing of information in this outbreak and the speed of the coordinated response both in the country and internationally suggest that lessons have been learned from SARS that improve global capacity. The international networks and forums that now exist have facilitated the bringing together of expertise from around the world to focus research and development efforts and maximise the impact. At this early stage in the outbreak information remains incomplete and key clinical and epidemiological questions have not yet been answered, but the deficit seems to be due more to the constraints of investigating an emerging disease than to any unwillingness to engage and share information with partners. There are some indications of areas where further improvement is necessary. The global media response to the unfolding events has been relatively balanced and informed but the nuances of the evolving situation have not been critically examined in partnership with the media and as a result the public perception of the risk may be exaggeratedalthough it of course remains possible that the outbreak will develop in a way that matches up to the perceived risk. The lack of appreciation of the uncertainties in determining a meaningful case fatality rate and the significance of ascertainment bias at the beginning of an outbreak, along with the impact of aggressive case finding on case numbers, are examples of where understanding could be improved. This is always a challenging process when balancing the resources focussed on analysing the situation on the ground with resources directed at interpreting the information for journalists but in SARS, the R 0 was seen to decrease in response to information reaching the public and the public then adopting risk reduction actions [6] ; so accurate public risk communication is critical to success. It would be helpful to find a forum where this can be explored with the media community after the event. The increase in access to early information from diverse sources including media and social media adds an important dimension to identifying and tracking new events globally and is a key part of the overall epidemic intelligence system. However, it is also a potential source of disinformation. When, as has been seen in this outbreak, the volume of information coming in exceeds any capacity to collate and analyse it and to attempt to cross-reference and verify separate items, there is a risk that the information fuels speculation and media and public concern. Again there is a fine balance between information that encourages appropriate risk avoidance actions and information that encourages inappropriate actions; however the public health is usually better served by more information rather than less. The role of a declaration of a PHEIC in managing a serious outbreak has been questioned in the light of Ebola in West Africa and in the Democratic Republic of Congo [9] and has been challenged again with this outbreak. The binary nature of a PHEIC declaration (either an event is a PHEIC or it isn'tthere are no intermediate options) and the specificity of the three defined criteria for a PHEIC have caused difficulty for Emergency Committees in considering whether a given event should be a PHEIC. The lack of a clear understanding of what a PHEIC declaration is meant to achieve adds to the Emergency Committee's difficulties, as does the relative paucity of clinical and epidemiological answers at this stage of the investigation. In this instance the Emergency Committee were divided in coming to a conclusion but decided on balance that the current situation, although an emergency, should not as yet be declared a PHEIC [2]. As with Ebola in the DRC, there has been criticism of the WHO for this decision but, as with Ebola, it is not immediately clear what would be different in the response if a PHEIC was declared. The WHO is working on improving the way in which Emergency Committees develop their advice for the Director General but, as recommended by this Emergency Committee and the post-Ebola IHR Review Committee in 2015, the development of an intermediate alert alongside WHO's risk assessment process may be helpful. A key function of a PHEIC declaration is that it is the (only) gateway to the WHO Temporary Recommendations on possible travel and trade restrictions to limit international spread of a disease. In this case several countries globally had already implemented entry screening at airports and China had begun closing down international travel from Wuhan before the Emergency Committee had finished their deliberations. While the WHO would not, and could not, interfere with the sovereign decisions of member states, the lack of influence on travel and trade decisions could prove problematic. Alongside the speed of the response in this outbreak, we have seen dramatic changes in the scale of the response. The imposition of very extensive quarantine measures on millions of people as an attempt to break the transmission of the virus is unprecedented. We do not know whether they will be effective; indeed we do not know how we will determine if they have been effectivewhat end point can we measure that will provide an answer to that question? If recent suggestions that people infected with this coronavirus may be infectious while incubating or asymptomatic, and the reports that up to 5 m people left Wuhan before the travel restrictions were imposed, are confirmed, the efficacy of these control measures will be more challenged. Given the likely impact on at least the Chinese economy and probably the global economy, it will be important to understand the role and the effectiveness of public health measures on this scale for the future. However, the imposition of these dramatic measures does also raise a wider question: if there is an impact from these measures, what other countries would (or could) implement such measures? Would other countries accept the self-imposed economic damage that China has accepted to try and contain this outbreak? Is it reasonable to consider that national governments would close down public transport into and out of London, New York or Paris in the week before Christmas even if it were shown to be an effective control measure? These decisions and questions cross the interface between public health, science and politics.\n\nQuestion: What was the R0 of SARS in absence of control measures?", "role": "user" }, { "content": "3", "role": "assistant" } ]
science.scientific_lay_summarisation_plos_single_doc_summ
science.scientific_lay_summarisation_plos_single_doc_summ.17203
[ { "content": "You will be presented with the abstract, introduction, and discussion section from a biomedical research article. Your task is to create a summary that a layperson can understand, capturing the essence of the research article. Keep essential scientific terms, but ensure that the language remains clear and the concepts are explained in an uncomplicated manner.\n\nTitle: Spatial Generalization in Operant Learning: Lessons from Professional Basketball\nArticle:\nAbstract:\nIn operant learning, behaviors are reinforced or inhibited in response to the consequences of similar actions taken in the past. However, because in natural environments the “same” situation never recurs, it is essential for the learner to decide what “similar” is so that he can generalize from experience in one state of the world to future actions in different states of the world. The computational principles underlying this generalization are poorly understood, in particular because natural environments are typically too complex to study quantitatively. In this paper we study the principles underlying generalization in operant learning of professional basketball players. In particular, we utilize detailed information about the spatial organization of shot locations to study how players adapt their attacking strategy in real time according to recent events in the game. To quantify this learning, we study how a make \\ miss from one location in the court affects the probabilities of shooting from different locations. We show that generalization is not a spatially-local process, nor is governed by the difficulty of the shot. Rather, to a first approximation, players use a simplified binary representation of the court into 2 pt and 3 pt zones. This result indicates that rather than using low-level features, generalization is determined by high-level cognitive processes that incorporate the abstract rules of the game.\nIntroduction:\nOf several responses made to the same situation, those which are accompanied or closely followed by satisfaction to the animal will… be more firmly connected with the situation, so that, when it recurs, they will be more likely to recur…’( Edward Thorndike, 1874–1949)[1]. No man ever steps into the same river twice, for it's not the same river and he's not the same man( Heraclitus of Ephesus, 535–475 BCE)[2]. Humans and animals modify their behavior in response to the consequences of their previous actions, a process known as operant learning. The standard account for this learning is based on a family of reinforcement learning( RL) algorithms that assert that the computational problem of learning from experience is achieved through the synergy of two processes: first, the values of the different actions( or more generally, state-actions) are learned from past actions and their subsequent rewards; second, these learned values are used to choose( or to learn to choose) among different actions such that actions associated with a higher values are more likely to be chosen[3]–[5] but see also[6]–[8]. This account is based, to a large extent, on a large number of laboratory experiments, in which participants repeatedly choose between the same small number of alternative actions( e. g., press a button) in repeated settings and are rewarded according to these actions. By contrast, in many natural environments, organisms learn from the consequences of their past actions in settings in which the same situation and action never recur( not even in the sense that two “identical” trials “recur” in a laboratory experiment). In these cases, generalization is an essential part of operant learning[9]. In this process of generalization, the organism determines which past situations, actions and their consequences are relevant for the current situation. In the language of RL algorithms discussed above, generalization is the process of determining which set of different situations defines a state and which set of responses defines an action. The level of generalization determines, roughly speaking, the density parsing of the set of situations into states and the set of responses into actions. A limited generalization would result in a large number of state and actions in the process of learning whereas broad generalization would result in a small number of states and actions. Too limited generalization implies that the organism learns values of states that are essentially identical, resulting in too-slow learning. Too broad generalization implies that the organism is inferring the outcome of future responses from irrelevant past experience, which may lead to suboptimal behavior even after very long learning. Thus, the proper level of generalization, which determines the tradeoff between the speed and the accuracy of learning, is of an utmost importance in the process of learning. It should be noted that the question of the proper level of generalization is present even in RL models that assume continuous states and actions[10],[11]. The problem of determining the proper level of generalization is not limited to operant learning. Indeed, this question has received considerable attention in the framework of Pavlovian learning and supervised learning( see[12] and references within). The goal of this paper is to elucidate the cognitive strategy underlying generalization in operant learning in natural conditions. Professional basketball, which is played by highly motivated and extensively-trained players, provides an exceptional opportunity to quantitatively study generalization in operant learning in complex natural environments. The objective of players in basketball is to gain points by shooting a ball through a hoop. If successful, the team is awarded with two or three points, depending on the distance of the shot attempt from the basket. In a previous study we demonstrated that players modify their shot selection policy in response to the recent history of their shots and their outcomes[13]. After a made( successful) 3-point( 3 pt) shot, the probability of attempting another 3 pt shot is 30% higher than that probability after a missed 3 pt. Moreover, some of the variability in players' shot selection can be accounted for using standard RL algorithms. However, lacking additional information about the shots, our previous study was unable to address the question of what is considered by the players as “the same situation” and “the same action”. Consider a player in possession of the ball. Multiple factors, including the locations, velocities and postures of team and opponent players, the score and the time in the game are relevant to the decision of whether or not to attempt a field goal( FG). In the framework of RL, all these factors determine the state of the world. In this paper we focus on the spatial location of the payers, which provides us with a low dimensional projection of the state of the world at the time of the FG. Quantifying how the outcome of a FG in one spatial location affects subsequent FGs in different locations is thus informative about the pattern and level of generalization between states. A spatially restricted generalization implies that the outcome of shots made in a particular location would have very little effect of behavior in other locations of the court. By contrast, learning could be independent of shot location, implying substantial spatial generalization. Between these two extremes, a made shot in one location may enhance the probability of another shot from the vicinity of that location, but not from further away locations. Alternatively, the pattern of generalization may be more complex. For example, a made shot in one location may enhance the probability of another shot from the same distance, the same angle or from the symmetrical location relative to the basket. Identifying the patterns of spatial generalization is thus the objective of this study.\nDiscussion:\nIn this study we used professional basketball to probe the cognitive strategy underlying generalization in operant learning in natural conditions. Generalization in learning indicates that events taking place in one state of the world affect the policy utilized in other states of the world. We used the spatial locations of players at the times of shot attempts as a proxy for the state of the world, and quantified the level of generalization between the states. Using hierarchical clustering analysis of the spatial conditional probabilities of shots we found that the pattern of generalization indicates that the separation of FGs into 2 pt and 3 pt shots, high-level abstract features of the game that seem of little relevance to the learning task, dominates the pattern of generalization. This result is also supported by the finding that while the matching law is a good approximation to the aggregate allocation of FGs into 2 pt and 3 pt shots, there are substantial deviations from this law of behavior on a finer spatial resolution. We interpret these findings as resulting from learning processes that are taking place in the brains of the shooting players. However, because these players operate in a complex strategic setting, other contributors to behavior should be considered. For example, the outcome of the shot may affect defensive maneuvers, which could change the spatial distribution of subsequent shots. However, this mechanism is unlikely to be a main contributor because defense is likely to respond to a made shot by preventing another shot from the same location, whereas we found that the diagonal elements of tend to be positive. Another potential contributor to the changes observed in behavior may be changes in the behavior of the player's coach or teammates. In a previous study we discussed this possibility at length. In short, if changes in a player's behavior result from actions taken by the player's coach or teammates then the learning observed in players within the same team are expected to be correlated. However, such correlations were not found[13]. Thus we conclude that while it is likely that many additional factors contribute to behavior, it is learning processes within the shooting players that dominate the generalization patterns reported in this paper. Models of operant learning often take as given that the learner has full information about the relevant states and actions in the problem. However, in real-life situations( and probably also in laboratory settings), the necessary preliminary step of identifying these states and actions is an essential part of operant learning. This is also often the case in machine learning, where classification is often preceded by feature extraction[26]. Our results highlight the importance of this preprocessing stage in the learning[9],[27]. Another contribution of this work is methodological. The distinction between the prospective and retrospective similarity may prove beneficial when studying generalization in other learning tasks. For example, in a standard supervised learning paradigm, the participant is instructed to learn the mapping between stimuli and their desired responses. The learning is quantified by computing the set of probabilities that denote the probability that the response of the participant to stimulus, is[28]. These probabilities can be used to define a matrix whose entries are. The prospective and retrospective clustering analysis that we used to study generalization in operant learning can be readily applied to this task to better quantify the pattern of generalization. Recent years have seen a growing interest in data acquired from professional sports. Basketball data, in particular, were used to study various phenomena, such as decision making in shot selection[13],[29]–[33], the ‘hot hand’ belief[34],[35], how coaching experience affects the effectiveness of timeouts[36], and the dynamics of scoring within a game[37]. Off the court, basketball players were subjects in imaging experiments examining the underlying neural mechanisms of action anticipation and evaluation[38],[39]. A major advantage of professional basketball is that large quantities of carefully collected behavioral data can be used to study the behavior of highly-motivated and extensively-trained humans in their natural settings, complementing the more controlled experiments in laboratory settings. With the development of high-speed cameras and automatic image processing, the extensive public interest in sports' statistics can be utilized to enhance our understanding of the computational principles underlying different cognitive processes.", "role": "user" }, { "content": "According to the law of effect , formulated a century ago by Edward Thorndike , actions which are rewarded in a particular situation are more likely to be executed when that same situation recurs . However , in natural settings the same situation never recurs and therefore , generalization from one state of the world to other states is an essential part of the process of learning . In this paper we utilize basketball statistics to study the computational principles underlying generalization in operant learning of professional basketball players . We show that players are more likely to attempt a field goal from the vicinity of a previously made shot than they are from the vicinity of a missed shot , as expected from the law of effect . However , the outcome of a shot can also affect the likelihood of attempting another shot at a different location . Using hierarchical clustering we characterize the spatial pattern of generalization and show that generalization is primarily determined by the type of shot , 3 pt vs . 2 pt . This result indicates that rather than using low-level features , generalization is determined by high-level cognitive processes that incorporate the abstract rules of the game .", "role": "assistant" } ]
science.multixscience_multidoc_summarization
science.multixscience_multidoc_summarization.17337
[ { "content": "You are provided with the abstract of a scientific paper and the abstracts of the papers it references. Your task is to write a Related Work section for the main paper. This section should concisely summarize and integrate the key points from the cited abstracts, reflecting how they relate to the main paper's contributions and context. In your summary, use numerical citation markers like [1], [2], etc., to refer to the corresponding papers.\n\nMain Abstract:\nThe search of spanning trees with interesting disjunction properties has led to the introduction of edge-disjoint spanning trees, independent spanning trees and more recently completely independent spanning trees. We group together these notions by defining (i, j)-disjoint spanning trees, where i (j, respectively) is the number of vertices (edges, respectively) that are shared by more than one tree. We illustrate how (i, j)-disjoint spanning trees provide some nuances between the existence of disjoint connected dominating sets and completely independent spanning trees. We prove that determining if there exist two (i, j)-disjoint spanning trees in a graph G is NP-complete, for every two positive integers i and j. Moreover we prove that for square of graphs, k-connected interval graphs, complete graphs and several grids, there exist (i, j)-disjoint spanning trees for interesting values of i and j.\n\nCited Abstract(s):\n[1]: In a data processing system comprising a plurality of systems each including a plurality of console type typewriters for establishing communications pertaining to data processing operations between an operator and the system, the communication information which is entered into and derived from a central processing unit in each system by the console type typewriter is stored in a character buffer unit corresponding to the console type typewriter in a monitor transfer control unit, and a buffer scanning unit scans the character buffers to transfer the contents therein into a statistical analyzer or processing unit so that communication information may be automatically analyzed and summarized to obtain the data per day or month required for determining whether or not the data processing system has been effectively operated.\n\n[2]: \n\n[3]: The dominating set problem in graphs asks for a minimum size subset of vertices with the following property: each vertex is required to be either in the dominating set, or adjacent to some vertex in the dominating set. We focus on the related question of finding a connected dominating set of minimum size, where the graph induced by vertices in the dominating set is required to be connected as well. This problem arises in network testing, as well as in wireless communication. Two polynomial time algorithms that achieve approximation factors of 2H(Δ)+2 and H(Δ)+2 are presented, where Δ is the maximum degree and H is the harmonic function. This question also arises in relation to the traveling tourist problem, where one is looking for the shortest tour such that each vertex is either visited or has at least one of its neighbors visited. We also consider a generalization of the problem to the weighted case, and give an algorithm with an approximation factor of (cn+1) n where cn ln k is the approximation factor for the node weighted Steiner tree problem (currently cn = 1.6103 ). We also consider the more general problem of finding a connected dominating set of a specified subset of vertices and provide a polynomial time algorithm with a (c+1) H(Δ) +c-1 approximation factor, where c is the Steiner approximation ratio for graphs (currently c = 1.644 ).\n\n[4]: A dominating set in a graph G is a connected dominating set of G if it induces a connected subgraph of G. The connected domatic number of G is the maximum number of pairwise disjoint, connected dominating sets in V(G). We establish a sharp lower bound on the number of edges in a connected graph with a given order and given connected domatic number. We also show that a planar graph has connected domatic number at most 4 and give a characterization of planar graphs having connected domatic number 3.", "role": "user" }, { "content": "Some subsets of vertices @math of a graph @math are @math if @math are pairwise disjoint and each subset is a connected dominating set in @math . There are some works about disjoint connected dominating sets that can be transcribed in terms of internally vertex-disjoint spanning trees (the disjoint connected dominating sets can be used to provide the inner vertices of internally vertex-disjoint spanning trees). The maximum number of disjoint connected dominating sets in a graph @math is the . This parameter is denoted by @math and has been introduced by Hedetniemi and Laskar [1] in 1984. An interesting result about connected domatic number concerns planar graphs, for which Hartnell and Rall have proven that, except @math (which has connected domatic number @math ), their connected domatic number is bounded by 3 [4] . The problem of constructing a connected dominating set is often motivated by wireless ad-hoc networks [3] [2] for which connected dominating sets are used to create a virtual backbone in the network.", "role": "assistant" } ]
science.scicite_classification
science.scicite_classification.1906
[ { "content": "You will be presented with a citation sentence from a research paper. Classify the intent behind this citation by choosing from one of the following categories:\n\n- Method: Cites the paper for its methodology or procedure.\n- Background: Cites the paper to provide background information.\n- Result: Cites the paper for its findings or results.\n\nYour answer should be a single word from the following list of options: [\"Method\", \"Background\", \"Result\"]. Do not include any other text in your response.\n\nCitation sentence:\n\nNevertheless, one area of ethics for HTA that indeed is considerably underdeveloped in comparison to other areas of HTA is quality assessment of ethics analyses in HTA [10], [11].", "role": "user" }, { "content": "Background", "role": "assistant" } ]
science.chemprot_re
science.chemprot_re.333
[ { "content": "Given an abstract from a research paper, your task is to identify and label relationships between chemical compounds and genes mentioned in the abstract.\n\nThe relationship type must be one of ['DOWNREGULATOR', 'SUBSTRATE', 'INDIRECT-REGULATOR', 'PART-OF', 'MODULATOR-ACTIVATOR', 'REGULATOR', 'INHIBITOR', 'COFACTOR', 'UPREGULATOR', 'ACTIVATOR', 'ANTAGONIST', 'NOT', 'INDIRECT-DOWNREGULATOR', 'SUBSTRATE_PRODUCT-OF', 'INDIRECT-UPREGULATOR', 'AGONIST', 'PRODUCT-OF', 'MODULATOR', 'DIRECT-REGULATOR', 'UNDEFINED', 'AGONIST-INHIBITOR', 'AGONIST-ACTIVATOR', 'MODULATOR-INHIBITOR'].\n\nPlease format your output as a JSON array. Each entry in the array should express a single relation, formatted as [\"<Entity_A>\", \"<RELATION_A_B>\", \"<Entity_B>\"]. If no relations can be found, please output an empty JSON array [].\n\nAbstract:\n\nRecent developments in the synthesis of acetylcholinesterase inhibitors. The acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) inhibitory activities of a series of pyrano[2,3-b]quinolines (2, 3), [1,8]naphthyridines (5, 6), 4-amino-2,3-diaryl-5,6,7,8-tetrahydrofuro[2,3-b]quinolines (11-13) / 4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]furo[2,3-b]pyridine (14), 4-amino-5,6,7,8-tetrahydro-2,3-diphenylthieno[2,3-b]quinoline (15) / 4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]thieno[2,3-b]pyridine (16) are described. These compounds are tacrine analogues that have been prepared from readily available polyfunctionalized ethyl [6-amino-5-cyano-4H-pyran]-3-carboxylates (9, 10), ethyl [6-amino-5-cyanopyridine]-3-carboxylates (7, 8), 2-amino-3-cyano-4,5-diarylfurans (17-19) and 2-amino-3-cyano-4,5-diphenylthiophene (20) via Friedlander condensation with selected ketones. These compounds are competitive and, in a few cases, non-competitive inhibitors for AChE, the most potent being compound (14), though three-fold less active than tacrine. The BuChE inhibitory activity is only significant in compounds 11 and 14, ten-fold less active than tacrine. Furthermore, the products 12 and 13 are selective and moderate AChE inhibitors.", "role": "user" }, { "content": "[[\"tacrine\", \"INHIBITOR\", \"BuChE\"], [\"[1,8]naphthyridines\", \"INHIBITOR\", \"butyrylcholinesterase\"], [\"[1,8]naphthyridines\", \"INHIBITOR\", \"AChE\"], [\"[1,8]naphthyridines\", \"INHIBITOR\", \"BuChE\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]thieno[2,3-b]pyridine\", \"INHIBITOR\", \"AChE\"], [\"pyrano[2,3-b]quinolines\", \"INHIBITOR\", \"BuChE\"], [\"4-amino-2,3-diaryl-5,6,7,8-tetrahydrofuro[2,3-b]quinolines\", \"INHIBITOR\", \"butyrylcholinesterase\"], [\"4-amino-2,3-diaryl-5,6,7,8-tetrahydrofuro[2,3-b]quinolines\", \"INHIBITOR\", \"acetylcholinesterase\"], [\"4-amino-5,6,7,8-tetrahydro-2,3-diphenylthieno[2,3-b]quinoline\", \"INHIBITOR\", \"acetylcholinesterase\"], [\"pyrano[2,3-b]quinolines\", \"INHIBITOR\", \"butyrylcholinesterase\"], [\"4-amino-5,6,7,8-tetrahydro-2,3-diphenylthieno[2,3-b]quinoline\", \"INHIBITOR\", \"AChE\"], [\"4-amino-5,6,7,8-tetrahydro-2,3-diphenylthieno[2,3-b]quinoline\", \"INHIBITOR\", \"BuChE\"], [\"4-amino-2,3-diaryl-5,6,7,8-tetrahydrofuro[2,3-b]quinolines\", \"INHIBITOR\", \"AChE\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]furo[2,3-b]pyridine\", \"INHIBITOR\", \"acetylcholinesterase\"], [\"[1,8]naphthyridines\", \"INHIBITOR\", \"acetylcholinesterase\"], [\"tacrine\", \"INHIBITOR\", \"AChE\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]furo[2,3-b]pyridine\", \"INHIBITOR\", \"butyrylcholinesterase\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]furo[2,3-b]pyridine\", \"INHIBITOR\", \"BuChE\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]thieno[2,3-b]pyridine\", \"INHIBITOR\", \"butyrylcholinesterase\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]furo[2,3-b]pyridine\", \"INHIBITOR\", \"AChE\"], [\"4-amino-5,6,7,8-tetrahydro-2,3-diphenylthieno[2,3-b]quinoline\", \"INHIBITOR\", \"butyrylcholinesterase\"], [\"4-amino-2,3-diaryl-5,6,7,8-tetrahydrofuro[2,3-b]quinolines\", \"INHIBITOR\", \"BuChE\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]thieno[2,3-b]pyridine\", \"INHIBITOR\", \"acetylcholinesterase\"], [\"4-amino-6,7,8,9-tetrahydro-2,3-diphenyl-5H-cyclohepta[e]thieno[2,3-b]pyridine\", \"INHIBITOR\", \"BuChE\"], [\"pyrano[2,3-b]quinolines\", \"INHIBITOR\", \"acetylcholinesterase\"], [\"pyrano[2,3-b]quinolines\", \"INHIBITOR\", \"AChE\"]]", "role": "assistant" } ]
science.covid_deepset_qa
science.covid_deepset_qa.1939
[ { "content": "You will be shown a paragraph from a biomedical research paper, followed by a question about the paper. Please extract text from the paragraph to answer the question. Do not include any text in your response other than the answer.\n\nParagraph: Prior to December 2019, 6 CoVs were known to infect human, including 2 αCoV (HCoV-229E and HKU-NL63) and 4 βCoV (HCoV-OC43 [ \n\nHCoV-OC43 and HCoV-HKU1 usually cause self-limiting upper respiratory infections in immunocompetent hosts and occasionally lower respiratory tract infections in immunocompromised hosts and elderly [4] . In contrast, SARS-CoV (lineage B βCoV) and MERS-CoV (lineage C βCoV) may cause severe lower respiratory tract infection with acute respiratory distress syndrome and extrapulmonary manifestations, such as diarrhea, lymphopenia, deranged liver and renal function tests, and multiorgan dysfunction syndrome, among both immunocompetent and immunocompromised hosts with mortality rates of ∼10% and ∼35%, respectively [5, 6] . On 31 December 2019, the World Health Organization (WHO) was informed of cases of pneumonia of unknown cause in Wuhan City, Hubei Province, China [7] . Subsequent virological testing showed that a novel CoV was detected in these patients. As of 16 January 2020, 43 patients have been diagnosed to have infection with this novel CoV, including two exported cases of mild pneumonia in Thailand and Japan [8, 9] . The earliest date of symptom onset was 1 December 2019 [10] . The symptomatology of these patients included fever, malaise, dry cough, and dyspnea. Among 41 patients admitted to a designated hospital in Wuhan, 13 (32%) required intensive care and 6 (15%) died. All 41 patients had pneumonia with abnormal findings on chest computerized tomography scans [10] . We recently reported a familial cluster of 2019-nCoV infection in a Shenzhen family with travel history to Wuhan [11] . In the present study, we analyzed a 2019-nCoV complete genome from a patient in this familial cluster and compared it with the genomes of related βCoVs to provide insights into the potential source and control strategies. The complete genome sequence of 2019-nCoV HKU-SZ-005b was available at GenBank (accession no. MN975262) ( Table 1 ). The representative complete genomes of other related βCoVs strains collected from human or mammals were included for comparative analysis. These included strains collected from human, bats, and Himalayan palm civet between 2003 and 2018, with one 229E coronavirus strain as the outgroup. Phylogenetic tree construction by the neighbour joining method was performed using MEGA X software, with bootstrap values being calculated from 1000 trees [12] . The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) was shown next to the branches [13] . The tree was drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Poisson correction method and were in the units of the number of amino acid substitutions per site [14] . All ambiguous positions were removed for each sequence pair (pairwise deletion option). Evolutionary analyses were conducted in MEGA X [15] . Multiple alignment was performed using CLUSTAL 2.1 and further visualized using BOX-SHADE 3.21. Structural analysis of orf8 was performed using PSI-blast-based secondary structure PREDiction (PSIPRED) [16] . For the prediction of protein secondary structure including beta sheet, alpha helix, and coil, initial amino acid sequences were input and analysed using neural networking and its own algorithm. Predicted structures were visualized and highlighted on the BOX-SHADE alignment. Prediction of transmembrane domains was performed using the TMHMM 2.0 server (http://www.cbs.dtu.dk/services/TMHMM/). Secondary structure prediction in the 5 ′ -untranslated region (UTR) and 3 ′ -UTR was performed using the RNAfold WebServer (http://rna.tbi.univie.ac.at/cgi-bin/ RNAWebSuite/RNAfold.cgi) with minimum free energy (MFE) and partition function in Fold algorithms and Table 2 . Putative functions and proteolytic cleavage sites of 16 nonstructural proteins in orf1a/b as predicted by bioinformatics. Putative function/domain Amino acid position Putative cleave site\n\ncomplex with nsp3 and 6: DMV formation\n\ncomplex with nsp3 and 4: DMV formation\n\nshort peptide at the end of orf1a basic options. The human SARS-CoV 5 ′ -and 3 ′ -UTR were used as references to adjust the prediction results. The single-stranded RNA genome of the 2019-nCoV was 29891 nucleotides in size, encoding 9860 amino acids. The G + C content was 38%. Similar to other (Table 2 ). There are no remarkable differences between the orfs and nsps of 2019-nCoV with those of SARS-CoV (Table 3) . The major distinction between SARSr-CoV and SARS-CoV is in orf3b, Spike and orf8 but especially variable in Spike S1 and orf8 which were previously shown to be recombination hot spots. Spike glycoprotein comprised of S1 and S2 subunits. The S1 subunit contains a signal peptide, followed by an N-terminal domain (NTD) and receptor-binding domain (RBD), while the S2 subunit contains conserved fusion peptide (FP), heptad repeat (HR) 1 and 2, transmembrane domain (TM), and cytoplasmic domain (CP). We found that the S2 subunit of 2019-nCoV is highly conserved and shares 99% identity with those of the two bat SARS-like CoVs (SL-CoV ZXC21 and ZC45) and human SARS-CoV (Figure 2 ). Thus the broad spectrum antiviral peptides against S2 would be an important preventive and treatment modality for testing in animal models before clinical trials [18] . Though the S1 subunit of 2019-nCoV shares around 70% identity to that of the two bat SARS-like CoVs and human SARS-CoV (Figure 3(A) ), the core domain of RBD (excluding the external subdomain) are highly conserved (Figure 3(B) ). Most of the amino acid differences of RBD are located in the external subdomain, which is responsible for the direct interaction with the host receptor. Further investigation of this soluble variable external subdomain region will reveal its receptor usage, interspecies transmission and pathogenesis. Unlike 2019-nCoV and human SARS-CoV, most known bat SARSr-CoVs have two stretches of deletions in the spike receptor binding domain (RBD) when compared with that of human SARS-CoV. But some Yunnan strains such as the WIV1 had no such deletions and can use human ACE2 as a cellular entry receptor. It is interesting to note that the two bat SARS-related coronavirus ZXC21 and ZC45, being closest to 2019-nCoV, can infect suckling rats and cause inflammation in the brain tissue, and pathological changes in lung & intestine. However, these two viruses could not be isolated in Vero E6 cells and were not investigated further. The two retained deletion sites in the Spike genes of ZXC21 and ZC45 may lessen their likelihood of jumping species barriers imposed by receptor specificity. A novel short putative protein with 4 helices and no homology to existing SARS-CoV or SARS-r-CoV protein was found within Orf3b ( Figure 4 ). It is notable that SARS-CoV deletion mutants lacking orf3b replicate to levels similar to those of wildtype virus in several cell types [19] , suggesting that orf3b is dispensable for viral replication in vitro. But orf3b may have a role in viral pathogenicity as Vero E6 but not 293T cells transfected with a construct expressing Orf3b underwent necrosis as early as 6 h after transfection and underwent simultaneous necrosis and apoptosis at later time points [20] . Orf3b was also shown to inhibit expression of IFN-β at synthesis and signalling [21] . Subsequently, orf3b homologues identified from three bat SARSrelated-CoV strains were C-terminally truncated and lacked the C-terminal nucleus localization signal of SARS-CoV [22] . IFN antagonist activity analysis demonstrated that one SARS-related-CoV orf3b still possessed IFN antagonist and IRF3-modulating activities. These results indicated that different orf3b proteins display different IFN antagonist activities and this function is independent of the protein's nuclear localization, suggesting a potential link between bat SARS-related-CoV orf3b function and pathogenesis. The importance of this new protein in 2019-nCoV will require further validation and study. Orf8 orf8 is an accessory protein found in the Betacoronavirus lineage B coronaviruses. Human SARS-CoVs isolated from early-phase patients, all civet SARS-CoVs, and other bat SARS-related CoVs contain fulllength orf8 [23] . However, a 29-nucleotide deletion,\n\nBat SL-CoV ZXC21 2018\n\nBat which causes the split of full length of orf8 into putative orf8a and orf8b, has been found in all SARS-CoV isolated from mid-and late-phase human patients [24] . In addition, we have previously identified two bat SARS-related-CoV (Bat-CoV YNLF_31C and YNLF_34C) and proposed that the original SARS-CoV full-length orf8 is acquired from these two bat SARS-related-CoV [25] . Since the SARS-CoV is the closest human pathogenic virus to the 2019-nCoV, we performed phylogenetic analysis and multiple alignments to investigate the orf8 amino acid sequences. The orf8 protein sequences used in the analysis derived from early phase SARS-CoV that includes full-length orf8 (human SARS-CoV GZ02), the mid-and late-phase SARS-CoV that includes the split orf8b (human SARS-CoV Tor2), civet SARS-CoV (paguma SARS-CoV), two bat SARS-related-CoV containing full-length orf8 (bat-CoV YNLF_31C and YNLF_34C), 2019-nCoV, the other two closest bat SARS-related-CoV to 2019-nCoV SL-CoV ZXC21 and ZC45), and bat SARS-related-CoV HKU3-1 ( Figure 5(A) ). As expected, orf8 derived from 2019-nCoV belongs to the group that includes the closest genome sequences of bat SARS-related-CoV ZXC21 and ZC45. Interestingly, the new 2019-nCoV orf8 is distant from the conserved orf8 or Figure 5(B) ) which was shown to trigger intracellular stress pathways and activates NLRP3 inflammasomes [26] , but this is absent in this novel orf8 of 2019-nCoV. Based on a secondary structure prediction, this novel orf8 has a high possibility to form a protein with an alpha-helix, following with a betasheet(s) containing six strands ( Figure 5(C) ). The genome of 2019-nCoV has overall 89% nucleotide identity with bat SARS-related-CoV SL-CoVZXC21 (MG772934.1), and 82% with human SARS-CoV BJ01 2003 (AY278488) and human SARS-CoV Tor2 (AY274119). The phylogenetic trees constructed using the amino acid sequences of orf1a/b and the 4 structural genes (S, E, M, and N) were shown (Figure 6(A-E) ). For all these 5 genes, the 2019-nCoV was clustered with lineage B βCoVs. It was most closely related to the bat SARS-related CoVs ZXC21 and ZC45 found in Chinese horseshoe \n\nAs shown in Figure 7 (A-C), the SARS-CoV 5 ′ -UTR contains SL1, SL2, SL3, SL4, S5, SL5A, SL5B, SL5C, SL6, SL7, and SL8. The SL3 contains trans-cis motif [27] . The SL1, SL2, SL3, SL4, S5, SL5A, SL5B, and SL5C structures were similar among the 2019-nCoV, human SARS-CoV and the bat SARS-related ZC45. In the 2019-nCoV, part of the S5 found was inside Figure 7 Continued the orf1a/b (marked in red), which was similar to SARS-CoV. In bat SARS-related CoV ZC45, the S5 was not found inside orf1a/b. The 2019-nCoV had the same SL6, SL7, and SL8 as SARS-CoV, and an additional stem loop. Bat SARS-related CoV ZC45 did not have the SARS-COV SL6-like stem loop. Instead, it possessed two other stem loops in this region. All three strains had similar SL7 and SL8. The bat SARS-like CoV ZC45 also had an additional stem loop between SL7 and SL8. Overall, the 5 ′ -UTR of 2019-nCoV was more similar to that of SARS-CoV than the bat SARS-related CoV ZC 45. The biological relevance and effects of virulence of the 5 ′ -UTR structures should be investigated further. The 2019-nCoV had various 3 ′ -UTR structures, including BSL, S1, S2, S3, S4, L1, L2, L3, and HVR (Figure 7(D-F) ). The 3 ′ -UTR was conserved among 2019-nCoV, human SARS-CoV and SARS-related CoVs [27] . In summary, 2019-nCoV is a novel lineage B Betacoronavirus closely related to bat SARS-related coronaviruses.\n\nQuestion: What would be the benefit of the identity of the S2 unit?", "role": "user" }, { "content": "the broad spectrum antiviral peptides against S2 would be an important preventive and treatment modality for testing in animal models before clinical trials", "role": "assistant" } ]
science.scierc_re
science.scierc_re.14
[ { "content": "You will be given an abstract from a computer science research paper. Given the abstract, your task is to extract all unique relationships between entities of the following, and ONLY the following, types:\n\n- \"USED-FOR\": For instance, \"Our method models user proficiency\" -> [\"method\", \"user proficiency\", \"USED-FOR\"].\n- \"EVALUATE-FOR\": For instance, \"We evaluate our method using F1-measure\" -> [\"F1-measure\", \"method\", \"EVALUATE-FOR\"].\n- \"FEATURE-OF\": For instance, \"prior knowledge of the model\" -> [\"prior knowledge\", \"model\", \"FEATURE-OF\"].\n- \"HYPONYM-OF\": For instance, \"TUIT is a software library\" -> [\"TUIT\", \"software library\", \"HYPONYM-OF\"].\n- \"PART-OF\": For instance, \"We incorporate NLU module into the system\" -> [\"NLU module\", \"system\", \"PART-OF\"].\n- \"COMPARE\": For instance, \"Unlike the quantitative prior, the qualitative prior is often ignored\" -> [\"quantitative prior\", \"qualitative prior\", \"COMPARE\"].\n- \"CONJUNCTION\": For instance, \"obtained from human expert or knowledge base\" -> [\"human expert\", \"knowledge base\", \"CONJUNCTION\"].\n\nFormat your output as a JSON array. Each entry in the array should express a single relation, formatted as [\"<Entity_A>\", \"<Entity_B>\", \"<RELATION_A_B>\"]. An example output might look like:\n\n[[\"neural networks\", \"classification\", \"USED-FOR\"], [\"neuron\", \"neural network\", \"PART-OF\"]].\n\nIf no relations are found, please return an empty JSON array \"[]\".\n\nAbstract:\n\nWe propose a draft scheme of the model formalizing the structure of communicative context in dialogue interaction. The relationships between the interacting partners are considered as system of three automata representing the partners of the dialogue and environment.", "role": "user" }, { "content": "[[\"dialogue interaction\", \"structure of communicative context\", \"FEATURE-OF\"], [\"model\", \"structure of communicative context\", \"USED-FOR\"]]", "role": "assistant" } ]
science.scitldr_aic
science.scitldr_aic.256
[ { "content": "Below are the Abstract, Introduction, and Conclusion sections of a computer science research paper. Please summarize the main contribution of the work in a single sentence. Your response should include the summary sentence and no additional text.\n\nPaper text:\n\nOur work addresses two important issues with recurrent neural networks: (1) they are over-parameterized, and (2) the recurrent weight matrix is ill-conditioned.\nThe former increases the sample complexity of learning and the training time.\nThe latter causes the vanishing and exploding gradient problem.\nWe present a flexible recurrent neural network model called Kronecker Recurrent Units (KRU).\nKRU achieves parameter efficiency in RNNs through a Kronecker factored recurrent matrix.\nIt overcomes the ill-conditioning of the recurrent matrix by enforcing soft unitary constraints on the factors.\nThanks to the small dimensionality of the factors, maintaining these constraints is computationally efficient.\nOur experimental results on seven standard data-sets reveal that KRU can reduce the number of parameters by three orders of magnitude in the recurrent weight matrix compared to the existing recurrent models, without trading the statistical performance.\nThese results in particular show that while there are advantages in having a high dimensional recurrent space, the capacity of the recurrent part of the model can be dramatically reduced.\nDeep neural networks have defined the state-of-the-art in a wide range of problems in computer vision, speech analysis, and natural language processing BID28 BID36 .\nHowever, these models suffer from two key issues.\n(1) They are over-parametrized; thus it takes a very long time for training and inference.\n(2) Learning deep models is difficult because of the poor conditioning of the matrices that parameterize the model.\nThese difficulties are especially relevant to recurrent neural networks.\nIndeed, the number of distinct parameters in RNNs grows as the square of the size of the hidden state conversely to convolutional networks which enjoy weight sharing.\nMoreover, poor conditioning of the recurrent matrices results in the gradients to explode or vanish exponentially fast along the time horizon.\nThis problem prevents RNN from capturing long-term dependencies BID22 BID5 .There\nexists an extensive body of literature addressing over-parametrization in neural networks. BID31\nfirst studied the problem and proposed to remove unimportant weights in neural networks by exploiting the second order information. Several\ntechniques which followed include low-rank decomposition BID13 , training a small network on the soft-targets predicted by a big pre-trained network BID2 , low bit precision training BID12 , hashing BID8 , etc. A notable\nexception is the deep fried convnets BID44 which explicitly parameterizes the fully connected layers in a convnet with a computationally cheap and parameter-efficient structured linear operator, the Fastfood transform BID29 . These techniques\nare primarily aimed at feed-forward fully connected networks and very few studies have focused on the particular case of recurrent networks BID1 .The problem of vanishing\nand exploding gradients has also received significant attention. BID23 proposed an effective\ngating mechanism in their seminal work on LSTMs. Later, this technique was adopted\nby other models such as the Gated Recurrent Units (GRU) BID10 and the Highway networks BID39 for recurrent and feed-forward neural networks respectively. Other popular strategies include\ngradient clipping BID37 , and orthogonal initialization of the recurrent weights . More recently BID1 proposed to use\na unitary recurrent weight matrix. The use of norm preserving unitary\nmaps prevent the gradients from exploding or vanishing, and thus help to capture long-term dependencies. The resulting model called unitary\nRNN (uRNN) is computationally efficient since it only explores a small subset of general unitary matrices. Unfortunately, since uRNNs can only\nspan a reduced subset of unitary matrices their expressive power is limited BID42 . We denote this restricted capacity\nunitary RNN as RC uRNN. Full capacity unitary RNN (FC uRNN\n) BID42 proposed to overcome this issue by parameterizing the recurrent matrix with a full dimensional unitary matrix, hence sacrificing computational efficiency. Indeed, FC uRNN requires a computationally\nexpensive projection step which takes O(N 3 ) time (N being the size of the hidden state) at each step of the stochastic optimization to maintain the unitary constraint on the recurrent matrix. BID35 in their orthogonal RNN (oRNN) avoided\nthe expensive projection step in FC uRNN by parametrizing the orthogonal matrices using Householder reflection vectors, it allows a fine-grained control over the number of parameters by choosing the number of Householder reflection vectors. When the number of Householder reflection vector\napproaches N this parametrization spans the full reflection set, which is one of the disconnected subset of the full orthogonal set. BID25 also presented a way of parametrizing unitary\nmatrices which allows fine-grained control on the number of parameters. This work called as Efficient Unitary RNN (EURNN),\nexploits the continuity of unitary set to have a tunable parametrization ranging from a subset to the full unitary set.Although the idea of parametrizing recurrent weight matrices with strict unitary linear operator is appealing, it suffers from several issues: (1) Strict unitary constraints severely restrict the search space of the model, thus making the learning process unstable. (2) Strict unitary constraints make forgetting irrelevant\ninformation difficult. While this may not be an issue for problems with non-vanishing\nlong term influence, it causes failure when dealing with real world problems that have vanishing long term influence 4.7. BID20 have previously pointed out that the good performance of\nstrict unitary models on certain synthetic problems is because it exploits the biases in these data-sets which favors a unitary recurrent map and these models may not generalize well to real world data-sets. More recently BID41 have also studied this problem of unitary\nRNNs and the authors found out that relaxing the strict unitary constraint on the recurrent matrix to a soft unitary constraint improved the convergence speed as well as the generalization performance.Our motivation is to address the problems of existing recurrent networks mentioned above. We present a new model called Kronecker Recurrent Units (KRU)\n. At the heart of KRU is the use of Kronecker factored recurrent\nmatrix which provide an elegant way to adjust the number of parameters to the problem at hand. This factorization allows us to finely modulate the number of\nparameters required to encode N × N matrices, from O(log(N )) when using factors of size 2 × 2, to O(N 2 ) parameters when using a single factor of the size of the matrix itself. We tackle the vanishing and exploding gradient problem through\na soft unitary constraint BID26 BID20 BID11 BID41 . Thanks to the properties of Kronecker matrices BID40 , this constraint\ncan be enforced efficiently. Please note that KRU can readily be plugged into vanilla real space RNN\n, LSTM and other variants in place of standard recurrent matrices. However in case of LSTMs we do not need to explicitly enforce the approximate\northogonality constraints as the gating mechanism is designed to prevent vanishing and exploding gradients. Our experimental results on seven standard data-sets reveal that KRU and KRU\nvariants of real space RNN and LSTM can reduce the number of parameters drastically (hence the training and inference time) without trading the statistical performance. Our core contribution in this work is a flexible, parameter efficient and expressive\nrecurrent neural network model which is robust to vanishing and exploding gradient problem.The paper is organized as follows, in section 2 we restate the formalism of RNN and detail the core motivations for KRU. In section 3 we present the Kronecker recurrent units (KRU). We present our experimental\nfindings in section 4 and section 5 concludes our work. DISPLAYFORM0\nWe have presented a new recurrent neural network model based on its core a Kronecker factored recurrent matrix.\nOur core reason for using a Kronecker factored recurrent matrix stems from it's elegant algebraic and spectral properties.\nKronecker matrices are neither low-rank nor block-diagonal but it is multi-scale like the FFT matrix.\nKronecker factorization provides a fine control over the model capacity and it's algebraic properties enable us to design fast matrix multiplication algorithms.\nIt's spectral properties allow us to efficiently enforce constraints like positive semi-definitivity, unitarity and stochasticity.\nAs we have shown, we used the spectral properties to efficiently enforce a soft unitary constraint.Experimental results show that our approach out-perform classical methods which uses O(N 2 ) parameters in the recurrent matrix.\nMaybe as important, these experiments show that both on toy problems ( § 4.1 and 4.2), and on real ones ( § 4.3, 4.4, , and § 4.6) , while existing methods require tens of thousands of parameters in the recurrent matrix, competitive or better than state-of-the-art performance can be achieved with far less parameters in the recurrent weight matrix.\nThese surprising results provide a new and counter-intuitive perspective on desirable memory-capable architectures: the state should remain of high dimension to allow the use of high-capacity networks to encode the input into the internal state, and to extract the predicted value, but the recurrent dynamic itself can, and should, be implemented with a low-capacity model.From a practical standpoint, the core idea in our method is applicable not only to vanilla recurrent neural networks and LSTMS as we showed, but also to a variety of machine learning models such as feed-forward networks BID46 , random projections and boosting weak learners.\nOur future work encompasses exploring other machine learning models and on dynamically increasing the capacity of the models on the fly during training to have a perfect balance between computational efficiency and sample complexity.", "role": "user" }, { "content": "Out work presents a Kronecker factorization of recurrent weight matrices for parameter efficient and well conditioned recurrent neural networks.", "role": "assistant" } ]
science.chia_ner
science.chia_ner.1308
[ { "content": "You will be shown inclusion or exclusion criteria for a clinical trial. Given these criteria, your task is to extract all unique entities of the following types: [\"Person\", \"Condition\", \"Value\", \"ReferencePoint\", \"Temporal\", \"Qualifier\", \"Observation\", \"Visit\", \"Measurement\", \"Mood\", \"Device\", \"Scope\", \"Negation\", \"Drug\", \"Procedure\", \"Multiplier\"].\n\nPlease return the output as a JSON object of the format: {\"Condition\": [\"breast cancer\", ...], \"Device\": [\"ports\", ...], \"Drug\": [\"Atenolol\", ...], \"Measurement\": [\"blood pressure\", ...], \"Mood\": [\"suspected\", ...], \"Multiplier\": [\"more than two\", ...], \"Negation\": [\"no\", ...], \"Observation\": [\"history\", ...], \"Person\": [\"male\", ...], \"Procedure\": [\"surgery\", ...], \"Qualifier\": [\"severe\", ...], \"ReferencePoint\": [\"enrollment\", ...], \"Scope\": [\"infection\", ...], \"Temporal\": [\"one day\", ...], \"Value\": [\"50%\", ...], \"Visit\": [\"ICU\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nAll patients who were wheelchair bound preoperatively\nAll patients who cannot participate in an outpatient physical therapy program for 3 days per week after surgery", "role": "user" }, { "content": "{\"Condition\": [], \"Device\": [], \"Drug\": [], \"Measurement\": [], \"Mood\": [], \"Multiplier\": [], \"Negation\": [\"cannot\"], \"Observation\": [\"wheelchair bound\"], \"Person\": [], \"Procedure\": [\"physical therapy\"], \"Qualifier\": [\"preoperatively\"], \"ReferencePoint\": [\"surgery\"], \"Scope\": [], \"Temporal\": [\"for 3 days per week after surgery\"], \"Value\": [], \"Visit\": [\"outpatient\"]}", "role": "assistant" } ]
science.scientific_lay_summarisation_elife_single_doc_summ
science.scientific_lay_summarisation_elife_single_doc_summ.3686
[ { "content": "You will be presented with the abstract, introduction, and discussion section from a biomedical research article. Your task is to create a summary that a layperson can understand, capturing the essence of the research article. Keep essential scientific terms, but ensure that the language remains clear and the concepts are explained in an uncomplicated manner.\n\nTitle: Chronic social isolation reduces 5-HT neuronal activity via upregulated SK3 calcium-activated potassium channels\nArticle:\nAbstract:\nThe activity of serotonin( 5-HT) neurons is critical for mood regulation. In a mouse model of chronic social isolation, a known risk factor for depressive illness, we show that 5-HT neurons in the dorsal raphe nucleus are less responsive to stimulation. Probing the responsible cellular mechanisms pinpoints a disturbance in the expression and function of small-conductance Ca2+-activated K+( SK) channels and reveals an important role for both SK2 and SK3 channels in normal regulation of 5-HT neuronal excitability. Chronic social isolation renders 5-HT neurons insensitive to SK2 blockade, however inhibition of the upregulated SK3 channels restores normal excitability. In vivo, we demonstrate that inhibiting SK channels normalizes chronic social isolation-induced anxiety/depressive-like behaviors. Our experiments reveal a causal link for the first time between SK channel dysregulation and 5-HT neuron activity in a lifelong stress paradigm, suggesting these channels as targets for the development of novel therapies for mood disorders.\nIntroduction:\nMajor depression is a prevalent and debilitating disease for which standard treatments remain ineffective. Social isolation has long been implicated as a risk factor for depression in humans( Cacioppo et al., 2002, 2006; Holt-Lunstad et al., 2010) and induces anxiety- and depressive-like behaviors in rodents( Koike et al., 2009; Wallace et al., 2009; Dang et al., 2015; Shimizu et al., 2016; for review see Fone and Porkess, 2008; Lukkes et al., 2009). The most frequently prescribed medications for depression and anxiety disorders target the serotonin( 5-HT) producing neurons( Blier et al., 1990; Nutt, 2005), the majority of which are located in the dorsal raphe nucleus( DRN)( Descarries et al., 1982). The activity of 5-HT neurons is highly vulnerable to stress( Lira et al., 2003; Bambico et al., 2009; Espallergues et al., 2012; Challis et al., 2013) and is critical for depressive-like, anxiogenic, and reward-associated behaviors( Liu et al., 2014; Teissier et al., 2015; Urban et al., 2016). Interestingly, social isolation in rodents has been shown to affect endogenous 5-HT release and 5-HT turnover in postsynaptic areas( Heidbreder et al., 2000; Muchimapura et al., 2002, 2003). However, it is unknown how social isolation affects the activity of DRN 5-HT neurons themselves. Identification of changes in the activity of these neurons may uncover novel therapeutic targets for depression and anxiety disorders. Here, we show that chronic social isolation leads to a reduction in the excitability of DRN 5-HT neurons. Their firing activity to optical, electrophysiological, and neuromodulatory stimulation are all reduced after social isolation. Specifically, we have identified that the reduction in the firing activity of 5-HT neurons results from alterations in the function and expression of small-conductance Ca2+-activated K+( SK) channels. Furthermore, inhibition of SK channels normalizes the activity of 5-HT neurons and restores the behavioral deficits observed after chronic social isolation.\nDiscussion:\nOur results demonstrate that chronic social isolation results in a reduction in the excitability of 5-HT neurons and renders them less responsive to stimulation. We demonstrate that inhibiting SK channels, critical regulators of AHP in these cells, can reverse the reduced excitability. This work provides the first direct link between a chronic social isolation paradigm and functional alterations in 5-HT neurons themselves. We also reveal that SK2 contributes to normal regulation of 5-HT neuronal excitability but ceases to modulate these neurons significantly after chronic stress. Behavioral analysis showed that chronic social isolation increases anxiety/depression-like behaviors, which can be normalized upon inhibition of SK channels by acute systemic apamin. In humans, polymorphisms on the gene encoding the SK3 channel have been associated with neuropsychiatric disorders characterized by emotional dysregulation including schizophrenia, bipolar disorder, and anorexia nervosa( Chandy et al., 1998; Koronyo-Hamaoui et al., 2007; Grube et al., 2011). Administration of the SK channel blocker apamin in mice and rats reduced immobility in a forced swim test( Galeotti et al., 1999; van der Staay et al., 1999), a measure of depressive-like behavior in rodents( Cryan and Slattery, 2007). Consistent with these reports, SK3 null mice show enhanced hippocampal 5-HT release and reduced immobility in forced swim and tail suspension tests, indicative of an antidepressant like phenotype( Jacobsen et al., 2008). Our work demonstrates for the first time that social isolation, a major risk factor for depression and anxiety, results in reduced 5-HT neuronal activity due to upregulated SK3 channel function in the DRN. Moreover, we show that systemic treatment with apamin, an inhibitor of SK channels improves behavioral deficits induced by chronic social isolation. Consistent with studies that have implicated SK channel function in neuropsychiatric conditions, our current findings suggest that SK channel modulation is a promising therapeutic target for disorders of emotional disturbance such as depression and anxiety. Apamin however has been reported to have adverse side effects in rats( van der Staay et al., 1999) close to the beneficial dose in the current mouse study. The diverse expression of SK channels in different tissue types and the lack of tissue/subtype specific modulators or inhibitors are currently the limiting factors for therapeutic interventions targeting SK channels. Further research focused on understanding the differential regulation, modulation and function of SK channel subtypes may shed light on the development of effective disease treatment strategies. Importantly, development of more specific drugs targeted at inhibiting SK3 function may have significant implications for treatment of depressive and anxiety disorders.", "role": "user" }, { "content": "Major depressive disorder is a common and debilitating disease that interferes with the afflicted person’s everyday life . While some patients do benefit from antidepressant treatments , these medications need to be taken for several weeks before they become effective . Still , a large proportion of patients do not recover fully and some do not respond at all to the existing treatments . As a result , there is a need to find new and more effective treatments for depression . The most widely used antidepressant drugs target the chemical messenger or neurotransmitter called serotonin . The majority of nerve cells that produce serotonin are located in a region deep in the brain known as the dorsal raphe nucleus . When active , these nerve cells release serotonin; this in turn controls the cells’ own activity as well as the activity of a large number of connected nerve cells located throughout the brain . Any disruption in this system will have a widespread impact and can potentially increase the risk of disturbed moods . However , it was not exactly clear what alters the activity of serotonin-producing nerve cells in depression . Now , Sargin et al . have identified a previously unknown mechanism that underlies changes to the activity of serotonin-producing nerve cells . Keeping mice isolated for a prolonged period elicits the symptoms of depression . Sargin et al . found that serotonin-producing nerve cells were dramatically less active in isolated mice and that a specific type of ion channel protein ( the SK3 channel ) was more abundant in these nerve cells . A higher amount of this channel inhibits the activity of nerve cells . Blocking these inhibitory SK3 channels ( using a drug that can be obtained from bee venom ) restored normal activity in the serotonin-producing cells . Moreover , this treatment alleviated the depressive symptoms of the isolated mice . The findings of Sargin et al . suggest a new way to treat the symptoms of depression . Yet to translate them into an accessible treatment for patients , future work will be required to develop drugs that can specifically and potently target the affected channel .", "role": "assistant" } ]
science.acl_arc_intent_classification
science.acl_arc_intent_classification.1498
[ { "content": "You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories:\n- Background: provides context or foundational information related to the topic.\n- Extends: builds upon the cited work.\n- Uses: applies the methods or findings of the cited work.\n- Motivation: cites the work as inspiration or rationale for the research.\n- CompareOrContrast: compares or contrasts the cited work with others.\n- FutureWork: cites the work as a direction for future research.\n\nYour answer should be a single word from the following list of options: [\"Background\", \"Extends\", \"Uses\", \"Motivation\", \"CompareOrContrast\", \"FutureWork\"]. Do not include any other text in your response.\n\nSection Title:\nintroduction\n\nContext before the citation:\nHowever, the literature on Linguistic Steganography, in which linguistic properties of a text are modified to hide information, is small compared with other media (Bergmair, 2007). Given the ubiquitous nature of natural languages and electronic text, text is an obvious medium to consider. Since the difference between 11111111 and 11111110 in the value for red/green/blue intensity is likely to be undetectable by the human eye, the LSB can be used to hide information other than colour, without being perceptable by a human observer.1 A key question for any steganography system is the choice of cover medium.\n\nCitation Sentence:\nHowever , the literature on Linguistic Steganography , in which linguistic properties of a text are modified to hide information , is small compared with other media ( Bergmair , 2007 ) .\n\nContext after the citation:\nThe likely reason is that it is easier to make changes to images and other nonlinguistic media which are undetectable by an observer. Language has the property that even small local changes to a text, e.g. replacing a word by a word with similar meaning, may result in text which is anomalous at the document level, or anomalous with respect to the state of the world. Hence finding linguistic transformations which can be applied reliably and often is a challenging problem for Linguistic Steganography. In this paper we focus on steganography rather than watermarking, since we are interested in the requirement that any changes to a text be imperceptible to an observer.", "role": "user" }, { "content": "Background", "role": "assistant" } ]
science.acl_arc_intent_classification
science.acl_arc_intent_classification.695
[ { "content": "You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories:\n- Background: provides context or foundational information related to the topic.\n- Extends: builds upon the cited work.\n- Uses: applies the methods or findings of the cited work.\n- Motivation: cites the work as inspiration or rationale for the research.\n- CompareOrContrast: compares or contrasts the cited work with others.\n- FutureWork: cites the work as a direction for future research.\n\nYour answer should be a single word from the following list of options: [\"Background\", \"Extends\", \"Uses\", \"Motivation\", \"CompareOrContrast\", \"FutureWork\"]. Do not include any other text in your response.\n\nSection Title:\nconclusion\n\nContext before the citation:\nFor projective parsing, it is significantly faster than exact dynamic programming, at the cost of small amounts of search error, We are interested in extending these ideas to phrase-structure and lattice parsing, and in trying other higher-order features, such as those used in parse reranking (Charniak and Johnson, 2005; Huang, 2008) and history-based parsing (Nivre and McDonald, 2008). Belief propagation improves non-projective dependency parsing with features that would make exact inference intractable.\n\nCitation Sentence:\nFor projective parsing , it is significantly faster than exact dynamic programming , at the cost of small amounts of search error , We are interested in extending these ideas to phrase-structure and lattice parsing , and in trying other higher-order features , such as those used in parse reranking ( Charniak and Johnson , 2005 ; Huang , 2008 ) and history-based parsing ( Nivre and McDonald , 2008 ) .\n\nContext after the citation:\nWe could also introduce new variables, e.g., nonterminal refinements (Matsuzaki et al., 2005), or secondary links Mid (not constrained by TREE/PTREE) that augment the parse with representations of control, binding, etc. (Sleator and Temperley, 1993; Buch-Kromann, 2006). Other parsing-like problems that could be attacked with BP appear in syntax-based machine translation. Decoding is very expensive with a synchronous grammar composed with an n-gram language model (Chiang, 2007)—but our footnote 10 suggests that BP might incorporate a language model rapidly. String alignment with synchronous grammars is quite expensive even for simple synchronous formalisms like ITG (Wu, 1997)—but Duchi et al. (2007) show how to incorporate bipartite matching into max-product BP.", "role": "user" }, { "content": "FutureWork", "role": "assistant" } ]
science.chemsum_single_document_summarization
science.chemsum_single_document_summarization.94215
[ { "content": "You will be shown the title of a chemistry paper, together with sections excerpted from the body of the paper. The paper sections may include HTML tags like `<p>` and `</p>`; please ignore these. Your task is to write an abstract for the paper. Your response should include the abstract and no other text.\n\nTitle:\nCatalytic Asymmetric Amino Acid and Its Derivatives by Chiral Aldehyde Catalysis\n\nPaper sections:\nIntroduction\n<p>In the last decades, the use of aldehyde catalysis is one of the most important and prevalent transformations in organic chemistry (Li et al., 2018; Wang et al., 2019; Yabushita et al., 2019). It has attracted great attention for synthesis chemists to develop efficient aldehyde catalytic systems to mimic a biological process such as amino acid metabolism (Meng et al., 2018). The use of chiral aldehyde catalysis allows the enzyme mimicry and becomes an important synthesis tool in organic chemistry (MacDonald et al., 2011; Gong, 2019). It has become an efficient asymmetric synthesis strategy in amino acid chemistry (Su et al., 2018). The main chiral aldehyde catalysts contain chiral pyridoxal–dependent catalysis (Hoegl et al., 2018; Chen et al., 2019) and chiral binaphthol (BINOL) aldehyde catalysis (Li et al., 2017). Despite chiral pyridoxal analogs being developed and have successfully catalyzed asymmetric aldol condensation of amino acids, nowadays the stereoselectivity needs to be further improved and the development of this catalysis limits substrate activation. Therefore, highly desirable new organo-catalysts are developed to allow flexible and various substrates.</p><p>Except for the research works, some literature reviews have referred to these areas. For example, Subrata Shaw and James D. White summarized recent advances in asymmetric catalysis using chiral salen–metal complexes. It puts an emphasis on asymmetric phase-separative resolution of amino acids. Sun Li et al. reviewed new options for asymmetric organocatalytic reactions of aldehyde catalysis. They described Zhao and his coworkers developed a novel asymmetric Mannich reaction by introducing chiral aldehydes. This review summarizes articles covering the period up to 2018. Based on the previous review works, combining with the latest research articles, the aspects about chiral aldehyde catalysis that used transition-metal catalysis or asymmetric 1,6-conjugated addition and Mannich reactions are reviewed and summarized.</p>\n\nCatalytic Asymmetric Amino Acid and Its Derivatives by Binary Chiral Aldehyde System\n<p>The asymmetric α-functionalization of primary amine and derivatives has an important impact in synthesis chemistry and is widely applied in the preparation of natural products and pharmaceuticals. Guo and his colleagues successfully synthesized a chiral BINOL aldehyde catalyst in 2014, which could directly catalyze the asymmetric α-functionalization reaction of the α-functional group of the N-unprotected amino ester (2-aminomalonate and A-alkylation of 3-indolyl methanol) via imine activation (Xu et al., 2014). In this organocatalytical system, the α-C-H bond of an amine can activate chiral BINOL aldehyde to form enamine through the corresponding 2-aza-allylic anion intermediates and deprotonation. Then the active imine carbon anion was formed and subsequently attacked electrophilic reagent. Finally, α-functional chiral primary amine was generated by the reversible in situ dissociation or amine exchange of imine. Although the enantioselectivity was improved by the utility of chiral BINOL aldehyde, the amino acid substrate could not be effectively extended, and the dosage of aldehyde catalyst is high. The development of efficient catalysts for amine acid transformation remains an elusive task. Inspired by this well-established enamine activation mode, synthetic chemists adopted this concept in the laboratory. Several biomimetic chiral BINOL aldehyde analogs or chiral pyridoxal analogs have been developed and employed in Mannich reactions (Dai et al., 2017), aldol reactions (Li et al., 2008), cyclization reaction (Itoh et al., 2015), and Cope-type hydroaminations of allylic amines (Guimond et al., 2012). In 2018, Zhao and coworkers designed N-quaternized pyridoxals as ideal enzyme mimics and successfully applied them in asymmetric Mannich reactions of glycinate with aryl N-diphenylphosphinyl imines (Chen et al., 2018). The catalyst exhibits high activity and stereoselectivity because of the bifunctional activation mode. It can activate not only glycine ester by the imine activation mode but also diphenylphosphine oxide–protected aromatic imines by hydrogen bond. In the same year, Guo and coworkers also reported a novel type of formyl chiral aldehyde derived from BINOL catalyst for the asymmetric activation of glycine esters through the asymmetric nucleophilic addition/cyclization reaction (Wen et al., 2018). Formyl chiral BINOL aldehyde served as the catalysts to directly activate the α-C–H bond of N-unprotected amines. It was based on the face control of the enolate intermediates. Chiral formyl BINOL aldehyde is an efficient nonpyridoxal-dependent catalyst that can promote the direct asymmetric α-functionalization of N-unprotected glycine esters. The successful usage of 2-formyl chiral BINOL aldehyde catalyst showed a high activity and stereoselectivity. It also expands the substrate including glycine esters, glycine amide, and glycine-derived dipeptides.</p>\n\nChiral Aldehyde Catalysis With Transition Metals\n<p>Despite chiral aldehyde catalysis has become a powerful asymmetric synthesis strategy in amine chemistry, the development of catalysts for amine chemistry remains a great challenge. Inspired by a level of reactivity that could be achievable with the combinations of organic and transition-metal catalysts, Lei Chen and his coworkers provided a good strategy for the combination of chiral aldehyde catalysts with transition metals (Chen et al., 2019). In 2019, they provided a ternary catalytic system that is the combination of the Zn–Schiff base complex, chiral BINOL aldehyde, and Lewis acid (as shown in Figure 1). They successfully applied them in α-allylation reaction of N-unprotected amino acid esters and allyl acetates, exhibiting a good yield and enantioselectivity. With the expansion of the application of chiral aldehyde catalyst, new catalytic technologies are emerging. Guo and his coworkers have disclosed chiral 3-formyl BINOL aldehyde catalyst of anti–1,6-conjugate addition and syn-Mannich reaction that was inspired by the intermediate ketimine in transamination. Another chiral 2-formyl BINOL aldehyde catalyst gave the syn-selective conjugate addition and anti-Mannich products. The good yields obtained both the syn- and anti-products of transformations with high diastereo- and enantioselectivity.</p>\n\n\n<p>Combining chiral aldehyde catalysis and transition-metal catalysis for enantioselective α-allylic alkylation of amino acid esters (Chen et al., 2019).</p>\n\nChiral Aldehyde Catalysis With Transition Metals\n<p>The novel chiral 3-formyl BINOL aldehyde catalyst was displayed in the study of the Mannich reaction of 1,6-conjugated addition of amino acids (as shown in Figure 2) (Wen et al., 2020). The reaction went well in the presence of 4g chiral aldehyde and mesitylene as the solvent and tBuOK as a base, leading to good yields with 90% ee with enantioselectivity. Notably, the catalyst loading was 10 mol%, and the reaction was completed after 7 h. Further studies showed that the substituted imines and pyridinylmethanamines in the syn-Mannich reaction could also be employed, although it required higher catalyst loading (20 mol%) and longer reaction time. Anti-product yields were affected by substituent steric and electronics. The developed catalytic system showed a quite broad substrate scope. A variety of para-quinone methide derivatives and amino acid substrates for the anti-selective reaction reacted well with chiral 3-formyl BINOL aldehyde catalyst to give the desired products in 25–84% yields with excellent enantioselectivities. However, it should be noted that when there is an imine bearing an electron-rich aryl substituent, the product yield decreased greatly. These indicated the aldehydes' influence on the yield and formation of product and that electron-rich aldehydes worked well in this reaction. To demonstrate a stereoselective control, four potential reaction models were produced by DFT calculations. The key factor for the diastereoselectivity and enantioselectivity of catalysts is the steric effect of the R group. In the syn-Mannich reaction, phenyl imines bearing a single substituent on the phenyl ring leads to favorable reactivity, giving syn-products with good yields, diastereoselectivities, and enantioselectivities. Another key factor for the catalytic system is the strength of alkali. Because Schiff base intermediates are formed by alkali compounds, it was a critical factor for the nucleophilic attack to stereoselectivity. The demand for highly active electrophiles of chiral aldehyde catalysis strongly limits the application.</p>\n\n\n<p>The 1,6-conjugated addition reaction by chiral BINOL aldehyde catalysts (Wen et al., 2020).</p>\n\nChiral Aldehyde Catalysis With Transition Metals\n<p>Chiral aldehyde catalysis promotes new methods to activate simple amines. However, no matter in the binary or ternary catalytic systems in asymmetric synthesis, the structurally diverse chiral aldehyde catalysts that were hardly achieved hampered the development of this field. Therefore, different aldehyde catalytic systems have been established. Professor Bing-Feng Shi has reported a system that synthesis of chiral aldehyde catalysts by Pd-catalyzed atroposelective C–H naphthylation (Liao et al., 2019; Liao et al., 2020). Jumpei Taguchi and his coworkers have synthetized potassium acyl trifluoroborates from aldehydes through copper (I)-catalyzed borylation/oxidation (Taguchi et al., 2019). Hirohisa Ohmiya et al. have gotten dialkyl ketones from aliphatic aldehydes through radical N-heterocyclic carbene catalysis (Yabushita et al., 2019; Kakeno et al., 2020). Sungwoo Hong group have employed an N-heterocyclic carbene (NHC) catalyst, which have developed a versatile catalytic system that enables deaminative cross-coupling reactions of aldehydes with redox-active pyridinium salts (Kim et al., 2020).</p>\n\nConclusion\n<p>In summary, the role of asymmetric catalysis by chiral aldehyde has grown rapidly in chemical reactions of amino acids. One limit is the reaction types. One more important limitation is how to get structurally diverse chiral aldehyde. Compared with binary catalytic systems, chiral aldehyde catalysts in ternary catalytic system not only acts as organic catalysts to activate amino acid and its derivatives through the formation of Schiff bases but also acts as ligands to promote the nucleophilic attack process. Importantly, compatibility also needs to be considered. Although the asymmetric catalysis involving chiral aldehyde has been grown in recent years, they are still rare. It is believed that the organic reactions catalyzed by chiral aldehydes are very important kind of asymmetric catalytic reactions, although some difficulties remain and new challenges will emerge. It is also believed that more researchers would involve and new and efficient chiral aldehyde catalysts will be developed.</p>\n\nAuthor Contributions\n<p>KL, AS, and JL conceived and wrote the manuscript. All other authors (CS and HL) provided comments and discussion on the manuscript to aid its preparation.</p>\n\nFunding\n<p>Natural Science Foundation of Fujian Province (2019I0014).</p>\n\nConflict of Interest\n<p>The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.</p>", "role": "user" }, { "content": "Amine acid transformation is an important chemical process in biological systems. As a well-developed and acknowledged tool, chiral aldehyde catalysis provides good catalytic activation and stereoselective control abilities in the asymmetric reaction of N-unprotected amino acid esters and amino acid esters analogs, in which the key to success is the design of the catalysts derived from chiral BINOL aldehyde, which is based on the face control of enolate intermediates. In this review, one of the co-catalytic systems that combined with a transition metal to form a multiplex catalytic system and the well-established multiplex stereocenters of chiral aldehyde catalysis have been reviewed. Finally, a novel organocatalysis is prospected.", "role": "assistant" } ]
science.bc7_litcovid_topic_classification
science.bc7_litcovid_topic_classification.20074
[ { "content": "You will be presented with a paragraph from a COVID-19 literature. Your task is to classify the topics of this piece of text by choosing from the following categories:\n\n- Epidemic Forecasting: The content relates to predictions about the spread of COVID-19.\n- Treatment: The paragraph discusses therapeutic methods or medication for COVID-19.\n- Prevention: The content covers measures to prevent the spread of COVID-19.\n- Mechanism: The paragraph explains the biological mechanisms of the virus.\n- Case Report: The content is a report of a COVID-19 case.\n- Transmission: The paragraph discusses how COVID-19 is transmitted.\n- Diagnosis: The content is about diagnosing COVID-19 in individuals.\n\nYour classification should be provided as an element within a JSON array. The array should be formatted with topics enclosed in double quotes and separated by commas, following the JSON format, for example: [\"Topic1\", \"Topic2\", ...]. Include all applicable categories within the array. \nYour output should be a valid JSON array, and no additional text should be included in your response.\n\nArticle:\n\nBACKGROUND: In December, 2019, a novel coronavirus disease 2019 (COVID-19) emerged in Wuhan, China. We aimed to clarify the epidemiology, laboratory examinations, imaging findings, and treatment of critically ill patients with COVID-19 in Hebei province, China. METHODS: In this retrospective study, the demographic, laboratory and imaging, and treatment data of patients with severe COVID-19 treated in 13 designated hospitals in Hebei were collected and analyzed. RESULTS: A total of 319 severe COVID-19 patients were treated at the 13 designated hospitals between 22 January, 2020 and 25 March, 2020. Eventually, 51 critically ill (31 severe cases and 20 critically severe cases) patients were included in the analysis. The patients had an average age of 58.9+/-13.7 years, and 27 (52.9%) were men. Twenty-one (41.2%) were familial cluster, and 33 (64.7%) had chronic illnesses. The patients in critically severe group had longer duration from symptom to confirmation, more severe infections, more severe lung injury, and a lower percentage of lymphocytes. All 51 patients received antiviral drugs, 47 (92.2%) received antibacterial agents, 49 (96.1%) received traditional Chinese drugs, and 46 (90.2%) received methylprednisolone. The critically severe patients received more fluid and more diuretic treatment; 14 (70.0%) required invasive mechanical ventilation, and 13 (65.0%) developed extrapulmonary complications. CONCLUSIONS: COVID-19 patients who had underlying diseases and longer confirmation times were more likely to progress to critically severe COVID-19. These patients also presented with a higher risk of respiratory depression, circulatory collapse, extrapulmonary complications, and infection.", "role": "user" }, { "content": "[\"Treatment\", \"Diagnosis\"]", "role": "assistant" } ]
science.scientific_papers_summarization_single_doc_arxiv
science.scientific_papers_summarization_single_doc_arxiv.173319
[ { "content": "Write an abstract for the research paper below. Your response should only include the abstract and no additional text.\n\nArticle text:\nsome of the hubble space telescope s most ground - braking discoveries have been about black holes , which are arguably the most extreme and mysterious objects in the universe . as a result\n, black holes appeal to a general audience in a way that almost no other scientific subject does .\nunfortunately , most people have little idea of what black holes actually are , and they are more likely to associate them with science fiction than with science .\nthese circumstances recommend black holes as a natural topic for an education and public outreach ( e / po ) activity . to this end\n, we have created a website called `` black holes : gravity s relentless pull , '' which serves as the e / po component of several hubble observing projects .\nthe new website is part of hubblesite , the internet home of all hubble news .\nthe url is .\nmany sites on the internet already explain black holes in one way or other .\nmost are encyclopedic , offering detailed text and graphics .\nby contrast , our site emphasizes user participation , and it is rich in animations and audio features .\nthis approach was facilitated by the availability of powerful software for authoring multimedia content .\nthe result is a website where scientific knowledge , learning theory , advanced technology , and pure fun all converge .\nthe opening animation at our website introduces the basic concept of a black hole .\nit shows how one could turn the earth into a black hole , if only one could shrink it to the size of a marble . by connecting enigma and commonplace ,\nthis scenario lends black holes a seeming familiarity .\nthe core of the website consists of three sequential , interactive modules , of which `` finding the invisible '' is the first .\nit shows the night sky with a viewfinder that can be dragged around to discover images of about a dozen objects , including some that should be familiar to the user ( the sun , moon , and saturn ) and others that only an astronomer might recognize ( betelgeuse , crab nebula , cygnus x-1 , andromeda galaxy , and quasar 3c273 ) . the user can select the wavelength range of the viewfinder : visible light , radio waves , or x - rays .\nthe goal is to teach which types of objects contain black holes and which do not .\ninterested users can move to pages that explain the various objects , the telescopes used to observe them , and the features in the images at different wavelengths that indicate the presence of a black hole . in this way\nthe user not only learns about black holes , but also about their relation to other objects in the universe and the methods that astronomers use to study them . when the user has found one or more black holes , he or she can choose to go to the second module , `` the voyage , '' which offers a multimedia trip in an animated starship to a nearby black hole , either a stellar - mass black hole ( cygnus x-1 ) or a supermassive black hole ( in the center of the andromeda galaxy ) .\nthe viewer traverses the solar system as well as our own milky way galaxy .\nvarious intriguing objects are encountered along the way , including many of the objects previously encountered in the first module .\na goal is to connect the two - dimensional , projected view of the night sky to the actual three - dimensional structure of the universe . in the process\n, the user learns about the distance scale and the layout of the local universe .\nafter the spaceship has arrived at the black hole , the user can proceed to the third module , `` get up close . ''\norbiting around a black hole , the sky outside the spaceship window looks strangely distorted , because of the strong gravitational lensing of distant starlight .\nthe module discusses this and many other fascinating phenomena to be encountered near black holes .\nfive interactive experiments are offered to explore specific issues . ``\ncreate a black hole '' allows the user to study the evolution of stars of different masses by trying to create a black hole , rather than a white dwarf or neutron star .\n`` orbit around a black hole '' plots relativistically correct orbits around a black hole , showing how it is possible to orbit a black hole without being sucked in .\n`` weigh a black hole '' shows how to use observations of a black hole in a binary system to calculate its mass .\nthis illustrates learning without seeing .\n`` drop a clock into a black hole '' explains time dilation and redshift .\nmysteriously , an outside observer never sees a falling clock disappear beyond the event horizon , which highlights the principle of relativity .\nthe last experiment , `` fall into a black hole '' allows the viewer to take the final plunge and witness how one s body gets stretched by tidal forces . with this , the user s journey is complete , from the backyard view of the night sky to a personal encounter with a singularity .\nour motivations for adopting a strongly interactive website were grounded in specific learning theories , `` constructivism '' in particular .\nconstructivism holds that learning is not merely an addition of items into a mental data bank , but rather requires a transformation of concepts in which the learner plays an active role making sense out of a range of phenomena .\nresearch has shown that people use various learning styles to perceive and process information .\nfor these reasons , we structured the scientific content in multiple ways at our website , to engage a range of learners in a variety of meaningful ways .\nnovices and children often prefer interactive learning experiences , which can motivate them to stay engaged until they achieve some payoff .\nthe interactive modules and experiments of the black hole website are tailored for this audience .\nthey offer goal - based scenarios , to motivate learning and a sense of accomplishment when completed .\nby contrast , experts and adults often prefer formal organization and ready access to information that they already know about and just want to locate . to reach this audience\n, the site also contains an encyclopedia of faqs frequently asked questions about the physics and astronomy of black holes , as well as a detailed glossary .\nwe are currently working on a modified version of the web site that can be used as a kiosk exhibit in museums , science centers and planetaria .\nthis will further broaden the audience of the project to include those people who do not have access to the internet , or who typically do not use it to broaden their knowledge .\nthe entire site is based on the most recent scientific knowledge .\nrecent results and ongoing projects are highlighted where possible .\nwe will continue to update the site as new discoveries emerge .\nastronomical images from state - of - the - art telescopes are used in abundance , with a particular focus on hubble .\nextensive links are provided to other websites on the internet , which allows users to explore specific subjects in more detail .\neven with its sharp focus on black holes , our site strives to teach much more .\nwe illustrate how humans can understand the universe by detailed observations of the night sky .\nwe teach basic concepts , like light and gravity .\nwe show how different perspectives can enrich our understanding of nature .\nwe illustrate the many wonders of our universe , and demonstrate its scale by alerting the user to what is really near and what is really far away . and we highlight the many things about black holes and our universe that we still do not understand .\nin the broadest sense , our goal has been to show that even the most mysterious of things can be understood with the combined application of human thinking and powerful technology .\nwe hope to convey the importance of scientific thought and to instill an appreciation for learning and an interest in science , especially in the younger generation of users .\nwe are grateful to the many people who assisted with the creation of the website , or who provided graphics , animations , or software .\na full list of credits is available at + http:/@xmath0/hubblesite.org / discoveries / black_holes / credits.html .\nwe would also like to thank bob brown for editorial assistance with this article .", "role": "user" }, { "content": "we have created a website , called `` black holes : gravity s relentless pull '' , which explains the physics and astronomy of black holes for a general audience . \n the site emphasizes user participation and is rich in animations and astronomical imagery . \n it won the top prize of the 2005 pirelli internetional awards competition for the best communication of science and technology using the internet . \n this article provides a brief overview of the site . \n the site starts with an opening animation that introduces the basic concept of a black hole . \n the user is then invited to embark on a journey from a backyard view of the night sky to a personal encounter with a singularity . \n this journey proceeds through three modules , which allow the user to : find black holes in the night sky ; travel to a black hole in an animated starship ; and explore a black hole from up close . \n there are also five `` experiments '' that allow the user to : create a black hole ; orbit around a black hole ; weigh a black hole ; drop a clock into a black hole ; or fall into a black hole . \n the modules and experiments offer goal - based scenarios tailored for novices and children . \n the site also contains an encyclopedia of frequently asked questions and a detailed glossary that are targeted more at experts and adults . \n the overall result is a website where scientific knowledge , learning theory , and fun converge . despite its focus on black holes \n , the site also teaches many other concepts of physics , astronomy and scientific thought . \n the site aims to instill an appreciation for learning and an interest in science , especially in the younger users \n . it can be used as an aid in teaching introductory astronomy at the undergraduate level .", "role": "assistant" } ]
science.scierc_re
science.scierc_re.155
[ { "content": "You will be given an abstract from a computer science research paper. Given the abstract, your task is to extract all unique relationships between entities of the following, and ONLY the following, types:\n\n- \"USED-FOR\": For instance, \"Our method models user proficiency\" -> [\"method\", \"user proficiency\", \"USED-FOR\"].\n- \"EVALUATE-FOR\": For instance, \"We evaluate our method using F1-measure\" -> [\"F1-measure\", \"method\", \"EVALUATE-FOR\"].\n- \"FEATURE-OF\": For instance, \"prior knowledge of the model\" -> [\"prior knowledge\", \"model\", \"FEATURE-OF\"].\n- \"HYPONYM-OF\": For instance, \"TUIT is a software library\" -> [\"TUIT\", \"software library\", \"HYPONYM-OF\"].\n- \"PART-OF\": For instance, \"We incorporate NLU module into the system\" -> [\"NLU module\", \"system\", \"PART-OF\"].\n- \"COMPARE\": For instance, \"Unlike the quantitative prior, the qualitative prior is often ignored\" -> [\"quantitative prior\", \"qualitative prior\", \"COMPARE\"].\n- \"CONJUNCTION\": For instance, \"obtained from human expert or knowledge base\" -> [\"human expert\", \"knowledge base\", \"CONJUNCTION\"].\n\nFormat your output as a JSON array. Each entry in the array should express a single relation, formatted as [\"<Entity_A>\", \"<Entity_B>\", \"<RELATION_A_B>\"]. An example output might look like:\n\n[[\"neural networks\", \"classification\", \"USED-FOR\"], [\"neuron\", \"neural network\", \"PART-OF\"]].\n\nIf no relations are found, please return an empty JSON array \"[]\".\n\nAbstract:\n\nThis paper examines the properties of feature-based partial descriptions built on top of Halliday 's systemic networks. We show that the crucial operation of consistency checking for such descriptions is NP-complete, and therefore probably intractable, but proceed to develop algorithms which can sometimes alleviate the unpleasant consequences of this intractability.", "role": "user" }, { "content": "[[\"consistency checking\", \"feature-based partial descriptions\", \"USED-FOR\"], [\"Halliday \\u0027s systemic networks\", \"feature-based partial descriptions\", \"USED-FOR\"]]", "role": "assistant" } ]
science.bioasq_general_qa
science.bioasq_general_qa.607
[ { "content": "You will be shown a question, followed by excerpts from biomedical research papers. Please answer the question based on the provided context. Do not include any text in your response other than the answer.\n\nQuestion: Describe JACUSA\n\nContext: JACUSA: site-specific identification of RNA editing events from replicate sequencing data.\nWe present JACUSA, a versatile one-stop solution to detect single nucleotide variant positions from comparing RNA-DNA and/or RNA-RNA sequencing samples. The performance of JACUSA has been carefully evaluated and compared to other variant callers in an in silico benchmark. JACUSA outperforms other algorithms in terms of the F measure, which combines precision and recall, in all benchmark scenarios. This performance margin is highest for the RNA-RNA comparison scenario. We further validated JACUSA's performance by testing its ability to detect A→I events using sequencing data from a human cell culture experiment and publicly available RNA-seq data from Drosophila melanogaster heads. To this end, we performed whole genome and RNA sequencing of HEK-293 cells on samples with lowered activity of candidate RNA editing enzymes. JACUSA has a higher recall and comparable precision for detecting true editing sites in RDD comparisons of HEK-293 data. Intriguingly, JACUSA captures most A→I events from RRD comparisons of RNA sequencing data derived from Drosophila and HEK-293 data sets.\nOur software JACUSA detects single nucleotide variants by comparing data from next-generation sequencing experiments (RNA-DNA or RNA-RNA). In practice, JACUSA shows higher recall and comparable precision in detecting A→I sites from RNA-DNA comparisons, while showing higher precision and recall in RNA-RNA comparisons.\nEven more surprising, condition-specific editing events, which would show up as differences in RNA-RNA comparisons (RRDs) and depend on particular cellular states, are rarely discussed in the literature.<br><b>RESULTS</b>: We present JACUSA, a versatile one-stop solution to detect single nucleotide variant positions from comparing RNA-DNA and/or RNA-RNA sequencing samples.\nIntriguingly, JACUSA captures most A→I events from RRD comparisons of RNA sequencing data derived from Drosophila and HEK-293 data sets.<br><b>CONCLUSION</b>: Our software JACUSA detects single nucleotide variants by comparing data from next-generation sequencing experiments (RNA-DNA or RNA-RNA).\nCONCLUSION Our software JACUSA detects single nucleotide variants by comparing data from next-generation sequencing experiments (RNA-DNA or RNA-RNA).\nRESULTS We present JACUSA, a versatile one-stop solution to detect single nucleotide variant positions from comparing RNA-DNA and/or RNA-RNA sequencing samples.\nWe present JACUSA, a versatile one-stop solution to detect single nucleotide variant positions from comparing RNA-DNA and/or RNA-RNA sequencing samples.", "role": "user" }, { "content": "JACUSA detects single nucleotide variants by comparing data from next-generation sequencing experiments (RNA-DNA or RNA-RNA). In practice, JACUSA shows higher recall and comparable precision in detecting A→I sites from RNA-DNA comparisons, while showing higher precision and recall in RNA-RNA comparisons.", "role": "assistant" } ]
science.data_reco_mcq_sc
science.data_reco_mcq_sc.1423
[ { "content": "Given a short query with keyphrases about a research idea in 2015 in the domain of artificial intelligence, recommend the relevant datasets to validate the research idea from the options provided. A short description of the datasets is also provided to help you.\n\nQuery: A fast, scalable algorithm to embed large-scale information networks.\n\nKeyphrases: network embedding information\n\nDataset description:\nAnimalWeb: A large-scale, hierarchical annotated dataset of animal faces, featuring 21.9K faces from 334 diverse species and 21 animal orders across biological taxonomy. These faces are captured `in-the-wild' conditions and are consistently annotated with 9 landmarks on key facial features.\nDBLP: The DBLP is a citation network dataset. The citation data is extracted from DBLP, ACM, MAG (Microsoft Academic Graph), and other sources.\nWebVision: The WebVision dataset is designed to facilitate the research on learning visual representation from noisy web data. It is a large scale web images dataset that contains more than 2.4 million of images crawled from the Flickr website and Google Images search.\nSketchyScene: SketchyScene is a large-scale dataset of scene sketches to advance research on sketch understanding at both the object and scene level. The dataset is created through a novel and carefully designed crowdsourcing pipeline, enabling users to efficiently generate large quantities of realistic and diverse scene sketches.\n\nOptions: A: AnimalWeb B: DBLP C: WebVision D: SketchyScene", "role": "user" }, { "content": "DBLP", "role": "assistant" } ]
science.chia_ner
science.chia_ner.1024
[ { "content": "You will be shown inclusion or exclusion criteria for a clinical trial. Given these criteria, your task is to extract all unique entities of the following types: [\"Person\", \"Condition\", \"Value\", \"ReferencePoint\", \"Temporal\", \"Qualifier\", \"Observation\", \"Visit\", \"Measurement\", \"Mood\", \"Device\", \"Scope\", \"Negation\", \"Drug\", \"Procedure\", \"Multiplier\"].\n\nPlease return the output as a JSON object of the format: {\"Condition\": [\"breast cancer\", ...], \"Device\": [\"ports\", ...], \"Drug\": [\"Atenolol\", ...], \"Measurement\": [\"blood pressure\", ...], \"Mood\": [\"suspected\", ...], \"Multiplier\": [\"more than two\", ...], \"Negation\": [\"no\", ...], \"Observation\": [\"history\", ...], \"Person\": [\"male\", ...], \"Procedure\": [\"surgery\", ...], \"Qualifier\": [\"severe\", ...], \"ReferencePoint\": [\"enrollment\", ...], \"Scope\": [\"infection\", ...], \"Temporal\": [\"one day\", ...], \"Value\": [\"50%\", ...], \"Visit\": [\"ICU\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nPreexisting ocular diseases or conditions other than age related cataracts, have contraindications for cataract surgery;\nPreexisting systemic diseases or conditions that may confound the results of the study;\nPrevious ocular surgery history or ocular trauma that may confound the results of the study;\nRequire combined surgery that may confound the results of the study;\nPrevious participation in other clinical trial within 30 days of this study start;\nSystemic or ocular medications that may confound the outcome of the intervention\nPregnant, lactating, or planning to become pregnant during the course of the trial;", "role": "user" }, { "content": "{\"Condition\": [\"conditions\", \"Pregnant\", \"contraindications\", \"ocular diseases\", \"lactating\", \"pregnant\", \"ocular trauma\", \"cataracts\", \"systemic diseases\"], \"Device\": [], \"Drug\": [\"mic or o ations that\", \"ocular medications\"], \"Measurement\": [], \"Mood\": [\"Require\", \"planning to become\"], \"Multiplier\": [], \"Negation\": [\"other than\"], \"Observation\": [], \"Person\": [], \"Procedure\": [\"cataract surgery\", \"ocular surgery\", \"combined surgery\"], \"Qualifier\": [\"may confound the results of the study\", \"may confound the outcome of the intervention\", \"age related\"], \"ReferencePoint\": [], \"Scope\": [\"Pregnant, lactating, or planning to become pregnant\", \"Systemic or ocular medications\", \"Previous ocular surgery history or ocular trauma\"], \"Temporal\": [\"Previous\", \"during the course of the trial\", \"Preexisting\"], \"Value\": [], \"Visit\": []}", "role": "assistant" } ]
science.chemdner_ner
science.chemdner_ner.854
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\": [\"Arsenic\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nMechanism of maprotiline-induced apoptosis: role of [Ca2+](i), ERK, JNK and caspase-3 signaling pathways. Antidepressants are generally used for treatment of various mood and anxiety disorders. Several studies have shown the anti-tumor and cytotoxic activities of some antidepressants, but the underlying mechanisms were unclear. Maprotiline is a tetracyclic antidepressant and possesses a highly selective norepinephrine reuptake ability. We found that maprotiline decreased cell viability in a concentration- and time-dependent manner in Neuro-2a cells. Maprotiline induced apoptosis and increased caspase-3 activation. The activation of caspase-3 by maprotiline appears to depend on the activation of JNK and the inactivation of ERK. Maprotiline also induced [Ca(2+)](i) increases which involved the mobilization of intracellular Ca(2+) stored in the endoplasmic reticulum. Pretreatment with BAPTA/AM, a Ca(2+) chelator, suppressed maprotiline-induced ERK phosphorylation, enhanced caspase-3 activation and increased maprotiline-induced apoptosis. In conclusion, maprotiline induced apoptosis in Neuro-2a cells through activation of JNK-associated caspase-3 pathways. Maprotiline also evoked an anti-apoptotic response that was both Ca(2+)- and ERK-dependent.", "role": "user" }, { "content": "{\"Chemical\": [\"Maprotiline\", \"BAPTA\", \"Ca(2+)\", \"norepinephrine\", \"maprotiline\", \"Ca2+\"]}", "role": "assistant" } ]
science.scientific_lay_summarisation_elife_single_doc_summ
science.scientific_lay_summarisation_elife_single_doc_summ.3668
[ { "content": "You will be presented with the abstract, introduction, and discussion section from a biomedical research article. Your task is to create a summary that a layperson can understand, capturing the essence of the research article. Keep essential scientific terms, but ensure that the language remains clear and the concepts are explained in an uncomplicated manner.\n\nTitle: Heart of glass anchors Rasip1 at endothelial cell-cell junctions to support vascular integrity\nArticle:\nAbstract:\nHeart of Glass( HEG1), a transmembrane receptor, and Rasip1, an endothelial-specific Rap1-binding protein, are both essential for cardiovascular development. Here we performed a proteomic screen for novel HEG1 interactors and report that HEG1 binds directly to Rasip1. Rasip1 localizes to forming endothelial cell( EC) cell-cell junctions and silencing HEG1 prevents this localization. Conversely, mitochondria-targeted HEG1 relocalizes Rasip1 to mitochondria in cells. The Rasip1-binding site in HEG1 contains a 9 residue sequence, deletion of which abrogates HEG1’s ability to recruit Rasip1. HEG1 binds to a central region of Rasip1 and deletion of this domain eliminates Rasip1’s ability to bind HEG1, to translocate to EC junctions, to inhibit ROCK activity, and to maintain EC junctional integrity. These studies establish that the binding of HEG1 to Rasip1 mediates Rap1-dependent recruitment of Rasip1 to and stabilization of EC cell-cell junctions.\nIntroduction:\nPrecisely regulated cell-cell adhesion is crucial for the development and functioning of multicellular organisms( Nelson, 2008). In vascular endothelial cells( EC), cell-cell contacts formed by adherens and tight junctions( Bazzoni, 2004), control vascular patterning and provide a barrier to the passage of water and solutes. Weakening of this endothelial barrier contributes to diseases such as sepsis, atherosclerosis, and multiple sclerosis( Bazzoni, 2004; Dejana et al., 2009). The actin cytoskeleton( Mège et al., 2006) and its regulatory Rho GTPases have profound roles in the assembly and disassembly of cell-cell contacts( Beckers et al., 2010; Yamada and Nelson, 2007). Rap1, a small GTPase, regulates epithelial and EC cell-cell adhesion( Pannekoek et al., 2014; Kooistra et al., 2007; Knox, 2002). In EC, binding of activated Rap1 to KRIT1( Gingras et al., 2013; Li et al., 2012), a scaffold that regulates RhoA signaling( Whitehead et al., 2009; Stockton et al., 2010), targets KRIT1 to cell-cell contacts( Liu et al., 2011; Glading et al., 2007; Béraud-Dufour et al., 2007) where it directly interacts with Heart of Glass( HEG1)( Gingras et al., 2012), a transmembrane receptor essential for cardiovascular development( Mably et al., 2003; Kleaveland et al., 2009). The 111 amino acid HEG1 cytoplasmic domain is highly conserved( e. g. 76% identity between human and zebrafish) whereas KRIT1 binding requires only the C-terminal 10% of HEG1 tail( Gingras et al., 2012). Thus, we suspected that other interactors of HEG1 might contribute to vascular function. Here, we report an unbiased proteomic screen for HEG1-binding endothelial-specific proteins that revealed that Rasip1, a second Rap1 binding protein essential for vascular integrity and development( Post et al., 2013; Xu et al., 2009; Xu et al., 2011; Wilson et al., 2013), directly interacts with the HEG1 cytoplasmic domain independent of KRIT1. Rasip1 recruitment to cell-cell junctions requires the presence of HEG1 and integrity of its HEG1 binding site. Moreover, targeting of HEG1 tail to mitochondria recruits Rasip1 to these organelles through the direct interaction of the proteins. We designed a structure-based Rasip1 RA-domain mutant( R182E) that failed to bind to Rap1 and did not associate with HEG1 in cells or maintain EC barrier integrity. Finally, the central region of Rasip1 interacts with HEG1 and deletion of this domain eliminates Rasip1’s ability to translocate to EC junctions, to inhibit Rho Kinase( ROCK) activity, and to maintain EC junctional integrity. These studies reveal that HEG1 interacts with two different Rap1 effectors important in EC, indicating that this transmembrane receptor serves as a physical nexus for Rap1 signaling in vascular integrity and development.\nDiscussion:\nHeart of Glass( HEG1), a transmembrane protein, is essential for vertebrate cardiovascular development( Mably et al., 2003; Kleaveland et al., 2009) and localizes KRIT1 to EC junctions. Here, we report that Rasip1, a Rap1 binding protein required for vascular development, binds directly to the cytoplasmic domain( tail) of HEG1, independent of KRIT1. Silencing HEG1 prevented Rasip1 targeting to EC junctions, furthermore the HEG1 cytoplasmic domain could re-localize Rasip1 in living cells. We mapped the Rasip1 binding site in HEG1 to a 9 residue peptide that is highly conserved in vertebrates and found that deletion of this sequence blocks the capacity of HEG1 to bind to Rasip1 and to recruit Rasip1 in cells. Similarly, we mapped the HEG1 binding site in Rasip1 to its central region, containing an FHA domain, and found that deletion of this domain abolished the capacity of Rasip1 to translocate to cell-cell contacts, to inhibit ROCK activity, and to maintain EC junctional integrity. Rap1 activity regulates the function and localization of Rasip1; using molecular modeling we designed and validated a well-folded mutant, Rasip1( R182E), that failed to bind Rap1. Rasip1( R182E) also failed to associate with the HEG1 tail in cells and lost the capacity to inhibit ROCK and maintain EC junctional integrity. These studies establish that the interaction of HEG1 and Rasip1, two proteins involved in vascular development, is regulated by Rap1 binding to Rasip1 and enables Rasip1 to localize to EC cell-cell junctions and to maintain vascular integrity. The binding of a Rasip1 FHA domain to a 9 amino acid( TDVYYSPTS) region in HEG1 mediates Rasip1 localization and function in EC. Rasip1 is recruited to assembling EC cell-cell junctions and we observed that HEG1 silencing blocked this recruitment and that targeting mito-mCherry-HEG1 to mitochondria re-localized endogenous Rasip1 to these organelles. Mapping studies identified HEG1( 1327-1335) as the critical region and deleting this site abrogated the capacity of HEG1 to bind Rasip1 and to recruit it to mito-mCherry-HEG1( 1327-1335). Rasip( 266-550), which contains an FHA domain, was necessary and sufficient for the interaction with the HEG1 tail. In sum, our studies show that the direct physical interaction of a HEG1 with Rasip1 targets Rasip1 to assembling EC junctions. Rasip1 regulation of ROCK activity and vascular permeability is dependent on the interaction with HEG1. In contrast to wild-type Rasip1, HEG1-binding deficient Rasip1( ∆334-539) was not recruited to endothelial cell-cell contacts during junction formation/remodeling. Groundbreaking studies( Xu et al., 2011) revealed that Rasip1 interacted with a RhoGAP( ARHGAP29) regulating RhoA and its effector ROCK1, the latter assayed by increased phospho-myosin light chain. More recent studies established that the Rasip1-ARHGAP29 axis also regulates endothelial permeability( Post et al., 2013; Wilson et al., 2013) in cooperation with Radil, which binds directly to the RhoGAP( Post et al., 2015). Our finding that HEG1-binding defective Rasip1( ∆334-539), which did not reach EC junctions, failed to rescue both increased phospho-myosin and vascular permeability caused by Rasip1 silencing, supports the idea( Post et al., 2015) that the junctional localization of Rasip1 is important for regulation of RhoA/ROCK and vascular permeability. We noted( Figure 5E) that Rasip1( ∆334-539) appears to accumulate on vesicles, suggesting that Rasip1( ∆334-539) transport to the cell periphery is intact but that incorporation into cell-cell contacts depends on HEG1. In sum, our data show that Rasip1-HEG1 interaction is important for Rasip1 junctional localization and Rasip1-mediated regulation of ROCK signaling and endothelial barrier integrity. The studies reported here establish HEG1 as a physical nexus for Rap1 signaling in the vasculature. We found that Rasip1 and KRIT1 bound to HEG1 at distinct binding sites. Rap1 binding regulates the capacity of both proteins to target to HEG1-containing junctions, inhibit ROCK activity, and maintain junctional integrity( Figure 10). Importantly, the Rasip1 murine null phenotype is lethal by day E9. 5 with much more profound vascular collapse( Xu et al., 2011; Wilson et al., 2013) than is seen in the Heg1 null( Kleaveland et al., 2009) suggesting that Rasip1 has functions that do not depend on HEG1 binding or recruitment to EC junctions. One compelling possibility is that Rasip1 null mice exhibit profound detachment of EC from their basement membranes( Xu et al., 2011), perhaps due to inhibition of integrin activation. Indeed we observed that silencing HEG1 did not impair β1 integrin activation, whereas silencing Rasip1 did so. Thus, the Rap1-regulated interaction of HEG1 with Rasip1 localizes Rasip1 to those junctions and stabilizes them. Remarkably, Rap1 regulated association of KRIT1 with a second site in HEG1 also contributes to vascular development and integrity. Thus, HEG1 serves as physical nexus of Rap effectors that control both RhoA/ROCK signaling and vascular development. 10. 7554/eLife. 11394. 018Figure 10. HEG1 supports vascular integrity by binding multiple Rap1 effectors. In endothelial cells, Rap1 activation targets Rasip1 and KRIT1 to cell-cell junctions through a direct interaction with the transmembrane receptor HEG1. Here, Rasip1 and KRIT1 regulate junction stability and vascular integrity in part by suppressing ROCK signaling. DOI: http://dx. doi. org/10. 7554/eLife. 11394. 018", "role": "user" }, { "content": "Blood vessels are lined with cells known as vascular endothelial cells . These cells are connected to each other at junctions that consist of several different proteins . The junctions help to control how the blood vessel develops and provide a barrier that controls the movement of water and certain other molecules through the vessel wall . This barrier becomes weakened in diseases like sepsis and atherosclerosis . Two proteins that are essential for the heart and blood vessels to develop correctly are called “Heart of Glass” ( HEG1 ) and Rasip1 . Although a protein has been identified that binds to HEG1 at the cell junctions , this binding only involves a small region of HEG1 . This led de Kreuk , Gingras et al . to look for other proteins that interact with HEG1 and that might be important for controlling the development of the blood vessels . This revealed that HEG1 binds directly to Rasip1 . Further experiments revealed that HEG1 is essential for targeting Rasip1 to the junctions between the endothelial cells , and that this helps to stabilize the cell junctions . Mutant forms of Rasip1 that lacked a particular sequence in the middle of the protein were unable to bind to HEG1 and did not localize to the cell junctions . These studies open the door to future work to define how the interaction of Rasip1 and HEG1 is controlled and how Rasip1 stabilizes blood vessels .", "role": "assistant" } ]
science.chemdner_ner
science.chemdner_ner.2163
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\": [\"Arsenic\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nSystematic review of atomoxetine data in childhood and adolescent attention-deficit hyperactivity disorder 2009-2011: Focus on clinical efficacy and safety. Background:Clinicians obtain critical prescribing knowledge from clinical papers and review articles. This is the first published systematic review of clinical atomoxetine data covering 2009-2011.Objective:We aim to update clinicians on current clinical atomoxetine data with specific reference to time of onset of efficacy and maximal efficacy. These data may allow critical analysis of comparative efficacy between atomoxetine and stimulant medications.Methods:A formal systematic review of atomoxetine data from January 2009-June 2011 was conducted. The search term used was \"atomoxetine\" in the English language. The search yielded 747 citations from which 106 are clinical data. This paper includes clinical efficacy and safety data and excludes quality-of-life and review papers.Results:Atomoxetine has an onset of action within 4 weeks (possibly within 1 week in subsequent responders) but requires at least 12 weeks for full response to be demonstrated. Treatment-naïve cohorts (6-12 weeks) report effect sizes of 0.6-1.3. Using minimum 6-week clinical trial criteria, atomoxetine may demonstrate similar efficacy to methylphenidate comparing reduction in core ADHD symptoms in meta-analysis, although the diversity of the data makes interpretation complex. From epidemiological databases, cardiovascular and suicide-related events were similar to those seen in patients taking methylphenidate.Conclusions:Incremental response time to atomoxetine should be considered in the design of future comparative efficacy trials.", "role": "user" }, { "content": "{\"Chemical\": [\"methylphenidate\", \"atomoxetine\", \"Atomoxetine\"]}", "role": "assistant" } ]
science.multixscience_multidoc_summarization
science.multixscience_multidoc_summarization.8176
[ { "content": "You are provided with the abstract of a scientific paper and the abstracts of the papers it references. Your task is to write a Related Work section for the main paper. This section should concisely summarize and integrate the key points from the cited abstracts, reflecting how they relate to the main paper's contributions and context. In your summary, use numerical citation markers like [1], [2], etc., to refer to the corresponding papers.\n\nMain Abstract:\nIn this paper, we propose an efficient algorithm to directly restore a clear image from a hazy input. The proposed algorithm hinges on an end-to-end trainable neural network that consists of an encoder and a decoder. The encoder is exploited to capture the context of the derived input images, while the decoder is employed to estimate the contribution of each input to the final dehazed result using the learned representations attributed to the encoder. The constructed network adopts a novel fusion-based strategy which derives three inputs from an original hazy image by applying White Balance (WB), Contrast Enhancing (CE), and Gamma Correction (GC). We compute pixel-wise confidence maps based on the appearance differences between these different inputs to blend the information of the derived inputs and preserve the regions with pleasant visibility. The final dehazed image is yielded by gating the important features of the derived inputs. To train the network, we introduce a multi-scale approach such that the halo artifacts can be avoided. Extensive experimental results on both synthetic and real-world images demonstrate that the proposed algorithm performs favorably against the state-of-the-art algorithms.\n\nCited Abstract(s):\n[1]: We present a comprehensive study and evaluation of existing single image dehazing algorithms, using a new large-scale benchmark consisting of both synthetic and real-world hazy images, called REalistic Single Image DEhazing (RESIDE). RESIDE highlights diverse data sources and image contents, and is divided into five subsets, each serving different training or evaluation purposes. We further provide a rich variety of criteria for dehazing algorithm evaluation, ranging from full-reference metrics, to no-reference metrics, to subjective evaluation and the novel task-driven evaluation. Experiments on RESIDE shed light on the comparisons and limitations of state-of-the-art dehazing algorithms, and suggest promising future directions.\n\n[2]: We propose a referenceless perceptual fog density prediction model based on natural scene statistics (NSS) and fog aware statistical features. The proposed model, called Fog Aware Density Evaluator (FADE), predicts the visibility of a foggy scene from a single image without reference to a corresponding fog-free image, without dependence on salient objects in a scene, without side geographical camera information, without estimating a depth-dependent transmission map, and without training on human-rated judgments. FADE only makes use of measurable deviations from statistical regularities observed in natural foggy and fog-free images. Fog aware statistical features that define the perceptual fog density index derive from a space domain NSS model and the observed characteristics of foggy images. FADE not only predicts perceptual fog density for the entire image, but also provides a local fog density index for each patch. The predicted fog density using FADE correlates well with human judgments of fog density taken in a subjective study on a large foggy image database. As applications, FADE not only accurately assesses the performance of defogging algorithms designed to enhance the visibility of foggy images, but also is well suited for image defogging. A new FADE-based referenceless perceptual image defogging, dubbed DEnsity of Fog Assessment-based DEfogger (DEFADE) achieves better results for darker, denser foggy images as well as on standard foggy images than the state of the art defogging methods. A software release of FADE and DEFADE is available online for public use: http: live.ece.utexas.edu research fog index.html .", "role": "user" }, { "content": "Another line of research tries to make use of a fusion principle to restore hazy images in [1] [2] . However, these methods need complex blending based on luminance, chromatic and saliency maps. In contrast, we introduce a gated fusion based single image dehazing technique that blends only the derived three input images.", "role": "assistant" } ]
science.scientific_papers_summarization_single_doc_arxiv
science.scientific_papers_summarization_single_doc_arxiv.75066
[ { "content": "Write an abstract for the research paper below. Your response should only include the abstract and no additional text.\n\nArticle text:\nabrupt variation in the stratification of a star ( relative to the scale of the inverse radial wavenumber of a seismic mode of oscillation ) , such as that resulting from the ( smooth , albeit acoustically relatively abrupt ) depression in the first adiabatic exponent @xmath0 caused by the ionization of helium , where @xmath1 , @xmath2 and @xmath3 are pressure , density and specific entropy , or from the sharp transition from radiative to convective heat transport at the base of the convection zone , induces small - amplitude oscillatory components ( with respect to frequency ) in the spacing of the cyclic eigenfrequencies @xmath4 of seismic oscillation .\none might hope that the variation of the sound speed induced by helium ionization might enable one to determine from the low - degree eigenfrequencies a measure that is directly related to , perhaps even almost proportional to , the helium abundance , with little contamination from other properties of the structure of the star .\na convenient and easily evaluated measure of the oscillatory component is the second multiplet - frequency difference with respect to order @xmath5 amongst modes of like degree @xmath6 : @xmath7 ( gough , 1990 ) .\nany localized region of rapid variation of either the sound speed or the density scale height , or a spatial derivative of them , which here we call an acoustic glitch , induces an oscillatory component in @xmath8 with a ` cyclic frequency ' approximately equal to twice the acoustic depth @xmath9 of the glitch , and with an amplitude which depends on the amplitude of the glitch and which decays with @xmath10 once the inverse radial wavenumber of the mode becomes comparable with or greater than the radial extent of the glitch .\nwe report here on a model formula for the oscillatory contribution to the second frequency difference that is directly related to the gross properties of the glitch , and compare how well it fits actual eigenfrequency differences with corresponding fits of previously published formulae .\nfor a nonrotating star with vanishing pressure gradient at the surface , the angular eigenfrequencies @xmath11 of adiabatic oscillation obey a variational principle ( e.g. chandrasekhar 1963 ) which can be used to estimate the contribution @xmath12 to @xmath11 from the acoustic glitch . in estimating the contribution from helium ionization we retain only the dominant glitch term ( @xmath13 at fixed pressure and temperature ) ,\nand we represent @xmath13 by a pair of ( negative ) gaussian functions of @xmath14 with widths @xmath15 and @xmath16 , whose integrals are @xmath17 and @xmath18 , and which are centred about the acoustic depths @xmath19 and @xmath20 of the first and second ionization zones of helium beneath the seismic surface @xmath21 of the star . after converting to cyclic frequency @xmath22 , the oscillatory component is given asymptotically by @xmath23\\nonumber\\\\ & & \\times\\bigl(\\mu\\beta\\kappa_{\\rm i}^{-1 } { \\rm e}^{-8\\pi^2\\mu^2\\kappa^2_{\\rm i}\\delta_{\\rm ii}^2\\nu^2}\\cos2\\psi_{\\rm i}\\nonumber\\\\ & & { } \\;\\;\\;+\\kappa_{\\rm ii}^{-1}{\\rm e}^{-8\\pi^2\\kappa^2_{\\rm ii}\\delta_{\\rm ii}^2\\nu^2}\\cos2\\psi_{\\rm ii}\\bigr)\\,\\end{aligned}\\ ] ] ( houdek & gough , 2007 ) .\nwe evaluate the phases @xmath24 and @xmath25 , where @xmath26 , by representing the envelope by a plane - parallel polytrope of index @xmath273.5 and adding a phase constant @xmath28 to @xmath29 to account for the deviation of the actual envelope from the polytrope : @xmath30 in equation ( 3 ) , @xmath31 etc , with @xmath32^{1/2}$ ] ; also @xmath33 is a characteristic value of the first frequency difference @xmath34 whose value is half the inverse total acoustic radius @xmath35 of the star : @xmath36 .\nwe have found that the ratios @xmath37 , @xmath38 and @xmath39 hardly vary amongst stellar models whose masses and radii vary by factors of at least five .\nwe have accordingly adopted constant values , namely @xmath40 , @xmath41 and @xmath42 for solar - like stars . to obtain a complete description of @xmath8\nwe must add an oscillatory contribution from the near discontinuity in the density scale height at the base @xmath43 of the convection zone : @xmath44 \\right\\},\\end{aligned}\\ ] ] with @xmath45 ^{r_{\\rm c}{+}}_{r_{\\rm c}{-}}\\,,\\ ] ] where @xmath46 is the polytropic phase at the base of the convection zone , the phase constant @xmath47 being allowed to differ from @xmath28 to account for the fact that the region of the convection zone beneath the helium ionization zones is not an exact polytrope ; the quantity @xmath48 is a characteristic scale over which the sound speed in the radiative interior relaxes to a smooth extrapolation from the convection zone ( and which we also take to be constant , 80s , for all the stars we consider ) . from these expressions it is straightforward to evaluate the second difference @xmath8 , to which must be added a smooth term which we represent by a third - degree polynomial in @xmath49 : @xmath50 the eleven parameters @xmath51 are adjusted to fit by least squares the theoretical curve to the second frequency differences of the actual eigenfrequencies of the modes .\nwe compare the current seismic diagnostic ( 3)(7 ) with diagnostics published previously by monteiro & thompson ( 1998 , 2005 ) and by basu etal .\n( 2004 ) .\nthis diagnostic is based on a single - triangular approximation to @xmath13 due to he ionization .\nthe resulting formula for the frequency perturbation is @xmath52^{1/2 } \\;\\cos(4\\pi\\nu\\tau_{\\rm c}+2\\epsilon_{c})\\ , , \\ ] ] in which @xmath53 is the half - width and @xmath54 is proportional to the height of the triangle , from which second differences can be computed .\nwe set to zero the fitting parameter @xmath55 , a nonzero value of which , in the original formulation , would account for an putative adiabatic extension to the convection zone produced by overshooting ( which is not present in our solar models ) . by so doing , the seismic diagnostic ( 8) has the same number , 11 , of fitting parameters as the seismic diagnostic ( 3)(7 ) when adopting the third - degree polynomial ( 7 ) to represent the smooth contribution . in this formulation the seismic diagnostic\nis that used by basu ( 1997 ) , which was invented by basu , antia and narasimha ( 1994 ) for studying the base of the convection zone .\nballot et al .\n( 2004 ) and piau et al .\n( 2005 ) have adopted essentially the same formula .\nit is not directly linked to the form of the ionization - induced acoustic glitch , but instead was motivated by the oscillatory signature of a discontinuity in sound speed , whose amplitude was modified to make it decay with frequency approximately in the manner of the second differences of the eigenfrequencies .\nthe functional form is given by @xmath56 here we set @xmath57 , to remove the effect of the erroneous sound - speed discontinuity at the base of the convection zone and to render the functional form of the contribution at @xmath58 ( nearly ) consistent with those in the other two signatures ( 5 ) and ( 8) ; and we adopt the third - degree polynomial ( 7 ) to represent the smooth contribution , thereby using the same number ( 11 ) of fitting parameters as the seismic diagnostics ( 3)(7 ) and ( 8) , ( 7 ) .\ntwo sets of calibrated evolutionary models of the sun , computed by gough and novotny ( 1990 ) , were used .\none set of models has a constant value for the heavy - element abundance @xmath59 but with age varying in uniform steps of approximately 0.051 in @xmath60 , and the other set has constant age but varying heavy - element abundance in uniform steps of approximately 0.016 in @xmath61 ( see fig . 1 ) . but varying age ; the second set of the models , ( 5)(8 ) , have constant age but varying @xmath59 .\nall three diagnostics were fitted by the same numerical algorithm .\nthe lower panels of figs 24 show the results for the three seismic diagnostics ( 3)(7 ) , ( 8) and ( 7 ) , and ( 9 ) and ( 7 ) , fitted by least - squares to the adiabatically computed eigenfrequencies of the nine test models . plotted are the frequency - averaged amplitudes , @xmath62 and @xmath63 , of the two oscillatory seismic contributions associated with ionization and with the base of the convection zone , obtained by averaging the appropriate calibrated formula uniformly over the frequency range of the data plotted in the upper panels . also plotted are the values for the acoustic depths @xmath20 and @xmath43 .\ndepending on the seismic diagnostic , the values of either @xmath16 , @xmath53 or the phase constant @xmath28 are shown , and also the standard measure @xmath64 of the goodness of the fit between @xmath8 of the modelled frequencies and the calibrated seismic diagnostic . in all three cases the frequency - averaged amplitude @xmath62 increases with the helium abundance @xmath65 of the test models\n; moreover , the amplitudes @xmath66 and @xmath54 inferred from diagnostics ( 3)(7 ) and ( 8) , ( 7 ) respectively also increase with @xmath65 ( @xmath66 varying almost linearly in the @xmath59-varying sequence , @xmath54 varying almost linearly in the @xmath67-varying sequence ) , the latter result being apparently at variance with the findings of monteiro and thompson ( 2005 ) . the seismic diagnostic ( 3)(7 ) fits @xmath8 of the model eigenfrequencies the best , as is evinced by its smallest values of @xmath64 .\nthe quality of the fits is also illustrated in the upper panels of figs 24 , which show the second frequency differences over the whole frequency range considered for the central model 0 .\nit is also clear from comparing these figures that the seismic diagnostic ( 3)(7 ) reproduces @xmath8 of the model frequencies the most faithfully .\nit is interesting to note that in determining the parameters in the formulae the fitting procedure gives greater weight to the low - frequency end of the ionization - induced component @xmath68 , where the amplitude is the greatest , and greater weight to the high - frequency end of the contribution @xmath69 from the base of the convection zone , where misfits can not be offset by adjustments to @xmath68 ; therefore , the diagnostics ( 8) , ( 7 ) and ( 9 ) , ( 7 ) fail principally in the mid - range .\nsome large - scale discrepancy can be seen also at high frequency for formula ( 9 ) , ( 7 ) because @xmath68 decays with frequency too slowly .\nwe acknowledge that these discrepancies might be either smaller or hardly greater than the random errors in imminent asteroseismic data ; nevertheless , fitting an erroneous formula can lead to the inferences from that fitting being biased , and should therefore be avoided .\ngh is grateful for support from the particle physics and astronomy research council .\nballot j. , turck - chize , s. , garca r.a . , 2004 , a&a , 423 , 1051 basu s. , 1997 , mnras , 288 , 527 basu s. , antia h. , narasimha d. , 1994 , mnras , 267 , 209 basu s. , mazumdar a. , antia h.m . , demarque p. , 2004\n, mnras 350 , 277 chandrasekhar s. , 1963 , apj 138 , 896 gough d.o . , 1990 , in progress of seismology of the sun and stars , lecture notes in physics , vol .\n367 , y. osaki , h. shibahashi , eds .\nspringer verlag , p. 283\ngough d.o .\nnovotny , e. , 1990 , solar phys .\n, 128 , 143 houdek g. , gough d.o .\n2007 , mnras , in press monteiro m.j.p.f.g .\n, thompson m. , 1998 , in deubner f .-\n, christensen - dalsgaard j. , kurtz d. , eds .\niau symp.185 , new eyes to see inside the sun and stars .\nkluwer , dordrecht , p. 317\nmonteiro m.j.p.f.g .\n, thompson m. , 2005 , mnras , 361 , 1187 piau l. ballot j. turck - chize s. , 2005 , a&a , 430 , 571", "role": "user" }, { "content": "we present an improved model for an asteroseismic diagnostic contained in the frequency spacing of low - degree acoustic modes . by modelling in a realistic manner regions of rapid variation of dynamically relevant quantities , which we call acoustic glitches \n , we can derive signatures of the gross properties of those glitches . \n in particular , we are interested in measuring properties that are related to the helium ionization zones and to the rapid variation in the background state associated with the lower boundary of the convective envelope \n . the formula for the seismic diagnostic is tested against a sequence of theoretical models of the sun , and is compared with seismic diagnostics published previously by monteiro & thompson ( 1998 , 2005 ) and by basuetal . \n ( 2004 ) . \n [ 2001/04/25 1.1 ( pwd ) ]", "role": "assistant" } ]
science.acl_arc_intent_classification
science.acl_arc_intent_classification.1541
[ { "content": "You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories:\n- Background: provides context or foundational information related to the topic.\n- Extends: builds upon the cited work.\n- Uses: applies the methods or findings of the cited work.\n- Motivation: cites the work as inspiration or rationale for the research.\n- CompareOrContrast: compares or contrasts the cited work with others.\n- FutureWork: cites the work as a direction for future research.\n\nYour answer should be a single word from the following list of options: [\"Background\", \"Extends\", \"Uses\", \"Motivation\", \"CompareOrContrast\", \"FutureWork\"]. Do not include any other text in your response.\n\nSection Title:\nintroduction\n\nContext before the citation:\nSuch approaches have been tried recently in restricted cases (McCallum et al., 2000; Eisner, 2001b; Lafferty et al., 2001). The resulting machine must be normalized, either per-state or globally, to obtain a joint or a conditional distribution as desired. This allows meaningful parameter tying: if certain arcs such asu:i �—*, �—*, and a:ae o:e �—* share a contextual “vowel-fronting” feature, then their weights rise and fall together with the strength of that feature.\n\nCitation Sentence:\nSuch approaches have been tried recently in restricted cases ( McCallum et al. , 2000 ; Eisner , 2001b ; Lafferty et al. , 2001 ) .\n\nContext after the citation:\nNormalization may be postponed and applied instead to the result of combining the FST with other FSTs by composition, union, concatenation, etc. A simple example is a probabilistic FSA defined by normalizing the intersection of other probabilistic FSAs f1, f2,. . .. (This is in fact a log-linear model in which the component FSAs define the features: string x has log fi(x) occurrences of feature i.) In short, weighted finite-state operators provide a language for specifying a wide variety of parameterized statistical models. Let us turn to their training. \u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000", "role": "user" }, { "content": "Background", "role": "assistant" } ]
science.data_reco_mcq_sc
science.data_reco_mcq_sc.7984
[ { "content": "Given a short query with keyphrases about a research idea in 2018 in the domain of artificial intelligence, recommend the relevant datasets to validate the research idea from the options provided. A short description of the datasets is also provided to help you.\n\nQuery: I want to train a supervised model for saliency detection.\n\nKeyphrases: saliency detection images\n\nDataset description:\nFaceForensics: FaceForensics is a video dataset consisting of more than 500,000 frames containing faces from 1004 videos that can be used to study image or video forgeries. All videos are downloaded from Youtube and are cut down to short continuous clips that contain mostly frontal faces.\nCoSal2015: Cosal2015 is a large-scale dataset for co-saliency detection which consists of 2,015 images of 50 categories, and each group suffers from various challenging factors such as complex environments, occlusion issues, target appearance variations and background clutters, etc. All these increase the difficulty for accurate co-saliency detection.\nECSSD: The Extended Complex Scene Saliency Dataset (ECSSD) is comprised of complex scenes, presenting textures and structures common to real-world images. ECSSD contains 1,000 intricate images and respective ground-truth saliency maps, created as an average of the labeling of five human participants.\nSNIPS: The SNIPS Natural Language Understanding benchmark is a dataset of over 16,000 crowdsourced queries distributed among 7 user intents of various complexity: SearchCreativeWork (e.g. Find me the I, Robot television show), GetWeather (e.g.\n\nOptions: A: FaceForensics B: CoSal2015 C: ECSSD D: SNIPS", "role": "user" }, { "content": "ECSSD", "role": "assistant" } ]
science.bioasq_general_qa
science.bioasq_general_qa.142
[ { "content": "You will be shown a question, followed by excerpts from biomedical research papers. Please answer the question based on the provided context. Do not include any text in your response other than the answer.\n\nQuestion: Which event results in the acetylation of S6K1?\n\nContext: S6K1 is acetylated at lysine 516 in response to growth factor stimulation\nIn addition to phosphorylation, we have recently shown that S6K1 is also targeted by lysine acetylation\nHere, using tandem mass spectrometry we have mapped acetylation of S6K1 to lysine 516, a site close to the C-terminus of the kinase that is highly conserved amongst vertebrate S6K1 orthologues\nUsing acetyl-specific K516 antibodies, we show that acetylation of endogenous S6K1 at this site is potently induced upon growth factor stimulation\nWe propose that K516 acetylation may serve to modulate important kinase-independent functions of S6K1 in response to growth factor signalling\nHistone acetyltransferases interact with and acetylate p70 ribosomal S6 kinases in vitro and in vivo\nFollowing mitogen stimulation, S6Ks interact with the p300 and p300/CBP-associated factor (PCAF) acetyltransferases. S6Ks can be acetylated by p300 and PCAF in vitro and S6K acetylation is detected in cells expressing p300\nAcetylation of S6K1 and 2 is increased upon the inhibition of class I/II histone deacetylases (HDACs) by trichostatin-A, while the enhancement of S6K1 acetylation by nicotinamide suggests the additional involvement of sirtuin deacetylases in S6K deacetylation\nFollowing mitogen stimulation, S6Ks interact with the p300 and p300/CBP-associated factor (PCAF) acetyltransferases. S6Ks can be acetylated by p300 and PCAF in vitro and S6K acetylation is detected in cells expressing p300.\nFurthermore, it appears that the acetylation sites targeted by p300 lie within the divergent C-terminal regulatory domains of both S6K1 and S6K2.\nFurthermore, it appears that the acetylation sites targeted by p300 lie within the divergent C-terminal regulatory domains of both S6K1 and S6K2\nWe propose that K516 acetylation may serve to modulate important kinase-independent functions of S6K1 in response to growth factor signalling.", "role": "user" }, { "content": "Using acetyl-specific K516 antibodies, we show that acetylation of endogenous S6K1 at this site is potently induced upon growth factor stimulation. We propose that K516 acetylation may serve to modulate important kinase-independent functions of S6K1 in response to growth factor signalling. Following mitogen stimulation, S6Ks interact with the p300 and p300/CBP-associated factor (PCAF) acetyltransferases. S6Ks can be acetylated by p300 and PCAF in vitro and S6K acetylation is detected in cells expressing p300", "role": "assistant" } ]
science.qasper_extractive_qa
science.qasper_extractive_qa.614.2
[ { "content": "You will be shown sections from a scientific research paper, together with a question about the paper. This is an extractive question-answering task, where you must find and extract relevant text spans directly from the paper to answer the question. Your response should strictly be a `json` object with two fields:\n\n\"answer\": An array of strings extracted directly from the paper which, collectively, answer the question.\n\"evidence\": An array of strings. Each should be an excerpt from the paper, in which one or more of the extracted answers can be found.\n\nFor example, for the question \"What baselines did the authors compare against?\", a sample response might be:\n{\n \"answer\": [\"BERT\", \"RoBERTa\"],\n \"evidence\": [\"In our experiments, we compare the performance of our model against BERT and RoBERTa.\"]\n}\n\nDo not include any text in your response other than the json.\n\nIf the question is unanswerable given the provided excerpts, respond with the single word \"null\".\n\nPaper: Make Lead Bias in Your Favor: A Simple and Effective Method for News Summarization\n\nLead bias is a common phenomenon in news summarization, where early parts of an article often contain the most salient information. While many algorithms exploit this fact in summary generation, it has a detrimental effect on teaching the model to discriminate and extract important information. We propose that the lead bias can be leveraged in a simple and effective way in our favor to pretrain abstractive news summarization models on large-scale unlabeled corpus: predicting the leading sentences using the rest of an article. Via careful data cleaning and filtering, our transformer-based pretrained model without any finetuning achieves remarkable results over various news summarization tasks. With further finetuning, our model outperforms many competitive baseline models. Human evaluations further show the effectiveness of our method.\n\nRelated work ::: Document Summarization\nEnd-to-end abstractive text summarization has been intensively studied in recent literature. To generate summary tokens, most architectures take the encoder-decoder approach BIBREF8. BIBREF9 first introduces an attention-based seq2seq model to the abstractive sentence summarization task. However, its output summary degenerates as document length increases, and out-of-vocabulary (OOV) words cannot be efficiently handled. To tackle these challenges, BIBREF4 proposes a pointer-generator network that can both produce words from the vocabulary via a generator and copy words from the source article via a pointer. BIBREF10 utilizes reinforcement learning to improve the result. BIBREF11 uses a content selector to over-determine phrases in source documents that helps constrain the model to likely phrases. BIBREF12 adds Gaussian focal bias and a salience-selection network to the transformer encoder-decoder structure BIBREF13 for abstractive summarization. BIBREF14 randomly reshuffles the sentences in news articles to reduce the effect of lead bias in extractive summarization.\n\nRelated work ::: Pretraining\nIn recent years, pretraining language models have proved to be quite helpful in NLP tasks. The state-of-the-art pretrained models include ELMo BIBREF15, GPT BIBREF7, BERT BIBREF6 and UniLM BIBREF16. Built upon large-scale corpora, these pretrained models learn effective representations for various semantic structures and linguistic relationships. As a result, pretrained models have been widely used with considerable success in applications such as question answering BIBREF17, sentiment analysis BIBREF15 and passage reranking BIBREF18. Furthermore, UniLM BIBREF16 leverages its sequence-to-sequence capability for abstractive summarization; the BERT model has been employed as an encoder in BERTSUM BIBREF19 for extractive/abstractive summarization.\nCompared to our work, UniLM BIBREF16 is a general language model framework and does not take advantage of the special semantic structure of news articles. Similarly, BERTSUM BIBREF19 directly copies the pretrained BERT structure into its encoder and finetunes on labelled data instead of pretraining with the large quantity of unlabeled news corpus available. Recently, PEGASUS BIBREF20 leverages a similar idea of summarization pretraining, but they require finetuning with data from target domains, whereas our model has a remarkable performance without any finetuning.\n\nModel\nIn this section, we introduce our abstractive summarization model, which has a transformer-based encoder-decoder structure. We first formulate the supervised summarization problem and then present the network architecture.\n\nModel ::: Network Structure\nWe utilize a transformer-based encoder-decoder structure that maximizes the conditional probability of the summary: $P(Y|X, \\theta )$, where $\\theta $ represents the parameters.\n\nModel ::: Network Structure ::: Encoder\nThe encoder maps each token into a fixed-length vector using a trainable dictionary $\\mathcal {D}$ randomly initialized using a normal distribution with zero mean and a standard deviation of 0.02. Each transformer block conducts multi-head self-attention. And we use sinusoidal positional embedding in order to process arbitrarily long input. In the end, the output of the encoder is a set of contextualized vectors:\n\nModel ::: Network Structure ::: Decoder\nThe decoder is a transformer that generates the summary tokens one at a time, based on the input and previously generated summary tokens. Each token is projected onto a vector using the same dictionary $\\mathcal {D}$ as the encoder.\nThe decoder transformer block includes an additional cross-attention layer to fuse in information from the encoder. The output of the decoder transformer is denoted as:\nTo predict the next token $w_{k}$, we reuse the weights of dictionary $\\mathcal {D}$ as the final linear layer to decode $u^D_{k-1}$ into a probability distribution over the vocabulary: $P(w_k|w_{<k},u^E_{1:m})=( \\mathcal {D}u^D_{k-1})$.\nTraining. During training, we seek to minimize the cross-entropy loss:\nWe use teacher-forcing in decoder training, i.e. the decoder takes ground-truth summary tokens as input. The model has 10 layers of 8-headed transformer blocks in both its encoder and decoder, with 154.4M parameters.\nInference. During inference, we employ beam search to select the best candidate. The search starts with the special token $\\langle \\mbox{BEGIN}\\rangle $. We ignore any candidate word which results in duplicate trigrams. We select the summary with the highest average log-likelihood per token.\n\nExperiments ::: Implementation Details\nWe use SentencePiece BIBREF24 for tokenization, which segments any sentence into subwords. We train the SentencePiece model on pretrained data to generate a vocabulary of size 32K and of dimension 720. The vocabulary stays fixed during pretraining and finetuning.\nPretraining. We collect three years of online news articles from June 2016 to June 2019. We filter out articles overlapping with the evaluation data on media domain and time range. We then conduct several data cleaning strategies.\nFirst, many news articles begin with reporter names, media agencies, dates or other contents irrelevant to the content, e.g. “New York (CNN) –”, “Jones Smith, May 10th, 2018:”. We therefore apply simple regular expressions to remove these prefixes.\nSecond, to ensure that the summary is concise and the article contains enough salient information, we only keep articles with 10-150 words in the top three sentences and 150-1200 words in the rest, and that contain at least 6 sentences in total. In this way, we filter out i) articles with excessively long content to reduce memory consumption; ii) very short leading sentences with little information which are unlikely to be a good summary. To encourage the model to generate abstrative summaries, we also remove articles where any of the top three sentences is exactly repeated in the rest of the article.\nThird, we try to remove articles whose top three sentences may not form a relevant summary. For this purpose, we utilize a simple metric: overlapping words. We compute the portion of non-stopping words in the top three sentences that are also in the rest of an article. A higher portion implies that the summary is representative and has a higher chance of being inferred by the model using the rest of the article. To verify, we compute the overlapping ratio of non-stopping words between human-edited summary and the article in CNN/DailyMail dataset, which has a median value of 0.87. Therefore, in pretraining, we keep articles with an overlapping word ratio higher than 0.65.\nThese filters rule out around 95% of the raw data and we end up with 21.4M news articles, 12,000 of which are randomly sampled for validation.\nWe pretrain the model for 10 epochs and evaluate its performance on the validation set at the end of each epoch. The model with the highest ROUGE-L score is selected.\nDuring pretraining, we use a dropout rate of 0.3 for all inputs to transformer layers. The batch size is 1,920. We use RAdam BIBREF25 as the optimizer, with a learning rate of $10^{-4}$. Also, due to the different numerical scales of the positional embedding and initialized sentence piece embeddings, we divide the positional embedding by 100 before feeding it into the transformer. The beam width is set to 5 during inference.\nFinetuning. During finetuning, we keep the optimizer, learning rate and dropout rate unchanged as in pretraining. The batch size is 32 for all datasets. We pick the model with the highest ROUGE-L score on the validation set and report its performance on the test set. Our strategy of Pretraining with unlabeled Lead-3 summaries is called PL. We denote the pretrained model with finetuning on target datasets as PL-FT. The model with only pretraining and no finetuning is denoted as PL-NoFT, which is the same model for all datasets.\n\nExperiments ::: Baseline\nTo compare with our model, we select a number of strong summarization models as baseline systems. $\\textsc {Lead-X}$ uses the top $X$ sentences as a summary BIBREF19. The value of $X$ is 3 for NYT and CNN/DailyMail and 1 for XSum to accommodate the nature of summary length. $\\textsc {PTGen}$ BIBREF4 is the pointer-generator network. $\\textsc {DRM}$ BIBREF10 leverages deep reinforcement learning for summarization. $\\textsc {TConvS2S}$ BIBREF2 is based on convolutional neural networks. $\\textsc {BottomUp}$ BIBREF11 uses a bottom-up approach to generate summarization. ABS BIBREF26 uses neural attention for summary generation. DRGD BIBREF27 is based on a deep recurrent generative decoder. To compare with our pretrain-only model, we include several unsupervised abstractive baselines: SEQ$^3$ BIBREF28 employs the reconstruction loss and topic loss for summarization. BottleSum BIBREF23 leverages unsupervised extractive and self-supervised abstractive methods. GPT-2 BIBREF7 is a large-scaled pretrained language model which can be directly used to generate summaries.\n\nExperiments ::: Metrics\nWe employ the standard ROUGE-1, ROUGE-2 and ROUGE-L metrics BIBREF29 to evaluate all summarization models. These three metrics respectively evaluate the accuracy on unigrams, bigrams and longest common subsequence. ROUGE metrics have been shown to highly correlate with the human judgment BIBREF29. Following BIBREF22, BIBREF23, we use F-measure ROUGE on XSUM and CNN/DailyMail, and use limited-length recall-measure ROUGE on NYT and DUC. In NYT, the prediction is truncated to the length of the ground-truth summaries; in DUC, the prediction is truncated to 75 characters.\n\nExperiments ::: Results\nThe results are displayed in tab:nyt, tab:xsumresults, tab:cnndaily and tab:duc. As shown, on both NYT and XSum dataset, PL-FT outperforms all baseline models by a large margin. For instance, PL-FT obtains 3.2% higher ROUGE-1, 1.6% higher ROUGE-2 and 2.1% higher ROUGE-L scores than the best baseline model on XSum dataset. We conduct statistical test and found that the results are all significant with p-value smaller than 0.05 (marked by *) or 0.01 (marked by **), compared with previous best scores. On CNN/DailyMail dataset, PL-FT outperforms all baseline models except BottomUp BIBREF11. PL-NoFT, the pretrained model without any finetuning, also gets remarkable results. On XSum dataset, PL-NoFT is almost 8% higher than Lead-1 in ROUGE-1 and ROUGE-L. On CNN/DailyMail dataset, PL-NoFT significantly outperforms unsupervised models SEQ$^3$ and GPT-2, and even surpasses the supervised pointer-generator network. PL-NoFT also achieves state-of-the-art results on DUC-2003 and DUC-2004 among unsupervised models (except ROUGE-1 on DUC-2004), outperforming other carefully designed unsupervised summarization models. It's worth noting that PL-NoFT is the same model for all experiments, which proves that our pretrain strategy is effective across different news corpus.\n\nExperiments ::: Abstractiveness Analysis\nWe measure the abstractiveness of our model via the ratio of novel n-grams in summaries, i.e. the percentage of n-grams in the summary that are not present in the article. fig:novel shows this ratio in summaries from reference and generated by PL-NoFT and PL-FT in NYT dataset. Both PL-NoFT and PL-FT yield more novel 1-grams in summary than the reference. And PL-NoFT has similar novelty ratio with the reference in other n-gram categories. Also, we observe that the novelty ratio drops after finetuning. We attribute this to the strong lead bias in the NYT dataset which affects models trained on it.\n\nConclusions\nIn this paper, we propose a simple and effective pretraining method for news summarization. By employing the leading sentences from a news article as its target summary, we turn the problematic lead bias for news summarization in our favor. Based on this strategy, we conduct pretraining for abstractive summarization in a large-scale news corpus. We conduct thorough empirical tests on five benchmark news summarization datasets, including both automatic and human evaluations. Results show that the same pretrained model without any finetuning can achieve state-of-the-art results among unsupervised methods over various news summarization datasets. And finetuning on target domains can further improve the model's performance. We argue that this pretraining method can be applied in more scenarios where structural bias exists.\n\nFigure 1: Using Lead-3 summary as target in pretraining.\n\nFigure 2: Ratio of novel n-grams in summaries from reference, PL-NoFT and PL-FT models in NYT test set.\n\nTable 1: ROUGE recall scores on NYT test set.\n\nTable 2: ROUGE F1 results on XSum test set.\n\nTable 3: ROUGE F1 results on CNN/DailyMail test set.\n\nTable 4: Scoring criteria for human evaluation of summaries.\n\nTable 5: Average and standard deviations of human evaluation scores for summaries on CNN/DailyMail test set. Scores range from 1 to 5 with 5 being perfect. Each summary is judged by 3 human evaluators. PL-FT’s result is statistically significant compared with pointer-generator network with coverage with a p-value less than 10−7.\n\nQuestion: What did they pretrain the model on?", "role": "user" }, { "content": "{\"answer\": [\"hree years of online news articles from June 2016 to June 2019\"], \"evidence\": [\"Pretraining. We collect three years of online news articles from June 2016 to June 2019. We filter out articles overlapping with the evaluation data on media domain and time range. We then conduct several data cleaning strategies.\"]}", "role": "assistant" } ]
science.pico_ner
science.pico_ner.3507
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Participant\", \"Outcome\", \"Intervention\"].\n\nPlease return the output as a JSON object of the format: {\"Participant\": [\"patients with COPD\", ...], \"Intervention\": [\"Atenolol\", ...], \"Outcome\": [\"blood pressure\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nEffect of enzalutamide on health-related quality of life , pain , and skeletal-related events in asymptomatic and minimally symptomatic , chemotherapy-naive patients with metastatic castration-resistant prostate cancer ( PREVAIL ) : results from a randomised , phase 3 trial . BACKGROUND Enzalutamide significantly increased overall survival and radiographic progression-free survival compared with placebo in the PREVAIL trial of asymptomatic and minimally symptomatic , chemotherapy-naive patients with metastatic castration-resistant prostate cancer . We report the effect of enzalutamide on health-related quality of life ( HRQoL ) , pain , and skeletal-related events observed during this trial . METHODS In this phase 3 , double-blind trial , patients were randomly assigned ( 1:1 ) to receive enzalutamide 160 mg/day ( n=872 ) or placebo ( n=845 ) orally . HRQoL was assessed at baseline and during treatment using the Functional Assessment of Cancer Therapy-Prostate ( FACT-P ) and EQ-5D questionnaires . Pain status was assessed at screening , baseline , week 13 , and week 25 with the Brief Pain Inventory Short Form ( BPI-SF ) . The primary analysis of HRQoL data used a mixed-effects model to test the difference between least square means change from baseline at week 61 . We assessed change from baseline , percentage improvement , and time to deterioration in HRQoL and pain , the proportion of patients with a skeletal-related event , and time to first skeletal-related event . Analysis was done on the intention-to-treat population . This study is registered with ClinicalTrials.gov , number NCT01212991 . FINDINGS Median treatment duration was 16·6 months ( IQR 10·1-21·1 ) in the enzalutamide group and 4·6 months ( 2·8-9·7 ) in the placebo group . The mixed-effects model analyses showed significant treatment differences in change from baseline to week 61 with enzalutamide compared with placebo for most FACT-P endpoints and EQ-5D visual analogue scale . Median time to deterioration in FACT-P total score was 11·3 months ( 95 % CI 11·1-13·9 ) in the enzalutamide group and 5·6 months ( 5·5-5·6 ) in the placebo groups ( hazard ratio [ HR ] 0·62 [ 95 % CI 0·54-0·72 ] ; p < 0·0001 ) . A significantly greater proportion of patients in the enzalutamide group than in the placebo group reported clinically meaningful improvements in FACT-P total score ( 327 [ 40 % ] of 826 vs 181 [ 23 % ] of 790 ) , in EQ-5D utility index ( 224 [ 28 % ] of 812 vs 99 [ 16 % ] of 623 ) , and visual analogue scale ( 218 [ 27 % ] of 803 vs 106 of [ 18 % ] 603 ; all p < 0·0001 ) . Median time to progression in BPI-SF pain at its worst was 5·7 months ( 95 % CI 5·6-5·7 ) in the enzalutamide group and 5·6 months ( 5·4-5·6 ) in the placebo group ( HR 0·62 [ 95 % CI 0·53-0·74 ] ; p < 0·0001 ) . Progression of pain at its worst was less common in the enzalutamide group than in the placebo group at week 13 ( 220 [ 29 % ] of 769 vs 257 [ 42 % ] of 610 ; p < 0·0001 ) , but not at week 25 ( 225 [ 32 % ] of 705 vs 135 [ 38 % ] of 360 ; p=0·068 ) . 278 ( 32 % ) of 872 patients in the enzalutamide group and 309 ( 37 % ) of 845 patients in the placebo group had experienced a skeletal-related event by data cutoff . Median time to first skeletal-related events in the enzalutamide group was 31·1 months ( 95 % CI 29·5-not reached ) and 31·3 months ( 95 % CI 23·9-not reached ) in the placebo group ( HR 0·72 [ 95 % CI 0·61-0·84 ] ; p < 0·0001 ) . INTERPRETATION In addition to improving overall survival relative to placebo , enzalutamide significantly improves patient-related outcomes and delays occurrence of first skeletal-related event in chemotherapy-naive men with metastatic castration-resistant prostate cancer . FUNDING Astellas Pharma and Medivation .", "role": "user" }, { "content": "{\"Intervention\": [\"enzalutamide 160 mg/day\", \"or placebo\", \"enzalutamide\", \"Enzalutamide\", \"placebo\"], \"Outcome\": [\"enzalutamide significantly improves patient-related outcomes and delays occurrence\", \"CI 0\\u00b753-0\\u00b774 ] ; p \\u003c 0\\u00b70001 )\", \"significant treatment differences\", \"placebo group had experienced\", \"[ 23 % ] of 790 ) , in EQ-5D\", \"health-related quality of life ( HRQoL )\", \"endpoints and EQ-5D visual analogue scale\", \"time to deterioration in HRQoL and pain\", \"603 ; all p \\u003c 0\\u00b70001 )\", \"pain\", \"health-related quality of life , pain , and skeletal-related events\", \"Median time to first\", \"Median treatment duration\", \"skeletal-related events\", \"percentage improvement\", \"Median time to deterioration in FACT-P total\", \"skeletal-related event by data cutoff\", \"proportion of patients with a skeletal-related event\", \"Pain status\", \"primary analysis of HRQoL data\", \"improvements in FACT-P\", \"overall survival\", \"change from baseline\", \"HRQoL\", \"INTERPRETATION In addition\", \"[ 16 % ] of 623 ) , and visual\", \"time to first skeletal-related event\", \"radiographic progression-free survival\", \"Brief Pain Inventory Short Form ( BPI-SF ) .\"], \"Participant\": [\"men\", \"with metastatic\", \"asymptomatic and minimally symptomatic\", \"asymptomatic and minimally symptomatic , chemotherapy-naive patients with metastatic castration-resistant prostate cancer ( PREVAIL\", \"chemotherapy-naive patients with metastatic castration-resistant prostate cancer\", \"or\", \"patients were randomly assigned ( 1:1 ) to\", \"occurrence of first skeletal-related event in chemotherapy-naive\"]}", "role": "assistant" } ]
science.scicite_classification
science.scicite_classification.3488
[ { "content": "You will be presented with a citation sentence from a research paper. Classify the intent behind this citation by choosing from one of the following categories:\n\n- Method: Cites the paper for its methodology or procedure.\n- Background: Cites the paper to provide background information.\n- Result: Cites the paper for its findings or results.\n\nYour answer should be a single word from the following list of options: [\"Method\", \"Background\", \"Result\"]. Do not include any other text in your response.\n\nCitation sentence:\n\nFrequent methylation for TIMP3 (Kang et al. 2006) and CDH13 (Suehiro et al. 2008) in the present study concurs with previously published results in endometrial cancer.", "role": "user" }, { "content": "Result", "role": "assistant" } ]
science.scientific_papers_summarization_single_doc_arxiv
science.scientific_papers_summarization_single_doc_arxiv.36084
[ { "content": "Write an abstract for the research paper below. Your response should only include the abstract and no additional text.\n\nArticle text:\nthe predicted ultra - high energy ( uhe ) neutrino fluxes from both hadronic processes in cosmic sources and interaction of high energy cosmic rays with the cosmic microwave background radiation are very low .\ntherefore extremely large detector volumes , on the order of 100 km@xmath0 or more , are needed to detect a significant amount of these neutrinos .\nthe idea of a large hybrid optical - radio - acoustic neutrino detector has been described and simulated in @xcite .\nthe density of detectors in such a possible future uhe neutrino telescope is dictated by the signal to noise behaviour with travelled distance of the produced signals .\nthe ice has optical attenuation lengths on the order of 100 m , enabling the construction and operation of the icecube @xcite detector .\nthe relatively short optical attenuation length makes building a detector much larger than icecube prohibitively expensive .\nin contrast , the attenuation lengths of both radio and acoustic waves are expected to be larger @xcite and may make construction of a larger detector feasible . the south pole acoustic test setup ( spats )\nhas been built and deployed to evaluate the acoustic attenuation length , background noise level , transient rates and sound speed in the south pole ice - cap in the 1 to 100 khz region so that the feasibility and specific design of an acoustic neutrino detection array could be assessed .\nacoustic waves are bent toward regions of lower propagation speed and the sound speed vertical profile dictates the refraction index and the resulting radius of curvature .\nan ultra - high energy neutrino interaction produces an acoustic emission disk that will be deformed more for larger sound speed gradients and so the direction reconstruction will be more difficult .\na good event vertex reconstruction allows accurate rejection of transient background .\nthe absolute level and spectral shape of the continuous background noise determine the threshold at which neutrino induced signals can be extracted from the background and thus set the lower energy threshold for a given detector configuration .\ntransient acoustic noise sources can be misidentified as possible neutrino candidates , therefore a study of transient sources and signal properties needs to be performed .\nthe acoustic signal undergoes a geometric 1/r attenuation and on top of that scattering and absorption . the overall acoustic attenuation influences detector design and hence cost .\nthe south pole acoustic test setup consists of four vertical strings that were deployed in the upper 500 meters of selected icecube holes @xcite to form a trapezoidal array , with inter - string distances from 125 to 543 m. each string has 7 acoustic stages .\nfigure [ fig:1 ] shows a schematic of the spats array and its in - ice and on - ice components .\nit also shows a schematic drawing of an acoustic stage comprised of a transmitter and sensor module .\nthe transmitter module consists of a steel pressure vessel that houses a high - voltage pulse generator board and a temperature or pressure sensor .\ntriggered hv pulses are sent to the transmitter , a ring - shaped piezo - ceramic element that is cast in epoxy for electrical insulation and positioned @xmath113 cm below the steel housing .\nazimuthally isotropic emission is the motivation for the use of ring shaped piezo - ceramics .\nthe actual emission directivity of such an element was measured in azimuthal and polar directions @xcite .\nthe sensor module has three channels , each 120@xmath2 apart in azimuth , to ensure good angular coverage .\na retrievable pinger was deployed in 10 water - filled icecube holes down to a depth of 500 m : 6 holes were pinged in december 2007-january 2008 and 4 more holes were pinged using an improved pinger design in december 2008-january 2009 . the emitter is a broad band omni - directional transmitter which has a transmission power of 149 db re ( @xmath3pa / v ) at 1 meter distance . upon receiving a trigger signal , a short ( @xmath4 ) hv pulse\nis sent to the piezo - ceramic transmitter that is suspended about 2 m below the housing .\na broadband acoustic pulse is then emitted .\nboth the spats array and the pinger are gps synchronized so that the arrival times of the pinger - pulses can be determined .\ndata was collected for all spats - sensor levels at 80 , 100 , 140 , 190 , 250 , 320 , 400 , 430 and 500 m depth and the pinger was pulsed at repetition rates of 1 , 8 or 10 hz . at the spats\ninstrumented depths , the pinger lowering was stopped for five minutes so that a signal could be recorded with every spats sensor at that pinger position . for the december 2008-january 2009 holes ,\nno stop was made at 80 , 100 , 140 and 430 m depth .\nthe sound speed analysis uses data from december 2007-january 2008 geometries where the pinger and sensor were at the same depth and 125 m apart .\nthe pinger emitter was situated in a column of water where no shear waves can propagate , nevertheless shear waves were generated at the water - ice boundary where mode conversion was expected . therefore transit times can be extracted from the data for both pressure and shear waves for many instrumented spats levels .\nthere is agreement with @xcite for the pressure wave measurement in the not fully compacted ice of the upper region ( firn ) .\nthe extracted shear wave speed is about half of the pressure wave speed , as expected @xcite .\na linear fit was made to the data in the deep and fully compacted ice between 250 and 500 m depth .\nwe find following results for the pressure and shear wave sound speeds ( @xmath5 and @xmath6 ) and their variation with depth ( gradient : @xmath7 and @xmath8 ) : @xmath9 the gradient for both pressure and shear waves is consistent with zero .\nboth sound speed measurements are performed with a better than 1@xmath10 precision , taking into account the errors on the horizontal distance , pinger and sensor depths , emission and arrival times . for more details on the spats\nsound speed analysis , see @xcite .\nthe noise is monitored in spats through a forced 200 khz read - out of all sensor channels for 0.5 s every hour .\nthe distribution of adc counts in each of the operational channels is gaussian and stable during the present observation time of more than 1 year with a typical deviation of the mean noise level of @xmath11 .\nthe spats sensors on strings a , b and c have been calibrated in liquid water at 0@xmath2c in the 10 to 80 khz frequency range prior to deployment @xcite .\nlab measurements have shown @xcite that the sensitivity of the spats sensor in air at atmospheric pressure increases by a factor of 1.5 @xmath12 0.2 when cooled down from 0@xmath2c to -50@xmath2c .\nwe use this value to estimate the sensitivity of the spats sensors after they were deployed in the cold antarctic ice .\na measurement of the sensitivity at room temperature as a function of static pressure was performed in a pressure vessel and the results indicate a change in sensitivity of @xmath13 in the 1 to 100 bar region @xcite . the noise level and fluctuations\nare high for all 4 strings in the firn region , where the transition from a snow / air mixture to compact bulk ice takes place .\nthis is consistent with the effect of a large sound speed gradient in that layer so that the surface noise is refracted back to the surface . in the fully compacted ice , below the firn\n, noise conditions are more stable and we can derive an average noise level below 10mpa in the relevant frequency range ( 10 to 50khz ) . for more details on the spats noise floor analysis , see @xcite . [ cols=\"^,^,^,^,^\",options=\"header \" , ] the spats detector has been operated in transient mode for 45 minutes of every hour since august 2008 .\nif the number of adc counts on any of the twelve monitored channels exceeds a certain level above noise , we record a 5ms window of data around the trigger on that channel .\nthe resulting trigger rate is stable and on the order of a few triggers every minute for each of the twelve monitored channels .\nmost of these events are gaussian noise events , where only one sample is outside the trigger boundaries .\nthe transient events are processed off - line and analysed for time - coincidence clustering .\nfigure [ transients ] shows the spatial distribution of a total of 4235 reconstructed transient events as detected by spats between 1 september 2008 and 23 april 2009 ( 4422 hours integrated lifetime ) .\nthe data shows clear and steady sources or hot spots that can be correlated with the refreezing process of the water in sub - surface caverns .\neach icecube drilling season , a `` rodriguez wel '' ( rw ) is used as a water reservoir .\nthe main source of transients is the 07 - 08 rw .\nthere is also steady detection of the 05 - 06/04 - 05 rw .\nthe 06/07 rw was a steady source until october 2008 .\nthe 08 - 09 rw has not yet been detected .\ntransient data - taking continued during the icecube 08 - 09 drilling season and the refreezing of 12 holes nearest to the spats array are audible whereas 7 of the farthest are not .\nno vertices were reconstructed deeper than 400 m. we present here 2 classes of attenuation analyses , all of which use the sensors on the frozen - in spats strings but with a different sound source : the retrievable pinger and the frozen - in spats transmitters .\ntable [ table_att ] gives an overview of the spats attenuation length studies with their respective uncertainties .\ndistance ) vs. distance)-fit for a single channel at 400 m depth.,width=283,height=170 ] the retrievable pinger was recorded simultaneously by all spats sensors for each of the 4 icecube holes in which it was deployed during the december 2008-january 2009 period .\nthe pinger holes were almost perfectly aligned relative to the spats array , making the single - channel analysis independent of polar and azimuthal sensitivity variation .\nthe energy of the signal , which dominates in the frequency range from 5 to 35 khz , was extracted both from the waveforms in the time domain ( 4 pinger holes ) and from the power spectrum ( 3 pinger holes ) .\nfigure [ singlefit ] shows an example of a single - level fit for the energy in the frequency domain ( single channel at 400 m depth ) .\nfor both analyses , the quoted value in table [ table_att ] is the mean of all fits with the standard deviation as error .\nthe result for the time domain analysis uses all levels as shown in fig .\n[ att - dist ] and points which have large error bar ( @xmath14 ) have been excluded ( 4/47 combinations ) .\nthere is no evidence for a depth - dependence of the attenuation length .\nthe complete spats inter - string set of april 2009 consists of all possible transmitter - sensor combinations excluding the shallowest transmitters .\nthe transmitters were fired at 25hz repetition rate and for each combination a total of 500 pulses was recorded simultaneously by the three channels of the sensor module .\nan averaged waveform can therefore be obtained and the pressure wave energies extracted .\nan averaged noise - waveform allows us to subtract the noise contribution .\nevery transmitter in spats can be detected by 3 sensors at the same depth and different distances ( strings ) , this means that only the unknown sensor sensitivities have to be included into the systematic error for a single - level attenuation length fit .\nwe have fitted all data of the transmitters that are detected at three different distances .\neach single fit does not constrain the attenuation length very well due to the large sensor - to - sensor variations in sensitivity . by combining all fits\n, we find a weighted mean and the standard deviation of the weighted mean as value for the attenuation length and its error ( table [ table_att ] ) .\na way to work around the missing calibration of the sensors and transmitters is to build ratios of amplitudes using two transmitter - sensor pairs . for isotropic sensors and transmitters , one such measurement should yield the attenuation length .\nto minimize the remaining variation of the sensitivity due to the varying polar angle , we limited the amplitude ratios to the levels at 250 , 320 and 400 m depth .\nthis ratio - method yields an attenuation length with a large systematic error due to unknown azimuthal and remaining polar sensor orientations , therefore only the best fit is quoted in table [ table_att ] .\nwe have presented the most recent results from the spats setup .\nboth pressure and shear wave speeds have been mapped versus depth in firn and bulk ice .\nthis is the first measurement of the pressure wave speed below 190 m depth in the bulk ice , and the first measurement of shear wave speed in the south pole ice .\nthe resulting vertical sound speed gradient for both pressure and shear waves is consistent with no refraction between 250 and 500 m depth .\nextrapolating sensitivities from laboratory calibrations gives a first estimation of the absolute noise level at depths larger than 200 m and indicates values below 10 mpa integrated over the 10 to 50 khz frequency range .\nthe in - ice transient rates are low , most of the transient events can be correlated with well - know anthropogenic sources .\nwe have presented an overview of the different spats attenuation length studies , the preliminary results show that the analyses favour attenuation lengths in the 200 - 350 m region .\nthe current dataset does not allow a distinction to be made between absorption- or scatter - dominated attenuation length and dedicated measurements are under consideration .\nwe are grateful for the support of the u.s . national science foundation and the hospitality of the nsf amundsen - scott south pole station .\nd. besson _ et al .\na * 21s1 * ( 2006 ) 259 .\nd. besson _ et al .\ninstr . and meth . a ( 2009 ) , doi:10.1016/j.nima.2009.03.047 .\na. achterberg _ et al .\n_ , astropart .\n* 26 * ( 2009 ) 155 .\ns. barwick _\net al . _ , j. glaciol . * 51 * ( 2005 ) 231 .\nprice , j. geophys.res . * 111 * ( 2006 ) b02201 .\nfisher , diploma thesis , humboldt - universitat zu berlin ( 2006 ) .\nj. g. weihaupt , geophys .\n28 * ( 4 ) * ( 1963 ) 582 .\nalbert , geophys .\n* 25 no 23 * ( 1998 ) 4257 .\nf. descamps _ et al .\n_ , _ nucl .\ninstr . and meth .\n_ a(2009 ) , doi:10.1016/j.nima.2009.03.062\n. t. karg _\net al . _ ,\nthese proceedings .\net al . _ ,\nnucl . instr . and meth . a ( 2009 ) , doi:10.1016/j.nima.2009.03.063 .", "role": "user" }, { "content": "the feasibility and specific design of an acoustic neutrino detection array at the south pole depend on the acoustic properties of the ice . \n the south pole acoustic test setup ( spats ) has been built to evaluate the acoustic characteristics of the ice in the 1 to 100 khz frequency range . \n the most recent results of spats are presented . \n spats , acoustic neutrino detection , acoustic ice properties", "role": "assistant" } ]
science.genia_ner
science.genia_ner.38
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"RNA\", \"CellType\", \"CellLine\", \"Protein\", \"DNA\"].\n\nPlease return the output as a JSON object of the format: {\"DNA\" : [\"TAL1\", ...], \"RNA\": [\"mRNAs\", ...], \"CellLine\": [\"T cells\", ...], \"CellType\": [\"monocytes\", ...], \"Protein\": [\"IL-2\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\ncDNA cloning of a NGFI-B / nur77-related transcription factor from an apoptotic human T cell line. A human T lymphoid cell line, PEER, dies by apoptosis in the presence of PMA and calcium ionophore. A new gene, TINUR, was cloned from apoptotic PEER cells. The expression of the TINUR gene is induced within 1 h after the cross-linking of the T cell Ag receptor complex. TINUR belongs to the NGFI-B / nur77 family of the steroid receptor superfamily and is an orphan receptor. TINUR binds to the same DNA sequence as NGFI-B / nur77. We also propose that the NGFI-B / nur77 family can be classified into two subtypes.", "role": "user" }, { "content": "{\"CellLine\": [\"PEER\", \"apoptotic human T cell line\", \"human T lymphoid cell line\", \"apoptotic PEER cells\"], \"CellType\": [], \"DNA\": [\"DNA sequence\", \"cDNA\", \"TINUR gene\", \"gene\", \"TINUR\"], \"Protein\": [\"NGFI - B / nur77 - related transcription factor\", \"NGFI - B / nur77\", \"orphan receptor\", \"T cell Ag receptor complex\", \"steroid receptor superfamily\", \"NGFI - B / nur77 family\"], \"RNA\": []}", "role": "assistant" } ]
science.bc7_litcovid_topic_classification
science.bc7_litcovid_topic_classification.24553
[ { "content": "You will be presented with a paragraph from a COVID-19 literature. Your task is to classify the topics of this piece of text by choosing from the following categories:\n\n- Epidemic Forecasting: The content relates to predictions about the spread of COVID-19.\n- Treatment: The paragraph discusses therapeutic methods or medication for COVID-19.\n- Prevention: The content covers measures to prevent the spread of COVID-19.\n- Mechanism: The paragraph explains the biological mechanisms of the virus.\n- Case Report: The content is a report of a COVID-19 case.\n- Transmission: The paragraph discusses how COVID-19 is transmitted.\n- Diagnosis: The content is about diagnosing COVID-19 in individuals.\n\nYour classification should be provided as an element within a JSON array. The array should be formatted with topics enclosed in double quotes and separated by commas, following the JSON format, for example: [\"Topic1\", \"Topic2\", ...]. Include all applicable categories within the array. \nYour output should be a valid JSON array, and no additional text should be included in your response.\n\nArticle:\n\nSARS-CoV-2 pneumonitis can quickly strike to incapacitate the lung, leading to severe disease and sometimes death. In this perspective, we suggest that vitamin D deficiency and the failure to activate the vitamin D receptor (VDR) can aggravate this respiratory syndrome by igniting a wounding response in stellate cells of the lung. The FDA-approved injectable vitamin D analog, paricalcitol, suppresses stellate cell-derived murine hepatic and pancreatic pro-inflammatory and pro-fibrotic changes. Therefore, we suggest a possible parallel program in the pulmonary stellate cells of COVID-19 patients and propose repurposing paricalcitol infusion therapy to restrain the COVID-19 cytokine storm. This proposed therapy could prove important to people of color who have higher COVID-19 mortality rates and lower vitamin D levels.", "role": "user" }, { "content": "[\"Treatment\", \"Mechanism\"]", "role": "assistant" } ]
science.anat_em_ner
science.anat_em_ner.534
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"AnatomicalEntity\"].\n\nPlease return the output as a JSON object of the format: {\"AnatomicalEntity\": [\"tumor cells\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be an empty JSON array \"[]\".\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\n[Study of fungus polysaccharides compounds (FPC) in inducing the apoptosis of liver cancer cell Bel-7402]. \nTo observe the influence of fungus polysaccharides compounds (FPC) in inducing human liver cancer cell Bel-7402 apoptosis in cell cultivating in vitro, the authors analyzed tumor inhibitive gene P53 expression in Bel-7402 apoptosis by applying double immuno-marker. The result showed that the multilevel of FPC could all apparently induce Bel-7402 apoptosis. With the enhancement of FPC concentration, the authors observed chromatin condensation in some phases companying with the characteristic apoptosis. In the meantime, it could also greatly reduce the G1 and S, with obviously dose-response relationship. The percentage of cell apoptosis increased with the enhancing of concentration. In the high-level group the authors found typical DNA ladder eletrophoresis stripe. The result showed that the mechanism of the FPC antineoplastic effect had an intimate relation with its induction to apoptosis and that the result of FPC inducing tumor cell apoptosis had the character of P53 independence.", "role": "user" }, { "content": "{\"AnatomicalEntity\": [\"liver cancer cell Bel-7402\", \"Bel-7402\", \"cell\", \"liver cancer cell Bel-7402]\", \"neoplastic\", \"tumor\", \"chromatin\", \"tumor cell\"]}", "role": "assistant" } ]
science.craftchem_ner
science.craftchem_ner.4533
[ { "content": "You will be presented with a sentence from biomedical research text. Extract all terms that correspond to chemical entities of biological interest (ChEBI).\nReturn your identified terms in a JSON array, formatted like this: [\"Entity1\", \"Entity2\", ...].\nIf no such entities are detected, respond by providing an empty JSON array [].\n\nBiomedical context:\nBar,20mum.", "role": "user" }, { "content": "[]", "role": "assistant" } ]
science.data_reco_mcq_sc
science.data_reco_mcq_sc.9294
[ { "content": "Given a short query with keyphrases about a research idea in 2015 in the domain of artificial intelligence, recommend the relevant datasets to validate the research idea from the options provided. A short description of the datasets is also provided to help you.\n\nQuery: We propose a method for image classification by transferring the discriminative information from other domains. Our approach is based\n\nKeyphrases: image classification images\n\nDataset description:\nSyn2Real: Syn2Real, a synthetic-to-real visual domain adaptation benchmark meant to encourage further development of robust domain transfer methods. The goal is to train a model on a synthetic \"source\" domain and then update it so that its performance improves on a real \"target\" domain, without using any target annotations.\nTableBank: To address the need for a standard open domain table benchmark dataset, the author propose a novel weak supervision approach to automatically create the TableBank, which is orders of magnitude larger than existing human labeled datasets for table analysis. Distinct from traditional weakly supervised training set, our approach can obtain not only large scale but also high quality training data.\nTHEODORE: Recent work about synthetic indoor datasets from perspective views has shown significant improvements of object detection results with Convolutional Neural Networks(CNNs). In this paper, we introduce THEODORE: a novel, large-scale indoor dataset containing 100,000 high- resolution diversified fisheye images with 14 classes.\nImageNet: The ImageNet dataset contains 14,197,122 annotated images according to the WordNet hierarchy. Since 2010 the dataset is used in the ImageNet Large Scale Visual Recognition Challenge (ILSVRC), a benchmark in image classification and object detection.\n\nOptions: A: Syn2Real B: TableBank C: THEODORE D: ImageNet", "role": "user" }, { "content": "ImageNet", "role": "assistant" } ]
science.chia_ner
science.chia_ner.1137
[ { "content": "You will be shown inclusion or exclusion criteria for a clinical trial. Given these criteria, your task is to extract all unique entities of the following types: [\"Person\", \"Condition\", \"Value\", \"ReferencePoint\", \"Temporal\", \"Qualifier\", \"Observation\", \"Visit\", \"Measurement\", \"Mood\", \"Device\", \"Scope\", \"Negation\", \"Drug\", \"Procedure\", \"Multiplier\"].\n\nPlease return the output as a JSON object of the format: {\"Condition\": [\"breast cancer\", ...], \"Device\": [\"ports\", ...], \"Drug\": [\"Atenolol\", ...], \"Measurement\": [\"blood pressure\", ...], \"Mood\": [\"suspected\", ...], \"Multiplier\": [\"more than two\", ...], \"Negation\": [\"no\", ...], \"Observation\": [\"history\", ...], \"Person\": [\"male\", ...], \"Procedure\": [\"surgery\", ...], \"Qualifier\": [\"severe\", ...], \"ReferencePoint\": [\"enrollment\", ...], \"Scope\": [\"infection\", ...], \"Temporal\": [\"one day\", ...], \"Value\": [\"50%\", ...], \"Visit\": [\"ICU\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nChildren, aged between one and 24 months. classified as (American Society of Anesthesiologists) ASA physical status I or II, undergoing TEE were enrolled in the study.", "role": "user" }, { "content": "{\"Condition\": [], \"Device\": [], \"Drug\": [], \"Measurement\": [\"American Society of Anesthesiologists\", \"ASA physical status\"], \"Mood\": [], \"Multiplier\": [], \"Negation\": [], \"Observation\": [], \"Person\": [\"Children\", \"aged\"], \"Procedure\": [\"TEE\"], \"Qualifier\": [], \"ReferencePoint\": [], \"Scope\": [], \"Temporal\": [], \"Value\": [\"between one and 24 months\", \"I or II\"], \"Visit\": []}", "role": "assistant" } ]
science.craftchem_ner
science.craftchem_ner.9706
[ { "content": "You will be presented with a sentence from biomedical research text. Extract all terms that correspond to chemical entities of biological interest (ChEBI).\nReturn your identified terms in a JSON array, formatted like this: [\"Entity1\", \"Entity2\", ...].\nIf no such entities are detected, respond by providing an empty JSON array [].\n\nBiomedical context:\n*Denotesasignificantchange(p<0.01)oftheaverageweightofthePygo2nullgroupcomparedwithboththePygowild-typeandheterozygousgroups.", "role": "user" }, { "content": "[]", "role": "assistant" } ]
science.mslr2022_ms2_multidoc_summarization
science.mslr2022_ms2_multidoc_summarization.12092
[ { "content": "You will be presented with a collection of medical randomized controlled trial (RCT) reports addressing a shared clinical question -- for instance, the efficacy of a medical intervention on a disease or condition. Given the titles and abstracts of these reports, your task is to generate the conclusion of the systematic review article synthesizing the key findings. Your conclusion should be a few sentences long. In your response, include only the conclusion and no other text. The RCT reports are below.\n\n\n\nTitle 1:\nValidation of intermediate end points in cancer research.\nAbstract 1:\nInvestigations using intermediate end points as cancer surrogates are quicker , smaller , and less expensive than studies that use malignancy as the end point . We present a strategy for determining whether a given biomarker is a valid intermediate end point between an exposure and incidence of cancer . C and i date intermediate end points may be selected from case series , ecologic studies , and animal experiments . Prospect i ve cohort and sometimes case-control studies may be used to quantify the intermediate end point-cancer association . The most appropriate measure of this association is the attributable proportion . The intermediate end point is a valid cancer surrogate if the attributable proportion is close to 1.0 , but not if it is close to 0 . Usually , the attributable proportion is close to neither 1.0 nor 0 ; in this case , valid surrogacy requires that the intermediate end point mediate an established exposure-cancer relation . This would in turn imply that the exposure effect would vanish if adjusted for the intermediate end point . We discuss the relative advantages of intervention and observational studies for the validation of intermediate end points . This validation strategy also may be applied to intermediate end points for adverse reproductive outcomes and chronic diseases other than cancer", "role": "user" }, { "content": "The use of biomarkers in phase I trials has increased over the period from 1991 to 2002 .\nTo date , biomarker utilization has made a limited and primarily supportive contribution to dose selection , the primary end point of phase I studies .", "role": "assistant" } ]
science.covid_deepset_qa
science.covid_deepset_qa.594
[ { "content": "You will be shown a paragraph from a biomedical research paper, followed by a question about the paper. Please extract text from the paragraph to answer the question. Do not include any text in your response other than the answer.\n\nParagraph: Etiology of Influenza-Like Illnesses from Sentinel Network Practitioners in Réunion Island, 2011-2012\n\nhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC5031398/\n\nSHA: f5ff89ebfdd0375d034c112c6c1c7e163fa69a0c\n\nAuthors: Brottet, Elise; Jaffar-Bandjee, Marie-Christine; Li-Pat-Yuen, Ghislaine; Filleul, Laurent\nDate: 2016-09-21\nDOI: 10.1371/journal.pone.0163377\nLicense: cc-by\n\nAbstract: In Réunion Island, despite an influenza surveillance established since 1996 by the sentinel general practitioner’s network, little is known about the etiology of Influenza like-illness (ILI) that differs from influenza viruses in a tropical area. We set up a retrospective study using nasal swabs collected by sentinel GPs from ILI patients in 2011 and 2012. A total of 250 swabs were randomly selected and analyzed by multiplex reverse transcriptase polymerase chain reaction (RT-PCR) including research of 18 viruses and 4 bacteria. We detected respiratory viruses in 169/222 (76.1%) samples, mostly rhinovirus (23.4%), influenza A virus (21.2%), influenza B virus (12.6%), coronavirus (4.9%) and Human metapneumovirus (3.6%). Nine swabs (5.3% of positive swabs) revealed co-infections with two viruses identified, among which six concerned co-infections with influenza viruses. We observed important seasonal differences, with circulation of Human Metapneumoviruses, RSV A and B and coronavirus only during summer; whereas parainfluenza viruses were identified only during winter. In conclusion, this study highlights a substantial circulation of multiple respiratory pathogens in Réunion Island throughout the year. It shows that ILI are not only attributable to influenza and underlines the need for biological surveillance. As the use of multiplex RT-PCR showed its efficacy, it is now used routinely in the surveillance of ILI. Text: Influenza like-illness (ILI) or acute respiratory infections can be caused by several types of respiratory viruses or bacteria in humans [1] . Influenza viruses, Respiratory Syncytial viruses (RSV) and Parainfluenza viruses are identified as major viruses mostly responsible for ILI and pneumonia in several studies [2] . However practitioners cannot diagnose the infection without a biological test confirmation. Unfortunately, these infections causes are identified in less than 50% [3] . Réunion Island, a French overseas territory with 850,000 inhabitants, is located in the southern hemisphere between Madagascar and Mauritius in the Indian Ocean (Latitude: 21°05.2920 S Longitude: 55°36.4380 E.). The island benefits from a healthcare system similar to mainland France and epidemiological surveillance has been developed by the regional office of the French Institute for Public Health Surveillance (Cire OI), based on the surveillance system of mainland France [4] . Influenza activity generally increases during austral winter, corresponding to summer in Europe [5] . Since 2011, influenza vaccination campaign in Reunion Island starts in April and the vaccine used corresponds to World Health Organization recommendations for the southern hemisphere. Since 1996, clinical and biological influenza surveillance has been based on a sentinel practitioner's network [6] . In 2014, this network was composed of 58 general practitioners (GPs) spread over the island and represented around 7% of all Réunion Island GPs. Nasal swabs are randomly collected all along the year and are tested by RT-PCR for influenza viruses . Among these surveillance samples, 40 to 50% are tested positive for influenza A virus, A(H1N1)pdm09 or B virus by the virological laboratory of the University Hospital Center of Réunion. Thus ILI samples tested negative for influenza are of unknown etiology. Several biological tools allow identifying respiratory pathogens from nasal swab. In recent years, multiplex reverse transcriptase polymerase chain reaction (RT-PCR) has been developed to identify several viruses simultaneously [7] [8] [9] [10] . We therefore used this new method to set up a retrospective study using swabs collected by sentinel GPs from 2011 to 2012. The main objective of our study was to characterize respiratory pathogens responsible for ILI consultations in sentinel GPs in 2011 and 2012. Secondary objectives were to highlight seasonal trends on respiratory pathogens circulation and to describe occurrence of co-infections, especially during the flu season. ILI was defined as a sudden onset of fever more than 38 degrees Celsius and cough, associated or not with other symptoms such as breathing difficulty, headache, etc. Every week, all GPs of the sentinel network were encouraged to collect a nasal swab from the first two patients who presented ILI since less than three days. After being tested for influenza viruses, the 994 swabs collected in 2011 and 2012 are frozen at -80°C at the university hospital center (CHU) laboratory. Based on the budget, a season-stratified sample of 250 swabs was randomly selected in order to describe circulating viruses including outside flu season. Random sampling was performed with Excel 1 using the anonymized surveillance database of the Cire OI. The sampling frame contained identification number of swab assigned by Cire OI, laboratory identification number, sex, age, date of onset of symptoms, date of swab collection and result of influenza RT-PCR. We used Respifinder 1 Smart 22 kits a multiplex RT-PCR (PathoFinder, Maastricht, The Netherlands) which can detect 22 respiratory pathogens. This assay is based on the multiplex ligation-dependent probe amplification (MLPA) technology. The reverse transcription and preamplification steps were performed on the epgradient Mastercycler 1 (Eppendorf) and the hybridization, ligation and detection steps on the LightCycler 1 480 system (Roche Applied Science). This method was chosen because of its high specificity, compared to other same methods (78% versus 33%) [3, 11] . Multiplex analysis allows for rapid production of diagnostic results. It thus allows highlighted the possible presence of eighteen respiratory viruses and four bacteria in one reaction by melt curve analysis: Influenza A not (H1N1 \n\nStatistical analyses were performed with Stata 1 and Excel 1 . Two seasons were defined to identify possible seasonal trends in circulation of the viruses: winter season during weeks 23 to 39 between June and September and summer season during the rest of the year. Data and swabs result from a surveillance system that received regulatory approvals, including the CNIL (National Commission for Information Technology and Civil Liberties Number 1592205) approval in July 2012. All the patients have received oral information and gave their consent for swab and data collection. Data were collected for surveillance purpose and are totally anonymous. Among the 250 randomly-selected swabs, 26 were not available anymore as they were sent to Influenza Reference Center for confirmation and characterization of the pathogenic agent. According to the sensitivity of the assay two samples could be discordant results between Influenza PCR initially realized and Multiplex PCR. Thus they were deleted from the analysis: one is positive for Influenza in singleplex and negative for all tested pathogens in multiplex and one is positive for Influenza in singleplex and positive for PIV2 in multiplex. In total, 222 analyses were considered. Moreover, 53 samples were negative for all analyzed respiratory pathogens (23.9%) and 169 samples had at least one detected pathogen (76.1%), finally a total of 178 pathogens was identified. During the study period, a minority of the weeks (21 i.e. 20%) did not include any sampled swab, mainly outside flu season. Patients' sex-ratio was 0.63 (86 men and 136 women) and mean age was 28.4 years [min 0; max 81]. Ten percent had less than 5 years, 24% 5-15 years, 63% 15-65 years and only 3% were 65 and older. The respiratory pathogens most frequently identified in ILI swabs were rhinovirus (23.4%), influenza A not H1N1 (21.2%) and influenza B (12.6%) ( Table 1) . Among the 22 respiratory pathogens tested by the multiplex, only three were not found in any analyzed sample: Parainfluenza3, Legionella pneumophila and Bordetella pertussis. Regarding co-infections, nine swabs revealed the presence of two viruses, among which6 involved influenza viruses (Table 2) . Analyses showed that some viruses are possibly seasonal and were circulating during a specific period of the year. They are detected only in summer for Human Metapneumovirus, RSV A and B, and influenza A(H1N1)pdm09. For the latter, it is specific to the studied period since the influenza A(H1N1)pdm09 virus reappeared in Réunion Island in October 2012 and was no longer circulating since late 2010. On the opposite, Parainfluenza 1,2 and 4 viruses were identified only in winter. For other pathogens, no specific period of detection was observed. A weekly description of samples was realized to study the distribution of respiratory pathogens in 2011 and 2012 (Fig 1) . Results of biological analyses were compared with data of ILI consultations declared by sentinel GPs in 2011 and 2012. We observed in 2011, after a first wave in June mainly due to influenza A not H1N1 virus, a second wave of ILI consultations with mainly identification of Parainfluenza viruses and not influenza viruses. In 2012, the second epidemic wave at the end of austral winter coincided with Influenza viruses and Rhinovirus circulation. Regarding negative swabs (Fig 2) , we observed no seasonality during the study period with a similar proportion whatever the season. This retrospective study based on a sentinel GPs network showed that not only influenza viruses are responsible for ILI consultations. Indeed, an important circulation of multiple pathogens was observed throughout the year, with 12 different types of pathogens identified in 2011 and 2012. Respiratory viral pathogens were present in 76.1% of samples, which is largely above results from annual influenza surveillance [12] . After influenza viruses, Rhinovirus and Coronavirus were the most common respiratory viruses in Réunion Island. Although samples were not taken every week, sample was representative of ILI activity and consistent with flu season. Nevertheless, according to the low number of samples, it is difficult to conclude about seasonality. However in our study, RSV was circulating in summer season which is hot and rainy, which is confirmed by other studies in tropical region [13] . This study also highlighted several co-infections, showing that concomitant the multiple etiology of ILI. Co-circulation was already observed in Réunion Island during the A(H1N1) pdm09 pandemic in addition to influenza virus, with identification of other respiratory viruses such as Rhinovirus or Coronavirus [14] . In mainland France, during this pandemic, circulation of major respiratory viruses was found, such as Rhinovirus, Parainfluenza, Coronavirus, Human Metapneumovirus, like in our publication [15] [16] . In our study, only 5.3% of positive swabs were co-infections whereas in two studies in Madagascar co-infections represented 27.3% and 29.4% [17] [18] . Despite the distance of 9,300 km between Réunion and France, the island is directly connected to Europe with four daily flights to France. These exchanges can impact respiratory pathogens circulation in southern and northern hemisphere. Results of this study can therefore be of interest to both Indian Ocean and Europe countries. Among the 148 swabs initially negative for influenza because not previously tested for any other viruses, the study found an etiology for 95 swabs. In total, only 53 swabs, representing 24% of the sample, remained without etiology with negative multiplex PCR results all along the year. Multiple hypotheses can explain this result: a poor quality of swabs, preventing from identifying a pathogen, noninfectious causes or other pathogens not included in the multiplex PCR. However, we couldn't test the negative swabs for RNAse P, a marker of human cells, which could provide a modicum of assurance that the swab contained human cells. Concerning the two samples divergent for influenza identification between the multiplex and singleplex PCR, we discarded them for the analysis; one was positive in Influenza with singleplex and positive in PIV with multiplex. It could be a false positive result from singleplex. Indeed, as the multiplex PCR assay has a good sensitivity and is considered as a gold-standard, we decided to keep seven negative results for Influenza in singleplex and positive in Influenza in multiplex [7] [8] [9] [10] . No case of Bordetella pertussis which causes whooping cough and Legionella pneumophila which causes Legionnaires' disease was identified in this study. However, these diseases are rare in Réunion Island, around three cases of Legionnaires' disease are declared each year. A limit of the study is that no clinical data were available in the virological surveillance system of influenza in Réunion Island. It was impossible to compare clinical symptoms according to each pathogen and to know if there are different pathogens which cause for instance rhinitis, laryngitis or bronchitis (diseases included in ILI). A specific prospective study including clinical data might provide useful elements in the semiotics of diseases. In conclusion, this study highlighted an important circulation of multiple pathogens in Réunion Island throughout the year. It shows that ILI is not specific to influenza and so it is essential to have biological results in order to establish the differential diagnosis and thus explain the etiology of symptoms. For a better understanding of respiratory pathogens circulating in Réunion Island, information from this study may also be useful to practitioners who see many patients in consultation with ILI. As the use of multiplex RT-PCR showed its efficacy in the ILI surveillance and allowed to highlight the circulation of other viruses and bacterial causes of respiratory infections, it is now used routinely in the surveillance of ILI. Moreover, it would be interesting to repeat this study every 3 or 5 years adding clinical data to monitor the evolution of respiratory pathogens in Réunion Island over time.\n\nQuestion: How are the influenza tests carried out?", "role": "user" }, { "content": "Nasal swabs are randomly collected all along the year and are tested by RT-PCR for influenza viruses", "role": "assistant" } ]
science.scitldr_aic
science.scitldr_aic.1161
[ { "content": "Below are the Abstract, Introduction, and Conclusion sections of a computer science research paper. Please summarize the main contribution of the work in a single sentence. Your response should include the summary sentence and no additional text.\n\nPaper text:\n\nWe review three limitations of BLEU and ROUGE – the most popular metrics\n\nused to assess reference summaries against hypothesis summaries, come up with\n\ncriteria for what a good metric should behave like and propose concrete ways to\n\nassess the performance of a metric in detail and show the potential of Transformers-based Language Models to assess reference summaries against hypothesis summaries.\nEvaluation metrics play a central role in the machine learning community.\nThey direct the efforts of the research community and are used to define the state of the art models.\nIn machine translation and summarization, the two most common metrics used for evaluating similarity between candidate and reference texts are BLEU [Papineni et al., 2002] and ROUGE [Lin, 2004] .\nBoth approaches rely on counting the matching n-grams in the candidates summary to n-grams in the reference text.\nBLEU is precision focused while ROUGE is recall focused.\nThese metrics have posed serious limitations and have already been criticized by the academic community [Reiter, 2018] [Callison-Burch et al., 2006] [Sulem et al., 2018] [Novikova et al., 2017] .\nIn this work, we formulate an empirical criticism of BLEU and ROUGE, establish a criteria that a sound evaluation metric should have and propose concrete ways to test any metric towards these criteria.\nWe also use recent advances in NLP to design a data-driven metric addressing the weaknesses found in BLEU and ROUGE and scoring high on the criteria for a sound evaluation metric.\n2 Related Work 2.1 BLEU, ROUGE and n-gram matching approaches BLEU (Bilingual Evaluation Understudy) [Papineni et al., 2002] and ROUGE (Recall-Oriented Understudy for Gisting Evaluation) [Lin, 2004] have been used to evaluate many NLP tasks for almost two decades.\nThe general acceptance of these methods depend on many factors including their simplicity and the intuitive interpretability.\nYet the main factor is the claim that they highly correlate with human judgement [Papineni et al., 2002] .\nThis has been criticised extensively by the literature and the shortcomings of these methods have been widely studied.\nReiter [Reiter, 2018] , in his structured review of BLEU, finds a low correlation between BLEU and human judgment.\nCallison et al [Callison-Burch et al., 2006] examines BLEU in the context of machine translation and find that BLEU does neither correlate with human judgment on adequacy(whether the hypothesis sentence adequately captures the meaning of the reference sentence) nor fluency(the quality of language in a sentence).\nSulem et al [Sulem et al., 2018] examines BLEU in the context of text simplification on grammaticality, meaning preservation and simplicity and report BLEU has very low or in some cases negative correlation with human judgment.\nIn this work, we have established a framework to assess metrics comparing the quality of reference and hypothesis summary/translations.\nBased on these criteria, we compare evaluators using recent Transformers to BLEU and ROUGE and highlight their potential to replace BLEU and ROUGE.", "role": "user" }, { "content": "New method for assessing the quaility of similarity evaluators and showing potential of Transformer-based language models in replacing BLEU and ROUGE.", "role": "assistant" } ]
science.bioasq_factoid_qa
science.bioasq_factoid_qa.396
[ { "content": "You will be shown biomedical passages and your task is to answer the provided question. Provide a concise and accurate response and do not include any additional text in the response other than the answer.\n\nQuestion: Mention the only available genomics and developmental transcriptomics resource for the urochordate Oikopleura dioica\n\nContext: OikoBase: a genomics and developmental transcriptomics resource for the urochordate Oikopleura dioica.\nWe report the development of OikoBase (http://oikoarrays.biology.uiowa.edu/Oiko/), a tiling array-based genome browser resource for Oikopleura dioica, a metazoan belonging to the urochordates, the closest extant group to vertebrates. OikoBase facilitates retrieval and mining of a variety of useful genomics information. First, it includes a genome browser which interrogates 1260 genomic sequence scaffolds and features gene, transcript and CDS annotation tracks. Second, we annotated gene models with gene ontology (GO) terms and InterPro domains which are directly accessible in the browser with links to their entries in the GO (http://www.geneontology.org/) and InterPro (http://www.ebi.ac.uk/interpro/) databases, and we provide transcript and peptide links for sequence downloads. Third, we introduce the transcriptomics of a comprehensive set of developmental stages of O. dioica at high resolution and provide downloadable gene expression data for all developmental stages. Fourth, we incorporate a BLAST tool to identify homologs of genes and proteins. Finally, we include a tutorial that describes how to use OikoBase as well as a link to detailed methods, explaining the data generation and analysis pipeline. OikoBase will provide a valuable resource for research in chordate development, genome evolution and plasticity and the molecular ecology of this important marine planktonic organism.\nWe report the development of OikoBase (http://oikoarrays.biology.uiowa.edu/Oiko/), a tiling array-based genome browser resource for Oikopleura dioica, a metazoan belonging to the urochordates, the closest extant group to vertebrates.\nOikoBase: a genomics and developmental transcriptomics resource for the urochordate Oikopleura dioica\nWe report the development of OikoBase (http://oikoarrays.biology.uiowa.edu/Oiko/), a tiling array-based genome browser resource for Oikopleura dioica, a metazoan belonging to the urochordates, the closest extant group to vertebrates\nWe report the development of OikoBase (http://oikoarrays.biology.uiowa.edu/Oiko/), a tiling array-based genome browser resource for Oikopleura dioica, a metazoan belonging to the urochordates, the closest extant group to vertebrates.", "role": "user" }, { "content": "OikoBase", "role": "assistant" } ]
science.chemsum_single_document_summarization
science.chemsum_single_document_summarization.15792
[ { "content": "You will be shown the title of a chemistry paper, together with sections excerpted from the body of the paper. The paper sections may include HTML tags like `<p>` and `</p>`; please ignore these. Your task is to write an abstract for the paper. Your response should include the abstract and no other text.\n\nTitle:\nApplication of a New Chiral Phosphepine to the Catalytic Asymmetric Synthesis of Highly Functionalized Cyclopentenes that Bear an Array of Heteroatom-Substituted Quaternary Stereocenters\n\nPaper sections:\nINTRODUCTION\n<p>Interest in the use of tertiary phosphines as nucleophilic catalysts,1 particularly for enantioselective processes,2 has increased substantially in recent years. One especially powerful method, initially reported by Lu, is a formal [3+2] cycloaddition reaction between electron-poor allenes (or alkynes) and olefins,3 which generates cyclopentenes that are useful as endpoints or as intermediates for further functionalization. 4 Zhang described the first asymmetric variant of this process catalyzed by a chiral phosphine,5 and we6 and others7 have subsequently expanded the scope of this valuable transformation (eq 1).</p><p>The presence of a heteroatom, rather than a hydrogen or a carbon, substituent on the olefin (A in eq 1) could be expected to have a significant impact on both reactivity and stereoselectivity in these [3+2] cycloadditions. Indeed, to the best of our know ledge, there have been few reports of the use of heteroatom-substituted olefins as partners in Lu annulations,8, 9 and there have been no descript ions of catalytic asymmetric processes that generate heteroatom-substituted quaternary stereocenters. 10 Nevertheless, the ability to accomplish bond construct ions of this type with good enantioselectivity 11 through the use of such olefins would substantially enhance the utility of these cyclopentene-forming reactions. In this report, we establish that this objective can be achieved with the aid of a new chiral phosphine catalyst (1) (eq 2).</p>\n\nRESULTS AND DISCUSSION\n<p>The most well-studied examples of phosphine-catalyzed [3+2] cycloadditions of electron-poor allenes/alkynes with heteroatom-substituted olefins involve nitrogen substituents.8 Consequently, as the starting point for our effort to develop the first such catalyticenantioselective annulations to generate heteroatom-bearing quaternary stereocenters, we chose to examine the reaction of a 5-methylenehydantoin8c,e (Table 1; product C can be transformed into a bioactive analogue of hydantocidin8e).</p><p>We surveyed a range of phosphines that have been useful in asymmetric catalysis,12 including nucleophilic catalysis. Several bidentate phosphines, such as DIPAMP,7g Ph-BPE, 13 and TangPhos,14 were ineffective, affording little or no ee (Table 1, entries 1–3), whereas phosphepines 2–415 provided somewhat promising enantioselectivity and yield (entries 4–6).</p><p>The addition of substituents to the 3,3′ positions of the 1,1′-binaphthyl framework has proved to be a valuable approach to enhancing the enantioselectivity afforded by an array of chiral catalysts.16 However, this strategy has not yet been exploited in the context of asymmetric nucleophilic catalysis by chiral phosphepines. We were therefore pleased to determine that replacement of the 3,3′ hydrogens of phosphepine 4 with phenyl groups leads to improved ee (Table 1, entry 7 vs. entry 6).17,18 An examination of the X-ray crystal structure of this new phosphepine catalyst reveals how the 3,3′ phenyl substituents extend the chiral environment of the binaphthyl unit in the direction of the nucleophilic phosphorus (Figure 1).</p><p>All studies to date of nucleophile-catalyzed enantioselective cycloadditions of allenes with olefins have focused primarily on the use of allenes that cannot undergo isomerization to 1,3-dienes (i.e., R=H in eq 1; the 1,3-dienes are unreactive toward cyclopentene formation).19,20 On the other hand, the ability to employ γ-substituted allenes as reaction partners would markedly increase the structural as well as the stereochemical (two stereocenters rather than one) diversity of the products generated through this powerful cycloaddition process.</p><p>We have established that the conditions described in entry 7 of Table 1 can in fact be applied directly to [3+2] annulations of a wide array of γ-substituted racemic allenes with a 5-methylene hydantoin, furnishing the desired highly functionalized cyclopentenes, which bear adjacent quaternary and tertiary stereocenters, in good stereoselectivity, regioselectivity, and yield (Table 2).21 With γ substituents that range in steric demand from Me to CH2 cyclopentyl, the cycloadditions proceed cleanly with excel lent enantioselectivity in the presence of 5% of new phosphepine 1 (entries 1–3); in the case of a bulky i-Pr group, an increased catalyst loading (10%) is required in order to obtain a high yield (92%), and the cyclopentene is formed with moderate ee (77%) but excellent regioselectivity (50:1; entry 4). Functionalized γ substituents are compatible with the reaction conditions, including a phthalimide-containing group (entries 5–7). These are stereoconvergent processes where in both enantiomers of the racemic allene are being converted by the chiral catalyst into the same stereoisomer of the product with good selectivity.</p><p>With the first method in hand for catalytic enantioselective [3+2] cycloadditions of allenes with a heteroatom-substituted olefin to produce quaternary stereo-centers, we turned our attention to determining the scope of this process with respect to other substituents, with the goal of providing a versatile approach. Indeed, we have established that not only 5-methylene hydantoin, but also other pnictogen-bearing olefins, are suitable cycloaddition partners in the presence of phosphepine catalyst 1 (Tables 3 and 4).22 Thus, an array of annulations of racemic γ-substituted allenes with nitrogen- and phosphorus-substituted olefins proceed in excel lent ee, regioselectivity, and diastereoselectivity (>20:1). We are not aware of a prior example of the use of a phosphorus-substituted alkene in such [3+2] cycloadditions (including non-asymmetric processes).</p><p>We have explored the possibility that the olefin coupling partner can bear other families of heteroatoms, e.g., chalcogens. In fact, C2-symmetric phosphepine 1 catalyzes [3+2] cyclo-additions of both oxygen- and sulfur-substituted olefins with γ-substituted allenes to generate highly functionalized cyclopentenes with good-to-excel lent stereoselectivity, regioselectivity (≥20:1), and yield (Tables 5 and 6). To the best of our know ledge, oxygen-substituted alkenes have not previously been employed in annulations of this type.</p><p>These catalytic asymmetric cycloadditions of heteroatom-substituted olefins are not limited to couplings with allenoates. For example, allenamides are also suitable partners (eq 3).23 As far as we are aware, allenamides have not been reported by others to serve as substrates in phosphine-catalyzed [3+2] reactions with olefins.</p><p>The products of these catalytic asymmetric [3+2] cycloadditions can be converted into other potentially useful compounds. For example, the phthalimide can be deprotected to reveal a quaternary amino-acid ester (eq 4),24 the thioether can be unmasked to a tertiary thiol (eq 5), and the double bond of the cyclopentene can be functionalized with high diastereoselectivity to afford a cyclopentane that bears four contiguous stereocenters (two tertiary and two quaternary; eq 6).</p><p>Detailed mechanistic experiments have not yet been described for catalytic asymmetric [3+2] cycloadditions of allenes with olefins.25 For the process catalyzed by phosphepine 1, specifically, the reaction illustrated in eq 7, the rate is dependent upon the concentration of the allene and the catalyst, but not the olefin. During the cycloaddition, the predominant phosphorus-containing species (>90% according to 31P NMR spectroscopy) is the phosphepine itself, not a covalent adduct (e.g., B in eq 1). Furthermore, at partial conversion, we observe modest enantiomeric enrichment of the unreacted allene (30% ee at 93% conversion); to the best of our know ledge, this is the first example of a kinetic resolution in a phosphine-catalyzed allene/olefin [3+2] cycloaddition. Finally, we do not detect a nonlinear effect: the ee of the product correlates linearly with the ee of the catalyst. All of these data can be reconciled with a mechanism in which the rate-determining step of the catalytic cycle is the addition of the phosphine to the allene.</p>\n\nCONCLUSIONS\n<p>Although there have been numerous reports of phosphine-catalyzed [3+2] cycloadditions of allenes with olefins, a variety of noteworthy challenges had not been addressed. For example, with respect to non-asymmetric reactions, no processes had been described that employ phosphorus- or oxygen-substituted olefins to provide heteroatom-substituted quaternary centers; with regard to enantioselective transformations, there had been no success with any heteroatom-bearing olefin to produce a quaternary carbon, and there had been only limited progress (R=Me in eq 1) with γ-substituted allenes. By synthesizing and applying a new chiral phosphepine (1), we have developed a versatile method that furnishes cyclopentenes with an array of heteroatom-substituted (N, P, O, and S) quaternary stereocenters in good ee, diastereoselectivity, regioselectivity, and yield. We have used catalyst 1 in stereoconvergent reactions of racemic γ-substituted allenes, which afford products with greater structural and stereochemical diversity as compared with unsubstituted allenes. Furthermore, for the first time, we have employed allenamides in phosphine-catalyzed [3+2] cyclodditions. The highly functionalized cyclopentenes that are generated in these catalytic asymmetric annulations are poised for transformation into a variety of other target molecules. Mechanistic studies, including a unique observation of a (modest) kinetic resolution of the racemic allene, are consistent with addition of phosphepine 1 to the allene being the turnover-limiting step of the catalytic cycle. This is the first report in which modification of the 3,3′ positions of the binaphthyl unit of a phosphepine has been exploited in the context of asymmetric nucleophilic catalysis, and we anticipate that an array of other processes may benefit from this approach.</p>", "role": "user" }, { "content": "Through the design and synthesis of a new chiral phosphepine, the first catalytic asymmetric method for the [3+2] cycloaddition of allenes with olefins has been developed that generates cyclopentenes that bear nitrogen-, phosphorus-, oxygen-, and sulfur-substituted quaternary stereocenters. A wide array of racemic \\xce\\xb3-substituted allenes can be employed in this stereoconvergent process, which occurs with good enantioselectivity, diastereoselectivity, regioselectivity, and yield. Mechanistic studies, including a unique observation of a (modest) kinetic resolution of a racemic allene, are consistent with addition of the phosphepine to the allene being the turnover-limiting step of the catalytic cycle.", "role": "assistant" } ]
science.scientific_lay_summarisation_plos_single_doc_summ
science.scientific_lay_summarisation_plos_single_doc_summ.11359
[ { "content": "You will be presented with the abstract, introduction, and discussion section from a biomedical research article. Your task is to create a summary that a layperson can understand, capturing the essence of the research article. Keep essential scientific terms, but ensure that the language remains clear and the concepts are explained in an uncomplicated manner.\n\nTitle: Tuft dendrites of pyramidal neurons operate as feedback-modulated functional subunits\nArticle:\nAbstract:\nDendrites of pyramidal cells exhibit complex morphologies and contain a variety of ionic conductances, which generate non-trivial integrative properties. Basal and proximal apical dendrites have been shown to function as independent computational subunits within a two-layer feedforward processing scheme. The outputs of the subunits are linearly summed and passed through a final non-linearity. It is an open question whether this mathematical abstraction can be applied to apical tuft dendrites as well. Using a detailed compartmental model of CA1 pyramidal neurons and a novel theoretical framework based on iso-response methods, we first show that somatic sub-threshold responses to brief synaptic inputs cannot be described by a two-layer feedforward model. Then, we relax the core assumption of subunit independence and introduce non-linear feedback from the output layer to the subunit inputs. We find that additive feedback alone explains the somatic responses to synaptic inputs to most of the branches in the apical tuft. Individual dendritic branches bidirectionally modulate the thresholds of their input-output curves without significantly changing the gains. In contrast to these findings for precisely timed inputs, we show that neuronal computations based on firing rates can be accurately described by purely feedforward two-layer models. Our findings support the view that dendrites of pyramidal neurons possess non-linear analog processing capabilities that critically depend on the location of synaptic inputs. The iso-response framework proposed in this computational study is highly efficient and could be directly applied to biological neurons.\nIntroduction:\nA key objective of theoretical neuroscience is the development of simplified neuron models that incorporate relevant features of neuronal function while abstracting away nonessential complexity. According to the classical point-neuron assumption, synaptic inputs to a neuron sum linearly at a single integrative node, the soma, where the resulting membrane potential is transformed non-linearly to generate the neuronal response[1, 2]. Although this simplified description might approximate the dynamics of certain neuron types, it is challenged for pyramidal cells that exhibit complex morphologies and spatially modulated distributions of ion channels( for recent reviews see[3, 4]). These characteristics generate regenerative events, such as Na+ or NMDA( N-methyl-D-aspartate) spikes, which are localized in specific branches or subtrees so that the neuron can no longer be described by a single voltage compartment. It has been previously proposed that individual dendritic branches act as independent functional subunits within a two-layer feedforward model[5–7]. The first layer represents the computations of the individual branches, i. e., the voltage responses to local synaptic inputs, which are linearly summed and passed through a final non-linearity representing the second layer. Experimental and simulation studies consolidated the two-layer model as a suitable abstraction of synaptic integration in basal and proximal apical dendrites[8–11]. Häusser and Mel[12] suggested that even the distal apical tuft may function as a separate two-layer model that feeds into a proximal two-layer model, but a rigorous test of this hypothesis is still missing. In support of it, uncaging experiments have shown that NMDA spikes elicited in the fine branches of tuft dendrites generate voltage transients that are largely confined to the same branch and strongly attenuated across branches[13]. However, NMDA spikes in different branches can sum to initiate an apical calcium spike that propagates back into the distal branches. The impact of this electrical coupling on the core assumption of the two-layer model—subunit independence—remains unclear. Related work[11] has shown that synaptic inputs to nearby branches in tuft dendrites cannot be described by a two-layer model if its integrative node is located at the soma. This observation hints at a partial breakdown of the functional independence of individual branches but might be attributed to an additional non-linearity between the site of the summation of the subunit outputs and the soma—the calcium spike initiation zone. Estimating this non-linearity poses a major problem for tests of the two-layer hypothesis. It requires that the result of the summation of the subunit outputs is known. But this information is rarely available because of the difficulty to record from the fine dendritic branches in the tuft. This obstacle can be bypassed by iso-response methods. For this approach inputs to a system are varied such that a chosen output measure stays constant. As shown by various examples in different neural systems, iso-response methods can readily be integrated into neurophysiological experiments( see[14]). All that is needed is a closed-loop setup with which measurements of the neural output variable can be used to tune the inputs. The stimuli that result in constant outputs define “iso-response curves” that are independent of the specific form of the system’s final non-linearity. An estimation of this non-linearity is not required. Here, we show that the shape of the subunit non-linearities of a two-layer model can be read off from only a few iso-response curves, and that the validity of a two-layer model can be assessed by additional test iso-response curves. We apply this theoretical framework to a detailed multi-compartment model of a CA1 pyramidal cell[6] and investigate the dendritic integration properties of different parts of the apical dendrites including proximal subtrees and the tuft. For this purpose, we analyze sub-threshold somatic membrane-potential responses to brief simultaneous synaptic inputs to pairs of dendritic branches. The two-layer model captures the somatic response to proximal inputs but fails for tuft dendrites. We generalize the two-layer model for the tuft, relax the core assumption of subunit independence, and incorporate additive and multiplicative feedback from a subsequent processing stage. This new model accurately predicts the somatic response to synaptic inputs on proximal apical branches as well as on tuft dendrites. The functional form and range of its non-linear feedback can be readily identified through the iso-response method.\nDiscussion:\nAdditive and multiplicative feedback terms correspond to horizontal shifts and horizontal rescalings of the subunit non-linearities, respectively. Various biophysical mechanisms have been associated with these operations( for a review see[16]). Shifts of dendritic non-linearities have been observed for additional current or conductance inputs in the proximity of a driving input[15, 17, 18]. In particular, it was shown for basal dendrites that distal modulatory input within the same branch mainly shifts the subunit non-linearities[15, 18]. In contrast, proximal modulatory input shifts and changes the slope of the subunit non-linearities. These asymmetric interactions have been linked to the voltage-dependence of NMDA channels and an increase in input resistance with increasing distance between the input and the soma. Our results indicate that a similar modulation occurs across dendritic branches in the tuft. Horizontal shifts of the subunit non-linearities alone are sufficient to explain the response properties of most of the dendritic branches. However, for a small fraction of branches a change in the slope of their subunit non-linearities is required as well to describe the synaptic integration properties. This modulation may be either attributed to a change in the gain of the subunit non-linearity, which changes its maximum value as well, or a horizontal rescaling, which does not change the maximum value. Horizontal rescaling has been observed for modulatory input co-localized with driving input[18], whereas gain control was reported for proximal modulatory input[15, 18]. Here, we cannot discriminate between both cases for all branch pairs that implement multiplicative feedback. The subunit non-linearities were not driven to saturation for somatic responses below the action potential threshold, and no conclusion can be drawn about the change of their maximum value. A major difference between the distal tuft and basal dendrites is the presence of hyperpolarization-activated cation currents( Ih) and voltage-gated calcium channels. In agreement with experiments[13] we find that Ih currents facilitate subunit independence in the dendritic tuft. Moreover, our simulation study predicts that feedback interactions in tuft dendrites are mediated by calcium currents at the calcium spike initiation zone. Previous work[11] applied the two-layer model to predict the firing rate of a detailed pyramidal cell model in response to homogeneous Poisson input. This feedforward two-layer model correctly predicted the somatic response to a distant pair of apical tuft dendrites but failed for synaptic inputs to nearby branches in the tuft( see Fig S4 in[11]). In contrast, our results show that computations based on firing rates can be well described by a feedforward two-layer model even for tuft dendrites. The failure of the two-layer feed-forward model in[11] to predict the firing-rate responses to synaptic inputs to nearby branches in the tuft can be attributed to either a disruption of the functional subunit independence or a distorted reconstruction of the model parameters. In[11], the final non-linearity of the two-layer model is approximated by the somatic f-I curve and the individual subunit outputs are assumed to contribute linearly to the total somatic current. If, under these conditions, the dendritic branches operate as independent functional subunits, their non-linearities can be estimated by a least-squares linear regression of the total somatic current on the synaptic inputs. However, for synaptic inputs to nearby distal branches, as shown in Fig S4C of[11], additional dendritic non-linearities occur between the location of the subunit output summation and the soma. In this case, the somatic current is no more a linear sum of the individual branch outputs and linear regression fails to correctly identify the dendritic non-linearities. As we find no evidence for a violation of the functional independence of dendritic branches in the tuft for computations based on firing rates our results suggest that the reduced performance of the two-layer model in[11] can be attributed to a distorted reconstruction of the final non-linearity. The iso-response method has two advantages when compared to the method proposed in[11]. First, it does not require an estimate of the final non-linearity of the two-layer model that is rarely available for synaptic integration in the tuft because of the difficulty to record from fine dendritic branches. Once the subunit non-linearities of a feedforward two-layer model are identified its final non-linearity can be reconstructed from measurements of the somatic responses to single branch stimulations of varying strength. Second, iso-response curves are optimally efficient in that they are based on the smallest amount of data needed to identify the subunit non-linearities. We have demonstrated that feedback in tuft dendrites impacts synaptic integration on the short time scales relevant for triggering single somatic spikes. The precise timing of spikes relative to hippocampal theta oscillations has been shown to carry information about the location of a rat within its environment[19, 20]. Moreover, recent work[21] supports the hypothesis that degraded spike timing in the hippocampus results in memory impairments. In contrast, we found no evidence for the functional relevance of feedback for mean somatic responses, i. e. spike counts within longer time intervals. This difference might be attributed to two reasons. First, the brief synaptic input to pairs of dendritic branches was injected simultaneously, whereas the Poisson spike trains used for rate inputs were not synchronized. Simultaneous synaptic stimulation has been shown to enhance the non-linear integrative properties of dendrites[22]. Second, for brief synaptic stimulation we investigated somatic sub-threshold responses in contrast to multiple spikes for rate responses. As proposed in[11], action potential generation might facilitate subunit independence. Overall, our findings consolidate the feedforward two-layer model as a suitable abstraction of neuronal computations based on firing rates[5–12] but not on precisely timed inputs. Other work has focused on non-linear interactions between tuft dendrites and the soma[23–27]. In contrast, we investigated dendritic integration on a finer spatial scale—within subtrees of the dendritic tuft. Notwithstanding, we observed for all but one proximal-distal branch pair that additive feedback was sufficient to describe the somatic response. In contrast, previous work found a multiplicative coupling between the soma and the tuft[25], i. e. distal depolarizing current injections increased the gain and reduced the rheobase of the firing rate-current response at the soma. However, the underlying mechanism relied on backpropagating action potentials that haven’t been triggered in our study. We have shown that feedback between the proximal apical dendrites and the tuft occurs also in the sub-threshold regime, and this coupling is additive in most cases. We hypothesize that this additive coupling between tuft dendrites serves as a gating mechanism similar to the proximal-distal coupling. Weakly stimulated dendritic branches in the tuft that would otherwise fail to detect individual input patterns with local membrane potentials sub-threshold to the non-linearities combine the collective evidence resulting in robust pattern recognition. We investigated dendritic integration properties of branches that each receive only a single synaptic input. Hence, the subunit non-linearities are one-dimensional functions of the synaptic input strength. It has been recently shown that pairs of synaptic inputs to the same branch in the basal dendrites are not purely summed but also processed non-linearly in a location dependent manner[15, 18]. Conceptually, the two-layer model proposed here can be augmented to incorporate this intra-branch integration effects. Moreover, the theoretical framework based on iso-response methods can be applied to identify all non-linearities involved. We focused on spatial integration properties of tuft dendrites, where pairs of synapses are activated simultaneously. A complete description of dendritic integration in pyramidal neurons has to incorporate the experimentally observed temporal coupling between the proximal dendritic compartment, including the basal dendrites, soma and apical obliques, and the distal tuft[23–27]. We expect that the two-layer model can be incorporated into this two-compartment schema. The theoretical framework presented here can be used to identify two-layer models without the necessity of additional information beyond the stimuli and the responses and without extensive datasets. Overall, our findings support the view that the information processing capabilities of pyramidal neurons depend sensitively on the spatial distribution of synaptic inputs. This conclusion has been reached using model neurons. The proposed iso-response techniques could, however, also be integrated into multi-electrode or photo-stimulation paradigms to directly reveal the dendritic computations of biological neurons.", "role": "user" }, { "content": "Pyramidal neurons are the principal cell type in the cerebral cortex . Revealing how these cells operate is key to understanding the dynamics and computations of cortical circuits . However , it is still a matter of debate how pyramidal neurons transform their synaptic inputs into spike outputs . Recent studies have proposed that individual dendritic branches or subtrees may function as independent computational subunits . Although experimental work consolidated this abstraction for basal and proximal apical dendrites , a rigorous test for tuft dendrites is still missing . By carrying out a computational study we demonstrate that dendritic branches in the tuft do not form independent subunits , however , their integrative properties can be captured by a model that incorporates modulatory feedback between these subunits . This conclusion has been reached using a novel theoretical framework that can be directly integrated into multi-electrode or photo-stimulation paradigms to reveal the dendritic computations of biological neurons .", "role": "assistant" } ]
science.cdr_ner
science.cdr_ner.148
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\", \"Disease\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\": [\"Naloxone\", ...], \"Disease\": [\"Hypotensive\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nOrganophosphate-induced convulsions and prevention of neuropathological damages. Such organophosphorus (OP) compounds as diisopropylfluorophosphate (DFP), sarin and soman are potent inhibitors of acetylcholinesterases (AChEs) and butyrylcholinesterases (BChEs). The acute toxicity of OPs is the result of their irreversible binding with AChEs in the central nervous system (CNS), which elevates acetylcholine (ACh) levels. The protective action of subcutaneously (SC) administered antidotes or their combinations in DFP (2.0 mg/kg BW) intoxication was studied in 9-10-weeks-old Han-Wistar male rats. The rats received AChE reactivator pralidoxime-2-chloride (2PAM) (30.0 mg/kg BW), anticonvulsant diazepam (2.0 mg/kg BW), A(1)-adenosine receptor agonist N(6)-cyclopentyl adenosine (CPA) (2.0 mg/kg BW), NMDA-receptor antagonist dizocilpine maleate (+-MK801 hydrogen maleate) (2.0 mg/kg BW) or their combinations with cholinolytic drug atropine sulfate (50.0 mg/kg BW) immediately or 30 min after the single SC injection of DFP. The control rats received atropine sulfate, but also saline and olive oil instead of other antidotes and DFP, respectively. All rats were terminated either 24 h or 3 weeks after the DFP injection. The rats treated with DFP-atropine showed severe typical OP-induced toxicity signs. When CPA, diazepam or 2PAM was given immediately after DFP-atropine, these treatments prevented, delayed or shortened the occurrence of serious signs of poisoning. Atropine-MK801 did not offer any additional protection against DFP toxicity. In conclusion, CPA, diazepam and 2PAM in combination with atropine prevented the occurrence of serious signs of poisoning and thus reduced the toxicity of DFP in rat.", "role": "user" }, { "content": "{\"Chemical\": [\"organophosphorus\", \"DFP\", \"ACh\", \"CPA\", \"NMDA\", \"Organophosphate\", \"OPs\", \"dizocilpine maleate\", \"adenosine\", \"Atropine\", \"atropine sulfate\", \"atropine\", \"2PAM\", \"N(6)-cyclopentyl adenosine\", \"diazepam\", \"soman\", \"diisopropylfluorophosphate\", \"OP\", \"pralidoxime-2-chloride\", \"MK801\", \"acetylcholine\", \"sarin\"], \"Disease\": [\"poisoning\", \"toxicity\", \"convulsions\", \"neuropathological damages\"]}", "role": "assistant" } ]
science.mslr2022_cochrane_multidoc_summarization
science.mslr2022_cochrane_multidoc_summarization.3551
[ { "content": "You will be presented with a collection of medical randomized controlled trial (RCT) reports addressing a shared clinical question -- for instance, the efficacy of a medical intervention on a disease or condition. Given the titles and abstracts of these reports, your task is to generate the conclusion of the systematic review article synthesizing the key findings. Your conclusion should be a few sentences long. In your response, include only the conclusion and no other text. The RCT reports are below.\n\n\n\nTitle 1:\nSchizophrenia--a follow-up study of results of treatment. I. Design and other problems.\nAbstract 1:\nThis is the first of a series of articles on a follow-up study of the results of treatment of schizophrenia, studied over a period of two to five years after first admission and first release. The study compares the follow-up outcome of five different treatment methods given to first-admission male and female schizophrenic patients in the hospital. The design of the study is used as a basis for description and discussion of the practical, ethical, and statistical problems involved. A distinction is made between follow-up and continued treatment design, and it is concluded that both pose massive problems in execution, analysis, presentation, and interpretation.\n\n\nTitle 2:\nEffects of psychotherapy in schizophrenia: II. Comparative outcome of two forms of treatment.\nAbstract 2:\nThis study compares the effects of exploratory, insight-oriented (EIO) and reality-adaptive, supportive (RAS) forms of psychotherapy on a sample of 95 schizophrenic patients. Analyses of 2-year outcomes revealed a complex interaction between the type of psychotherapy provided and the domain of psychopathology affected. RAS psychotherapy exerted clear preferential effects in the areas of recidivism and role performance. The EIO psychotherapy exerted preferential, albeit more modest, action in the areas of ego functioning and cognition. Overall, however, the magnitude of the differences was low. The results highlighted the need for more focused studies of subgroups, and of process and contextual influences on outcome.\n\n\nTitle 3:\nGroup vs individual psychotherapy with schizophrenics. A controlled outcome study.\nAbstract 3:\nnan", "role": "user" }, { "content": "Current data do not support the use of psychodynamic psychotherapy techniques for hospitalised people with schizophrenia. If psychoanalytic therapy is being used for people with schizophrenia there is an urgent need for trials.", "role": "assistant" } ]
science.craftchem_ner
science.craftchem_ner.9969
[ { "content": "You will be presented with a sentence from biomedical research text. Extract all terms that correspond to chemical entities of biological interest (ChEBI).\nReturn your identified terms in a JSON array, formatted like this: [\"Entity1\", \"Entity2\", ...].\nIf no such entities are detected, respond by providing an empty JSON array [].\n\nBiomedical context:\nThefollowingprimersweregeneratedtoincludethesequenceobtainedfromtheIlluminaprobe:Actb(TTGCTGACAGGATGCAGAAG,ACATCTGCTGGAAGGTGGAC);Aldh1a7(CCAGAAAGTGGTGTTTTGCT,GAGTTACAGAGAGCTTGCACCTT);Col8a1(GCAGACAGGCATCTTCACCT,TGTGTACATCATGGGCTCGT);Csrp1(CAGCATAGCCCAGGGTAGAG,TGGGCAAGGTAGTGAAGGTT);Klk5(GCAGCATTACACCCGTCATA,TTGCCTCCATCCATATCTCC);Picalm(GGGAGGGAACAGAAATCCTT,GCACCGATCAACAGTGCAG);Pygo2(TTCTGGGAACTTGTGCACTG,AACTTCCTCCAGCCCATTTT);Ren2(TTGTGAGCTGTAGCCAGGTG,TGTGCACAGCTTGTCTCTCC)andTia1(TGATTGAAGGGCTACTAGAGTGGT,AGCCCCAGAGGACAATTTTT)usingPrimer3software[37].", "role": "user" }, { "content": "[]", "role": "assistant" } ]
science.chemprot_ner
science.chemprot_ner.427
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\", \"Protein\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\" : [\"Dexamethasone\", ...], \"Protein\" : [\"BRCA-1\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nBioactive microconstituents and antioxidant properties of wild edible mushrooms from the island of Lesvos, Greece.\nCrude composition, fatty acids, sterols, total phenolic content (TPC), individual polyphenols and terpenic acids were determined in five wild edible mushrooms species (Lactarius deliciosus, Lactarius sanguifluus, Lactarius semisanguifluus, Russula delica, Suillus bellinii) from Lesvos Island, Greece. In addition, the DPPH scavenging capacity, the ferric ion reducing power (FRAP) and the ferrous ion chelating activity of mushroom methanolic extracts were assessed. Among sterols, ergosterol predominated at concentrations 9.2-18.0mg/100g fw. Total phenolic content of mushroom extracts ranged from 6.0 to 20.8mg GAE/100g fw. Up to 19 simple polyphenols were determined in mushrooms extracts, the more abundant being p-OH-benzoic acid, p-OH-phenylacetic acid, o-coumaric acid, ferulic acid and chrysin. In addition, the triterpenic acids oleanolic and ursolic were detected for the first time in mushrooms. All species exerted antioxidant activity and ferrous ion chelating capacity. Principal component analysis revealed good correlations between TPC, DPPH and FRAP but not with metal chelating activity. It seems that mushrooms polyphenols exert antiradical and reducing activities, but they are not strong metal chelators, the observed chelating ability being probably due to other classes of compounds. To our knowledge, this is the first report on the bioactive microconstituents and antioxidant activity of wild Greek edible mushrooms.", "role": "user" }, { "content": "{\"Chemical\": [\"DPPH\", \"chrysin\", \"ursolic\", \"p-OH-benzoic acid\", \"triterpenic acids\", \"oleanolic\", \"ferric\", \"sterols\", \"terpenic acids\", \"polyphenols\", \"ferulic acid\", \"p-OH-phenylacetic acid\", \"ferrous\", \"fatty acids\", \"o-coumaric acid\", \"ergosterol\"], \"Protein\": []}", "role": "assistant" } ]
science.data_reco_mcq_mc
science.data_reco_mcq_mc.2056
[ { "content": "You are provided with a research question, keyphrases about the question, a description of candidate datasets and dataset options. Read the description of popular datasets provided below and select the ones that can be used to validate the following research question. Use your knowledge of machine learning datasets to make the best judgement.\nYour response should be formatted as a `json` array. For instance, for the query \"Semi supervised image classification\", a sample response might be: [\"CIFAR-10\", \"CIFAR-100\"]. Do not include any extra text in the response other than the answer array.\n\nQuery: In this paper, we propose a deep learning basedapproach for facial action unit detection by enhancing\n\nKeyphrases: facial action unit detection videos\n\nDataset description:\nSAMM Long Videos: The SAMM Long Videos dataset consists of 147 long videos with 343 macro-expressions and 159 micro-expressions. The dataset is FACS-coded with detailed Action Units.\nDISFA: The Denver Intensity of Spontaneous Facial Action (DISFA) dataset consists of 27 videos of 4844 frames each, with 130,788 images in total. Action unit annotations are on different levels of intensity, which are ignored in the following experiments and action units are either set or unset.\nSEWA DB: A database of more than 2000 minutes of audio-visual data of 398 people coming from six cultures, 50% female, and uniformly spanning the age range of 18 to 65 years old. Subjects were recorded in two different contexts: while watching adverts and while discussing adverts in a video chat.\nHollywood 3D dataset: A dataset for benchmarking action recognition algorithms in natural environments, while making use of 3D information. The dataset contains around 650 video clips, across 14 classes.\nBP4D: The BP4D-Spontaneous dataset is a 3D video database of spontaneous facial expressions in a diverse group of young adults. Well-validated emotion inductions were used to elicit expressions of emotion and paralinguistic communication.\nHAA500: HAA500 is a manually annotated human-centric atomic action dataset for action recognition on 500 classes with over 591k labeled frames. Unlike existing atomic action datasets, where coarse-grained atomic actions were labeled with action-verbs, e.g., \"Throw\", HAA500 contains fine-grained atomic actions where only consistent actions fall under the same label, e.g., \"Baseball Pitching\" vs \"Free Throw in Basketball\", to minimize ambiguities in action classification.\n\nOptions:- A: SAMM Long Videos B: DISFA C: SEWA DB D: Hollywood 3D dataset E: BP4D F: HAA500", "role": "user" }, { "content": "[\"DISFA\", \"BP4D\"]", "role": "assistant" } ]
science.genia_ner
science.genia_ner.1526
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"RNA\", \"CellType\", \"CellLine\", \"Protein\", \"DNA\"].\n\nPlease return the output as a JSON object of the format: {\"DNA\" : [\"TAL1\", ...], \"RNA\": [\"mRNAs\", ...], \"CellLine\": [\"T cells\", ...], \"CellType\": [\"monocytes\", ...], \"Protein\": [\"IL-2\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nThe human prointerleukin 1 beta gene requires DNA sequences both proximal and distal to the transcription start site for tissue-specific induction. In these studies, we have identified DNA sequences and specific protein interactions necessary for transcriptional regulation of the human prointerleukin 1 beta (proIL-1 beta) gene. A cell-type-independent lipopolysaccharide (LPS) -responsive enhancer element located between-3757 and-2729 bp upstream from the transcription start site (cap site) consisted of at least six discrete subregions which were essential to the maximal induction by LPS in transfected monocytes. The enhancer also appeared to mediate phorbol myristate acetate induction in monocytes and IL-1 responsiveness in fibroblasts. Deletion and base substitution mutations along with DNA binding studies demonstrated that the enhancer contained a minimum of three functional protein binding sequences, two of which appeared to be important for gene induction. One of the essential proteins which bound to the enhancer was similar or identical to members of the C / EBP family of transcription factors required for both IL-1-and LPS-specific induction of the IL-6 gene (i. e., the NF-IL6 proteins). When ligated to the proIL-1 beta cap site-proximal region (located between-131 to + 12), both the proIL-1 beta and the simian virus 40 enhancer elements functioned more efficiently in monocytes than in HeLa cells, which are not normally competent for IL-1 beta expression. When ligated to the murine c-fos promoter, however, the proIL-1 beta enhancer was inducible in phorbol myristate acetate-stimulated HeLa cells, suggesting the existence of a proIL-1 beta promoter-proximal requirement for tissue specificity.", "role": "user" }, { "content": "{\"CellLine\": [\"phorbol myristate acetate\", \"transfected monocytes\", \"phorbol myristate acetate - stimulated HeLa cells\"], \"CellType\": [\"HeLa cells\", \"monocytes\", \"fibroblasts\"], \"DNA\": [\"- proximal region\", \"cell - type - independent lipopolysaccharide (LPS) - responsive enhancer element\", \"human prointerleukin 1 beta gene\", \"proIL - 1 beta\", \"transcription start site\", \"murine c - fos promoter\", \"DNA sequences\", \"NF - IL6 proteins\", \"proIL - 1 beta cap site\", \"proIL - 1 beta enhancer\", \"IL - 6 gene\", \"human prointerleukin 1 beta (proIL - 1 beta) gene\", \"proIL - 1 beta and the simian virus 40 enhancer elements\", \"protein binding sequences\", \"cap site\", \"- 3757 and - 2729 bp upstream\", \"prointerleukin 1 beta (proIL - 1 beta)\", \"LPS\", \"c - fos\"], \"Protein\": [\"C / EBP family\", \"human prointerleukin 1 beta\", \"IL - 1 beta\", \"transcription factors\", \"proIL - 1 beta\", \"IL - 1\"], \"RNA\": []}", "role": "assistant" } ]
science.bioasq_factoid_qa
science.bioasq_factoid_qa.1343
[ { "content": "You will be shown biomedical passages and your task is to answer the provided question. Provide a concise and accurate response and do not include any additional text in the response other than the answer.\n\nQuestion: What is the most frequent evolution (next stage) when Aortic intramural hematoma (IMH) is not treated?\n\nContext: We consider that IMH may represent a part of a disease (aortic dissection), depicted by radiological images in a specific single instant of its clinical evolution.\nIMH may progress to classic dissection, frank rupture, or aneurysmal dilation; yet, IMH may also regress and be completely resorbed.\nType B acute aortic dissection (AAD) and intramural hematoma (IMH) can both present as potentially catastrophic lesions of the descending aorta. IMH is distinguished from AAD by the absence of an intimal tear and flap. With short-term outcomes being similar to type B AAD, IMH is treated identically to AAD in the corresponding segment of the aorta.\nIntramural hematomas (IMHs) are reported to dynamically evolve into different clinical outcomes ranging from regression to aortic rupture,\nIMH displays a typical of dissection progress, and could be considered as a precursor of classic aortic dissection.\nAortic intramural hematoma (IMH) evolves very dynamically in the short-term to regression, dissection, or aortic rupture\nIntramural hematoma of the aorta is a condition increasingly observed in clinical practice. Uncertainty exists whether such lesions represent a different pathology or simply the precursors of classic dissecting aneurysm\nWe stress that intramural hematoma of the ascending aorta has to be managed as an aortic type A dissection and that aggressive treatment is advisable.\nBACKGROUND: Aortic intramural hematoma (IMH) evolves very dynamically in the short-term to regression, dissection, or aortic rupture.\nLUSIONS: The most frequent long-term evolution of IMH is to aortic aneurysm or pseudoaneurysm. Com\nIMH also frequently leads to aortic emergency, which can be fatal unless rapidly diagnosed and treated.\nAortic intramural hematoma (IMH) is an acute, potentially lethal disorder that is similar to but pathologically distinct from acute aortic dissection.\nBACKGROUND: It has been reported that early surgery should be required for patients with type A aortic intramural hematoma (IMH) because it tends to develop classic aortic dissection\nBACKGROUND: Aortic intramural hematoma (IMH) evolves very dynamically in the short-term to regression, dissection, or aortic\nome cases there is partial or complete regression of the hematoma under medical treatment, but most progress to dissection, aneurysmal dilatation or aortic rupture. The\nase evolution, IMH may progress to classic AD, frank rupture, or aneurysmal dilation; yet, IMH may also regress and be completely resorbed. However, since the nat\nCONCLUSIONS: The most frequent long-term evolution of IMH is to aortic aneurysm or pseudoaneurys\nlater developed pseudoaneurysm. At the end of follow-up, the IMH had regressed completely without dilatation in 17 patients (34%), progressed to classical dissection in 6 (12%), evolved to fusiform aneurysm in 11 (22%), evolved to saccular aneurysm in 4 (8%), and evolv", "role": "user" }, { "content": "classic aortic dissection", "role": "assistant" } ]
science.mslr2022_cochrane_multidoc_summarization
science.mslr2022_cochrane_multidoc_summarization.2232
[ { "content": "You will be presented with a collection of medical randomized controlled trial (RCT) reports addressing a shared clinical question -- for instance, the efficacy of a medical intervention on a disease or condition. Given the titles and abstracts of these reports, your task is to generate the conclusion of the systematic review article synthesizing the key findings. Your conclusion should be a few sentences long. In your response, include only the conclusion and no other text. The RCT reports are below.\n\n\n\nTitle 1:\nA comparison of percutaneous cryosurgery and percutaneous radiofrequency for unresectable hepatic malignancies.\nAbstract 1:\nThe complication and success rates in patients treated with either percutaneous cryosurgery (PCS) or percutaneous radiofrequency (PRF) for unresectable hepatic malignancies are similar.\n Retrospective study.\n University hospital.\n Sixty-four patients were treated with either PCS (n = 31) or PRF (n = 33). Patient treatment was based on the random availability of the probes. Tumors were evaluated by a blinded comparison of pretreatment and posttreatment helical computed tomographic scans. All living patients had at least a 6-month follow-up.\n Complication rate, initial treatment success (complete devascularization of the tumor), and local recurrence (tumor revascularization within or at its periphery).\n The distribution of tumor types was similar in the 2 groups (P =.76). One patient with cirrhosis died of variceal hemorrhage on day 30 after PCS (mortality, 3.2%), while no mortality was observed after PRF (P =.48). Complications occurred in 9 (29%) of the patients following PCS and in 8 (24%) of the patients following PRF (P =.66). Initial treatment success was comparable in the 2 treatment groups (30 [83%] of 36 tumors following PCS vs 34 [83%] of 41 tumors following PRF). However, local recurrences occurred more frequently after PCS than after PRF (16 [53%] of 30 vs 6 [18%] of 34; P =.003). The higher rate of local recurrence was identified for metastases (10 [71%] of 14 after PCS vs 3 [19%] of 16 after PRF; P =.004), while the difference was not significant for hepatocellular carcinoma (6 [38%] of 16 after PCS vs 3 [17%] of 18 after PRF; P =.25). Multivariate analysis demonstrated that the use of PCS (P =.003) and more than 1 treatment (P =.05) were independent risk factors for local tumor recurrence.\n While similar initial treatment success and complication rates are observed following either PCS or PRF, local recurrences occur more frequently following PCS, particularly for metastases.\n\n\nTitle 2:\nIntraoperative radiofrequency ablation or cryoablation for hepatic malignancies.\nAbstract 2:\nThe majority of patients with primary or metastatic malignancies confined to the liver are not candidates for resection because of tumor size, location, multifocality, or inadequate functional hepatic reserve. Cryoablation has become a common treatment in select groups of these patients with unresectable liver tumors. However, hepatic cryoablation is associated with significant morbidity. Radiofrequency ablation (RFA) is a technique that destroys liver tumors in situ by localized application of heat to produce coagulative necrosis. In this study, we compared the complication and early local recurrence rates in patients with unresectable malignant liver tumors treated with either cryoablation or RFA.\n Patients with hepatic malignancies were entered into two consecutive prospective, nonrandomized trials. The liver tumors were treated intraoperatively with cryoablation or RFA; intraoperative ultrasonography was used to guide placement of cryoprobes or RFA needles. All patients were followed up postoperatively to assess complications, treatment response, and local recurrence of malignant disease.\n Cryoablation was performed on 88 tumors in 54 patients, and RFA was used to treat 138 tumors in 92 patients. Treatment-related complications, including 1 postoperative death, occurred in 22 of the 54 patients treated with cryoablation (40.7% complication rate). In contrast, there were no treatment-related deaths and only 3 complications after RFA (3.3% complication rate, P<0.001). With a median follow-up of 15 months in both patient groups, tumor has recurred in 3 of 138 lesions treated with RFA (2.2%), versus 12 of 88 tumors treated with cryoablation (13.6%, P<0.01).\n RFA is a safe, well-tolerated treatment for patients with unresectable hepatic malignancies. This study indicates that (1) complications occur much less frequently following RFA of liver tumors compared with cryoablation of liver tumors, and (2) early local tumor recurrence is infrequent following RFA.\n\n\nTitle 3:\nCryosurgical ablation and radiofrequency ablation for unresectable hepatic malignant neoplasms: a proposed algorithm.\nAbstract 3:\nThermal ablation of unresectable hepatic tumors can be achieved by cryosurgical ablation (CSA) or radiofrequency ablation (RFA). The relative advantages and disadvantages of each technique have not yet been determined.\n Radiofrequency ablation of malignant hepatic neoplasms can be performed safely, but is currently limited by size. Cryosurgical ablation, while associated with higher morbidity, is more effective for larger unresectable hepatic malignant neoplasms.\n Retrospective analysis of prospective patient database.\n Between July 1992 and September 1999, 308 patients with liver tumors not amenable to curative surgical resection were treated with CSA and/or RFA (percutaneous, laparoscopic, celiotomy). No patient had preoperative evidence of extrahepatic disease. All patients underwent laparoscopy with intraoperative ultrasound if technically possible. Both RFA and CSA were performed under ultrasound guidance. Resection, as an adjunctive procedure, was combined with ablation in certain patients.\n Laparoscopy identified extrahepatic disease in 12% of patients, and intraoperative hepatic ultrasound identified additional lesions in 33% of patients, despite extensive preoperative imaging. Radiofrequency ablation alone or combined with resection or CSA resulted in reduced blood loss (P<.05), thrombocytopenia (P<.05), and shorter hospital stay compared with CSA alone (P<.05). Median ablation times for lesions greater than 3 cm were 60 minutes with RFA and 15 minutes with CSA (P<.001). Local recurrence rates for lesions greater than 3 cm were also greater with RFA (38% vs 17%).\n Laparoscopy and intraoperative ultrasound are essential in staging patients with hepatic malignant neoplasms. Radiofrequency ablation when combined with CSA reduces the morbidity of multiple freezes. Although RFA is safer than CSA and can be performed via different approaches (percutaneously, laparoscopically, or at celiotomy), it is limited by tumor size (<3 cm). Percutaneous RFA should be considered in high-risk patients or those with small local recurrences.", "role": "user" }, { "content": "At present, there is no evidence to recommend or refute cryotherapy for patients with hepatocellular carcinoma. Randomised clinical trials with low-risk of bias may help in defining the role of cryotherapy in the treatment of hepatocellular carcinoma.", "role": "assistant" } ]
science.chia_ner
science.chia_ner.896
[ { "content": "You will be shown inclusion or exclusion criteria for a clinical trial. Given these criteria, your task is to extract all unique entities of the following types: [\"Person\", \"Condition\", \"Value\", \"ReferencePoint\", \"Temporal\", \"Qualifier\", \"Observation\", \"Visit\", \"Measurement\", \"Mood\", \"Device\", \"Scope\", \"Negation\", \"Drug\", \"Procedure\", \"Multiplier\"].\n\nPlease return the output as a JSON object of the format: {\"Condition\": [\"breast cancer\", ...], \"Device\": [\"ports\", ...], \"Drug\": [\"Atenolol\", ...], \"Measurement\": [\"blood pressure\", ...], \"Mood\": [\"suspected\", ...], \"Multiplier\": [\"more than two\", ...], \"Negation\": [\"no\", ...], \"Observation\": [\"history\", ...], \"Person\": [\"male\", ...], \"Procedure\": [\"surgery\", ...], \"Qualifier\": [\"severe\", ...], \"ReferencePoint\": [\"enrollment\", ...], \"Scope\": [\"infection\", ...], \"Temporal\": [\"one day\", ...], \"Value\": [\"50%\", ...], \"Visit\": [\"ICU\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nCurrent inpatient hospitalization or active suicidal ideation requiring referral for inpatient hospitalization for safety.\nHistory of psychotic disorder or manic episode diagnosed by MINI-KID\nHistory of substance dependence diagnosis by MINI-KID (excluding tobacco) or positive urine toxicology.\nPregnancy (urine pregnancy tests on the day of scans for menstruating girls).\nInability to provide written informed consent according to the Yale Human Investigation Committee (HIC) guidelines in English.", "role": "user" }, { "content": "{\"Condition\": [\"substance dependence\", \"suicidal ideation\", \"psychotic disorder\", \"Pregnancy\", \"manic episode\"], \"Device\": [], \"Drug\": [\"tobacco\"], \"Measurement\": [\"urine toxicology\"], \"Mood\": [\"requiring\"], \"Multiplier\": [], \"Negation\": [\"excluding\", \"Inability to provide\"], \"Observation\": [\"History\", \"written informed consent\"], \"Person\": [\"menstruating girls\"], \"Procedure\": [\"MINI-KID\", \"urine pregnancy tests\", \"referral\", \"hospitalization\"], \"Qualifier\": [\"in English\", \"Yale Human Investigation Committee (HIC) guidelines\"], \"ReferencePoint\": [], \"Scope\": [\"substance dependence diagnosis by MINI-KID (excluding tobacco) or positive urine toxicology\", \"psychotic disorder or manic episode\"], \"Temporal\": [\"on the day of scans\", \"Current\", \"active\"], \"Value\": [\"positive\"], \"Visit\": [\"inpatient\", \"inpatient hospitalization\"]}", "role": "assistant" } ]
science.craftchem_ner
science.craftchem_ner.10180
[ { "content": "You will be presented with a sentence from biomedical research text. Extract all terms that correspond to chemical entities of biological interest (ChEBI).\nReturn your identified terms in a JSON array, formatted like this: [\"Entity1\", \"Entity2\", ...].\nIf no such entities are detected, respond by providing an empty JSON array [].\n\nBiomedical context:\nAtP14,GATA4andGATA1levelsaresimilarinwild-typeandmutantSertolicell.", "role": "user" }, { "content": "[]", "role": "assistant" } ]
science.medmentions_ner
science.medmentions_ner.2204
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Finding\", \"Bacterium\", \"ProfessionalOrOccupationalGroup\", \"InjuryOrPoisoning\", \"Eukaryote\", \"BodySystem\", \"Virus\", \"ResearchActivity\", \"BodySubstance\", \"PopulationGroup\", \"Chemical\", \"AnatomicalStructure\", \"SpatialConcept\", \"HealthCareActivity\", \"Organization\", \"BiomedicalOccupationOrDiscipline\", \"Food\", \"ClinicalAttribute\", \"MedicalDevice\", \"BiologicFunction\", \"IntellectualProduct\"].\n\nPlease return the output as a JSON object of the format: {\"Virus\": [\"HIV\", ...], \"Bacterium\": [\"MRSA\", ...], \"AnatomicalStructure\": [\"Lung\", ...], \"BodySystem\": [\"CNS\", ...], \"BodySubstance\": [\"Serum\", ...], \"Finding\": [\"Headache\", ...], \"InjuryOrPoisoning\": [\"Fracture\", ...], \"BiologicFunction\": [\"Death\", ...], \"HealthCareActivity\": [\"Biopsy\", ...], \"ResearchActivity\": [\"Clinical trial\", ...], \"MedicalDevice\": [\"Lenses\", ...], \"SpatialConcept\": [\"Camps\", ...], \"BiomedicalOccupationOrDiscipline\": [\"Forensic medicine\", ...], \"Organization\": [\"WHO\", ...], \"ProfessionalOrOccupationalGroup\": [\"Provider\", ...], \"PopulationGroup\": [\"Swimmers\", ...], \"Chemical\": [\"Gold\", ...], \"Food\": [\"Rice\", ...], \"IntellectualProduct\": [\"RPAM\", ...], \"ClinicalAttribute\": [\"Biomarker\", ...], \"Eukaryote\": [\"Dogs\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nImpact of changes in pill appearance in the adherence to angiotensin receptor blockers and in the blood pressure levels: a retrospective cohort study To assess the level of adherence to angiotensin receptor blockers (ARBs) in patients regularly attending a community pharmacy and the influence of a change in patients' adherence to pharmacological treatment. Retrospective cohort study of a random sample of consecutive patients collecting their medication. 40 community pharmacies in Alicante (Southeast Spain). 602 consecutive ≥18 years old patients following treatment with ARBs at least 3 previous refills were included. Prevalence of uncontrolled blood pressure (BP) and adherence to prescribed pharmacological treatment (measured through both the Batalla and the Morisky-Green tests). A multivariate Poisson regression model was used to estimate the adjusted risk ratio (RRa) for non-adherence to pharmacological treatment by the presence of a change in patient's adherence and other significant variables. 161/602 (13.7%) patients presented uncontrolled BP. According to the Morisky test, 410/602 (68.2%) patients were considered adherent to pharmacological treatment and 231/602 (38.4%) patients according to the Batalla test. According to the Morisky-Green test, in the multivariable analysis, patients with a previous change in pill appearance were less likely to be adherent than those patients with no change in their pharmacological treatment (RRa 0.45; CI 95% 0.22 to 0.90; p=0.024). Systolic BP was higher in patients with a change in pill appearance in the previous 3 refills (median BP 142 mm Hg; IQR 136-148) than in those who did not have a change (median BP 127 mm Hg; IQR 118-135; p<0.001). There was a low percentage of adherence and nearly 15% of uncontrolled BP in patients who regularly collected their medication. Switching between pills of different appearances was associated with lower patient adherence to pharmacological treatment and a higher uncontrolled BP than no change in pharmacological treatment or change only in package but not in pill appearance.", "role": "user" }, { "content": "{\"AnatomicalStructure\": [], \"Bacterium\": [], \"BiologicFunction\": [\"blood pressure\", \"BP\"], \"BiomedicalOccupationOrDiscipline\": [], \"BodySubstance\": [], \"BodySystem\": [], \"Chemical\": [\"pills\", \"pill\", \"angiotensin receptor blockers\", \"medication\", \"ARBs\"], \"ClinicalAttribute\": [\"Systolic BP\"], \"Eukaryote\": [], \"Finding\": [\"non-adherence\", \"blood pressure levels\", \"presence\"], \"Food\": [], \"HealthCareActivity\": [\"Morisky test\", \"treatment\", \"pharmacological treatment\", \"Morisky-Green test\", \"Batalla test\", \"Batalla and the Morisky-Green tests\"], \"InjuryOrPoisoning\": [], \"IntellectualProduct\": [], \"MedicalDevice\": [], \"Organization\": [\"community pharmacy\", \"community pharmacies\"], \"PopulationGroup\": [], \"ProfessionalOrOccupationalGroup\": [], \"ResearchActivity\": [\"retrospective cohort study\", \"random sample\", \"Retrospective cohort study\"], \"SpatialConcept\": [\"Southeast Spain\", \"median\"], \"Virus\": []}", "role": "assistant" } ]
science.bioasq_yesno_qa
science.bioasq_yesno_qa.526
[ { "content": "You have to answer a biomedical question in binary format, i.e., only yes and no are acceptable answers. A list of paragraphs is provided as context to help you answer the question.\n\nQuestion: Do T-Cells regulate neuropathic pain?\n\nContext: here is evidence for a considerable impact of the immune system also in neuropathic pain. However, the role of the adaptive immune system is still unclear.\nOur investigation revealed a clear shift of T-cell subsets towards anti-inflammation in patients with neuropathic pain.\nGITRL expressed on macrophages drives cytokine release and T cell activation, resulting in neuropathic pain via GITR-dependent actions.\nThus, this T-cell subset may be specifically targeted to alleviate chronic neuropathic pain.<CopyrightInformation>Copyright © 2012 International Association for the Study of Pain\nRecent studies show that T cells play an important role in neuropathic pain following nerve injury in rats\nThese results show a peripheral pivotal role of CatS in the development of neuropathic pain through the antigen-specific activation of CD4(+) T-cells\nChemokine (C-C motif) ligand 1 (CCL-1), a well-characterized chemokine secreted by activated T cells, has been shown to play an important role in neuropathic pain induced by nerve injury and is also produced in various cell types in the CNS, especially in dorsal root ganglia (DRG)\nIn the present study, we investigated systemic T-cell subset responses and T-cell related cytokine profiles in patients with chronic neuropathic pain.\nAnti-inflammatory T-cell shift in neuropathic pain.\nThus, this T-cell subset may be specifically targeted to alleviate chronic neuropathic pain..\nRegulatory T cells attenuate neuropathic pain following peripheral nerve injury and experimental autoimmune neuritis.\nMacrophage-T cell interactions mediate neuropathic pain through the glucocorticoid-induced tumor necrosis factor ligand system.", "role": "user" }, { "content": "yes", "role": "assistant" } ]
science.chemdner_ner
science.chemdner_ner.13
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Chemical\"].\n\nPlease return the output as a JSON object of the format: {\"Chemical\": [\"Arsenic\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nSeasonal variation in biomarkers in blue mussel (Mytilus edulis), Icelandic scallop (Chlamys islandica) and Atlantic cod (Gadus morhua): implications for environmental monitoring in the Barents Sea. In the Barents Sea, the limited data on biological relevant indicators and their responses to various anthropogenic stressors have hindered the development of a consistent scientific basis for selecting indicator species and developing practical procedures for environmental monitoring. Accordingly, the main aim of the present study was to develop a common set of baseline values for contaminants and biomarkers in three species, and to identify their strengths and limitations in monitoring of the Barents Sea. Blue mussel (Mytilus edulis), Icelandic scallop (Chlamys islandica) and Atlantic cod (Gadus morhua) were sampled from a north Norwegian fjord in March, June, September and December 2010. Digestive glands from the bivalve species and liver from Atlantic cod were analysed for biomarkers of oxidative stress (catalase [CAT], glutathione peroxidase [GPX], glutathione-S-transferase activities [GST], lipid peroxidation as thiobarbituric reactive substances [TBARS] and total oxyradical scavenging capacity [TOSC]), biotransformation (ethoxyresorufine-O-deethylase activity [EROD]) and general stress (lysosomal membrane stability [LMS]). Concentrations of polycyclic aromatic hydrocarbons (PAHs) and metals in the bivalves and PAH metabolites in fish bile were quantified. Finally, energy reserves (total lipids, proteins and carbohydrates) and electron transport system (ETS) activity in the digestive gland of the bivalves and liver of Atlantic cod provided background information for reproductive cycle and general physiological status of the organisms. Blue mussel and Icelandic scallop showed very similar trends in biological cycle, biomarker expression and seasonality. Biomarker baselines in Atlantic cod showed weaker seasonal variability. However, important biological events may have been undetected due to the large time intervals between sampling occasions. Physiological biomarkers such as energy reserves and ETS activity were recommended as complementary parameters to the commonly used stress biomarkers, as they provided valuable information on the physiological status of the studied organisms. Interpretation of the seasonality in oxidative stress biomarkers was in general difficult but TOSC and lipid peroxidation were preferred over the antioxidant enzyme activities. This study is the first reporting seasonal baseline in these three species in a sub-Arctic location. Overall, the Icelandic scallop was considered the most adequate organism for environmental monitoring in the Barents Sea due to the interpretability of the biomarker data as well as its abundance, ease to handle and wide distribution from the southern Barents Sea to Svalbard.", "role": "user" }, { "content": "{\"Chemical\": [\"O\", \"S\", \"PAHs\", \"glutathione\", \"PAH\", \"ethoxyresorufine\", \"polycyclic aromatic hydrocarbons\", \"thiobarbituric\"]}", "role": "assistant" } ]
science.medmentions_ner
science.medmentions_ner.2143
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"Finding\", \"Bacterium\", \"ProfessionalOrOccupationalGroup\", \"InjuryOrPoisoning\", \"Eukaryote\", \"BodySystem\", \"Virus\", \"ResearchActivity\", \"BodySubstance\", \"PopulationGroup\", \"Chemical\", \"AnatomicalStructure\", \"SpatialConcept\", \"HealthCareActivity\", \"Organization\", \"BiomedicalOccupationOrDiscipline\", \"Food\", \"ClinicalAttribute\", \"MedicalDevice\", \"BiologicFunction\", \"IntellectualProduct\"].\n\nPlease return the output as a JSON object of the format: {\"Virus\": [\"HIV\", ...], \"Bacterium\": [\"MRSA\", ...], \"AnatomicalStructure\": [\"Lung\", ...], \"BodySystem\": [\"CNS\", ...], \"BodySubstance\": [\"Serum\", ...], \"Finding\": [\"Headache\", ...], \"InjuryOrPoisoning\": [\"Fracture\", ...], \"BiologicFunction\": [\"Death\", ...], \"HealthCareActivity\": [\"Biopsy\", ...], \"ResearchActivity\": [\"Clinical trial\", ...], \"MedicalDevice\": [\"Lenses\", ...], \"SpatialConcept\": [\"Camps\", ...], \"BiomedicalOccupationOrDiscipline\": [\"Forensic medicine\", ...], \"Organization\": [\"WHO\", ...], \"ProfessionalOrOccupationalGroup\": [\"Provider\", ...], \"PopulationGroup\": [\"Swimmers\", ...], \"Chemical\": [\"Gold\", ...], \"Food\": [\"Rice\", ...], \"IntellectualProduct\": [\"RPAM\", ...], \"ClinicalAttribute\": [\"Biomarker\", ...], \"Eukaryote\": [\"Dogs\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nToll-like receptors genes polymorphisms and the occurrence of HCMV infection among pregnant women Human cytomegalovirus (HCMV) is the most common cause of intrauterine infections worldwide. The toll-like receptors (TLRs) have been reported as important factors in immune response against HCMV. Particularly, TLR2, TLR4 and TLR9 have been shown to be involved in antiviral immunity. Evaluation of the role of single nucleotide polymorphisms (SNPs), located within TLR2, TLR4 and TLR9 genes, in the development of human cytomegalovirus (HCMV) infection in pregnant women and their fetuses and neonates, was performed. The study was performed for 131 pregnant women, including 66 patients infected with HCMV during pregnancy, and 65 age - matched control pregnant individuals. The patients were selected to the study, based on serological status of anti-HCMV IgG and IgM antibodies and on the presence of viral DNA in their body fluids. Genotypes in TLR2 2258 A > G, TLR4 896 G > A and 1196 C > T and TLR9 2848 G > A SNPs were determined by self-designed nested PCR - RFLP assays. Randomly selected PCR products, representative for distinct genotypes in TLR SNPs, were confirmed by sequencing. A relationship between the genotypes, alleles, haplotypes and multiple variants in the studied polymorphisms, and the occurrence of HCMV infection in pregnant women and their offsprings, was determined, using a logistic regression model. Genotypes in all the analyzed polymorphisms preserved the Hardy-Weinberg equilibrium in pregnant women, both infected and uninfected with HCMV (P > 0.050). G G homozygotic and G A heterozygotic status in TLR9 2848 G > A SNP decreased significantly the occurrence of HCMV infection (OR 0.44 95% CI 0.21-0.94 in the dominant model, P ≤ 0.050). The G allele in TLR9 SNP was significantly more frequent among the uninfected pregnant women than among the infected ones (χ(2) = 4.14, P ≤ 0.050). Considering other polymorphisms, similar frequencies of distinct genotypes, haplotypes and multiple-SNP variants were observed between the studied groups of patients. TLR9 2848 G > A SNP may be associated with HCMV infection in pregnant women.", "role": "user" }, { "content": "{\"AnatomicalStructure\": [\"multiple variants\", \"fetuses\", \"TLR9 genes\", \"TLR4\", \"TLR2\", \"multiple-SNP variants\", \"alleles\", \"allele\", \"TLR9\"], \"Bacterium\": [], \"BiologicFunction\": [\"human cytomegalovirus (HCMV) infection\", \"HCMV infection\", \"antiviral immunity\", \"intrauterine infections\", \"pregnancy\", \"immune response\", \"analyzed polymorphisms\"], \"BiomedicalOccupationOrDiscipline\": [], \"BodySubstance\": [\"body fluids\"], \"BodySystem\": [], \"Chemical\": [\"Toll-like receptors\", \"C\", \"TLRs\", \"toll-like receptors\", \"T\", \"anti-HCMV IgG\", \"A\", \"viral DNA\", \"G\", \"IgM antibodies\"], \"ClinicalAttribute\": [], \"Eukaryote\": [], \"Finding\": [\"infected\"], \"Food\": [], \"HealthCareActivity\": [\"RFLP assays\", \"sequencing\"], \"InjuryOrPoisoning\": [], \"IntellectualProduct\": [\"dominant model\"], \"MedicalDevice\": [], \"Organization\": [], \"PopulationGroup\": [\"pregnant women\", \"pregnant individuals\"], \"ProfessionalOrOccupationalGroup\": [], \"ResearchActivity\": [\"self-designed nested PCR\", \"PCR\"], \"SpatialConcept\": [\"SNPs\", \"located within\", \"SNP\", \"single nucleotide polymorphisms\"], \"Virus\": [\"HCMV\", \"Human cytomegalovirus\"]}", "role": "assistant" } ]
science.craftchem_ner
science.craftchem_ner.8403
[ { "content": "You will be presented with a sentence from biomedical research text. Extract all terms that correspond to chemical entities of biological interest (ChEBI).\nReturn your identified terms in a JSON array, formatted like this: [\"Entity1\", \"Entity2\", ...].\nIf no such entities are detected, respond by providing an empty JSON array [].\n\nBiomedical context:\nQ1isacosmidcontainingtheRanBP2geneuptoexon20[4].", "role": "user" }, { "content": "[]", "role": "assistant" } ]
science.qasper_extractive_qa
science.qasper_extractive_qa.634.0
[ { "content": "You will be shown sections from a scientific research paper, together with a question about the paper. This is an extractive question-answering task, where you must find and extract relevant text spans directly from the paper to answer the question. Your response should strictly be a `json` object with two fields:\n\n\"answer\": An array of strings extracted directly from the paper which, collectively, answer the question.\n\"evidence\": An array of strings. Each should be an excerpt from the paper, in which one or more of the extracted answers can be found.\n\nFor example, for the question \"What baselines did the authors compare against?\", a sample response might be:\n{\n \"answer\": [\"BERT\", \"RoBERTa\"],\n \"evidence\": [\"In our experiments, we compare the performance of our model against BERT and RoBERTa.\"]\n}\n\nDo not include any text in your response other than the json.\n\nIf the question is unanswerable given the provided excerpts, respond with the single word \"null\".\n\nPaper: Global Greedy Dependency Parsing\n\nMost syntactic dependency parsing models may fall into one of two categories: transition- and graph-based models. The former models enjoy high inference efficiency with linear time complexity, but they rely on the stacking or re-ranking of partially-built parse trees to build a complete parse tree and are stuck with slower training for the necessity of dynamic oracle training. The latter, graph-based models, may boast better performance but are unfortunately marred by polynomial time inference. In this paper, we propose a novel parsing order objective, resulting in a novel dependency parsing model capable of both global (in sentence scope) feature extraction as in graph models and linear time inference as in transitional models. The proposed global greedy parser only uses two arc-building actions, left and right arcs, for projective parsing. When equipped with two extra non-projective arc-building actions, the proposed parser may also smoothly support non-projective parsing. Using multiple benchmark treebanks, including the Penn Treebank (PTB), the CoNLL-X treebanks, and the Universal Dependency Treebanks, we evaluate our parser and demonstrate that the proposed novel parser achieves good performance with faster training and decoding.\n\nGlobal Greedy Parsing Model ::: Scorers\nThe traditional easy-first model relies on an incremental tree scoring process with stepwise loss backpropagation and sub-tree removal facilitated by local scoring, relying on the scorer and loss backpropagation to hopefully obtain the parsing order. Communicating the information from the scorer and the loss requires training a dynamic oracle, which exposes the model to the configurations resulting from erroneous decisions. This training process is done at the token level, not the sentence level, which unfortunately means incremental scoring prevents parallelized training and causes error propagation. We thus forego incremental local scoring, and, inspired by the design of graph-based parsing models, we instead choose to score all of the syntactic arc candidates in one-shot, which allows for global featuring at a sentence level; however, the introduction of one-shot scoring brings new problems. Since the graph-based method relies on a tree space search algorithm to find the tree with the highest score, the parsing order is not important at all. If we apply one-shot scoring to greedy parsing, we need a mechanism like a stack (as is used in transition-based parsing) to preserve the parsing order.\nBoth transition-based and easy-first parsers build parse trees in an incremental style, which forces tree formation to follow an order starting from either the root and working towards the leaf nodes or vice versa. When a parser builds an arc that skips any layer, certain errors will exist that it will be impossible for the parent node to find. We thus implement a parsing order prediction module to learn a parsing order objective that outputs a parsing order score addition to the arc score to ensure that each pending node is attached to its parent only after all (or at least as many as possible) of its children have been collected.\nOur scorer consists of two parts: a biaffine scorer for one-shot scoring and a parsing order scorer for parsing order guiding. For the biaffine scorer, we adopt the biaffine attention mechanism BIBREF18 to score all possible head-dependent pairs:\nwhere $\\textbf {W}_{arc}$, $\\textbf {U}_{arc}$, $\\textbf {V}_{arc}$, $\\textbf {b}_{arc}$ are the weight matrix of the bi-linear term, the two weight vectors of the linear terms, and the bias vector, respectively.\nIf we perform greedy inference only on the $s_{arc}$ directly, as in Figure FIGREF6, at step $i$, the decoder tests every pair in the pending list, and although the current score fits the correct tree structure for this example, because backtracking is not allowed in the deterministic greedy inference, according to the maximum score $s_{arc}$, the edge selected in step $i$+1 is “root\"$\\rightarrow $“come\". This prevents the child nodes (“today\" and “.\") from finding the correct parent node in the subsequent step. Thus, the decoder is stuck with this error. This problem can be solved or mitigated by using a max spanning tree (MST) decoder or by adding beam search method to the inference, but neither guarantees maintaining linear time decoding. Therefore, we propose a new scorer for parsing order $s_{order}$. In the scoring stage, the parsing order score is passed to the decoder to guide it and prevent (as much as possible) resorting to erroneous choices.\nWe formally define the parsing order score for decoding. To decode the nodes at the bottom of the syntax tree first, we define the the parsing order priority as the layer “level\" or “position\" in the tree. The biaffine output score is the probability of edge (dependency) existence, between 0 and 1, so the greater the probability, the more likely an edge is to exist. Thus, our parsing order scorer gives a layer score for a node, and then, we add this layer score to the biaffine score. Consequently, the relative score of the same layer can be kept unchanged, and the higher the score of a node in the bottom layer, the higher its decoding priority will be. We therefore define $s_{order}$ as:\nwhere $\\textbf {W}_{order}$ and $\\textbf {b}_{order} $ are parameters for the parsing order scorer. Finally, the one-shot arc score is:\nSimilarly, we use the biaffine scorer for dependency label classification. We apply MLPs to the contextualized representations before using them in the label classifier as well. As with other graph-based models, the predicted tree at training time has each word as a dependent of its highest-scoring head (although at test time we ensure that the parse is a well-formed tree via the greedy parsing algorithm).\n\nGlobal Greedy Parsing Model ::: Training Objectives\nTo parse the syntax tree $y$ for a sentence $x$ with length $l$, the easy-first model relies on an action-by-action process performed on pending. In the training stage, the loss is accumulated once per step (action), and the model is updated by gradient backpropagation according to a preset frequency. This prohibits parallelism during model training lack between and within sentences. Therefore, the traditional easy-first model was trained to maximize following probability:\nwhere $\\emph {pending}_i$ is the pending list state at step $i$.\nWhile for our proposed model, it uses the training method similar to that of graph-based models, in which the arc scores are all obtained in one-shot. Consequently, it does not rely on the pending list in the training phase and only uses the pending list to promote the process of linear parsing in the inference stage. Our model is trained to optimize the probability of the dependency tree $y$ when given a sentence $x$: $P_\\theta (y|x)$, which can be factorized as:\nwhere $\\theta $ represents learnable parameters, $l$ denotes the length of the processing sentence, and $y^{arc}_i$, $y^{rel}_i$ denote the highest-scoring head and dependency relation for node $x_i$. Thus, our model factors the distribution according to a bottom-up tree structure.\nCorresponding to multiple objectives, several parts compose the loss of our model. The overall training loss is the sum of three objectives:\nwhere the loss for arc prediction $\\mathcal {L}^{arc}$ is the negative log-likelihood loss of the golden structure $y^{arc}$:\nthe loss for relation prediction $\\mathcal {L}^{rel}$ is implemented as the negative log-likelihood loss of the golden relation $y^{rel}$ with the golden structure $y^{arc}$,\nand the loss for parsing order prediction $\\mathcal {L}^{order}$:\nBecause the parsing order score of each layer in the tree increases by 1, we frame it as a classification problem and therefore add a multi-class classifier module as the order scorer.\n\nExperiments ::: Treebanks\nFor English, we use the Stanford Dependency (SD 3.3.0) BIBREF37 conversion of the Penn Treebank BIBREF38, and follow the standard splitting convention for PTB, using sections 2-21 for training, section 22 as a development set and section 23 as a test set. We use the Stanford POS tagger BIBREF39 generate predicted POS tags.\nFor Chinese, we adopt the splitting convention for CTB BIBREF40 described in BIBREF19. The dependencies are converted with the Penn2Malt converter. Gold segmentation and POS tags are used as in previous work BIBREF19.\nFor the CoNLL Treebanks, we use the English treebank from the CoNLL-2008 shared task BIBREF41 and all 13 treebanks from the CoNLL-X shared task BIBREF42. The experimental settings are the same as BIBREF43.\nFor UD Treebanks, following the selection of BIBREF23, we take 12 treebanks from UD version 2.1 (Nivre et al. 2017): Bulgarian (bg), Catalan (ca), Czech (cs), Dutch (nl), English (en), French (fr), German (de), Italian (it), Norwegian (no), Romanian (ro), Russian (ru) and Spanish (es). We adopt the standard training/dev/test splits and use the universal POS tags provided in each treebank for all the languages.\n\nExperiments ::: CoNLL Results\nTable TABREF11 presents the results on 14 treebanks from the CoNLL shared tasks. Our model yields the best results on both UAS and LAS metrics of all languages except the Japanese. As for Japanese, our model gives unsatisfactory results because the original treebank was written in Roman phonetic characters instead of hiragana, which is used by both common Japanese writing and our pre-trained embeddings. Despite this, our model overall still gives 1.0% higher average UAS and LAS than the previous best parser, BIAF.\n\nExperiments ::: UD Results\nFollowing BIBREF23, we report results on the test sets of 12 different languages from the UD treebanks along with the current state-of-the-art: BIAF and STACKPTR. Although both BIAF and STACKPTR parsers have achieved relatively high parsing accuracies on the 12 languages and have all UAS higher than 90%, our model achieves state-of-the-art results in all languages for both UAS and LAS. Overall, our model reports more than 1.0% higher average UAS than STACKPTR and 0.3% higher than BIAF.\n\nConclusion\nThis paper presents a new global greedy parser in which we enable greedy parsing inference compatible with the global arc scoring of graph-based parsing models instead of the local feature scoring of transitional parsing models. The proposed parser can perform projective parsing when only using two arc-building actions, and it also supports non-projective parsing when introducing two extra non-projective arc-building actions. Compared to graph-based and transition-based parsers, our parser achieves a better tradeoff between parsing accuracy and efficiency by taking advantages of both graph-based models' training methods and transition-based models' linear time decoding strategies. Experimental results on 28 treebanks show the effectiveness of our parser by achieving good performance on 27 treebanks, including the PTB and CTB benchmarks.\n\nTable 1: Comparison of results on the test sets. “T”, “G” and “E” indicate transition-based, graph-based and easy-first models, respectively. The “G + E” represents the graph-based training while the easy-first algorithm is used for inference. Acronyms used: (g) – greedy, (b) – beam search, (re) – re-ranking, (3rd) – 3rd-order, (1st) – 1st-order. The “*” in the upper right corner of the results is because the original ELMo (Peters et al. 2018) has no Chinese version. We instead used the multilingual version “HIT-ELMo” pre-trained by (Che et al. 2018).\n\nTable 4: Training time and decoding speed. The experimental environment is on the same machine with Intel i9 9900k CPU and NVIDIA 1080Ti GPU.\n\nTable 3: UAS and LAS on test datasets of 12 treebanks from UD Treebanks, together with BIAF and STACKPTR for comparison.\n\nQuestion: What are performance compared to former models?", "role": "user" }, { "content": "{\"answer\": [\"model overall still gives 1.0% higher average UAS and LAS than the previous best parser, BIAF\", \"our model reports more than 1.0% higher average UAS than STACKPTR and 0.3% higher than BIAF\"], \"evidence\": [\"Table TABREF11 presents the results on 14 treebanks from the CoNLL shared tasks. Our model yields the best results on both UAS and LAS metrics of all languages except the Japanese. As for Japanese, our model gives unsatisfactory results because the original treebank was written in Roman phonetic characters instead of hiragana, which is used by both common Japanese writing and our pre-trained embeddings. Despite this, our model overall still gives 1.0% higher average UAS and LAS than the previous best parser, BIAF.\", \"Following BIBREF23, we report results on the test sets of 12 different languages from the UD treebanks along with the current state-of-the-art: BIAF and STACKPTR. Although both BIAF and STACKPTR parsers have achieved relatively high parsing accuracies on the 12 languages and have all UAS higher than 90%, our model achieves state-of-the-art results in all languages for both UAS and LAS. Overall, our model reports more than 1.0% higher average UAS than STACKPTR and 0.3% higher than BIAF.\"]}", "role": "assistant" } ]
science.drug_combo_extraction_re
science.drug_combo_extraction_re.759
[ { "content": "You will be shown a paragraph from a biomedical research article. Your task is to extract all drug combinations appearing in a specific sentence within the paragraph. Each drug combination should be assigned to one of the following three categories:\n\n- POS: Indicates a positive interaction between the drugs in the combination; for instance, drugs that have an additive or synergistic effect.\n- NEG: Indicates a negative interaction between the drugs in the combination; for instance, combining these drugs might result in adverse side effects, or might reduce their efficacy.\n- COMB: Indicates that the drugs are given in combination, but the type of interaction between the drugs cannot be determined based on the provided context.\n\nPlease format your answer as a JSON array. Each entry in the array should itself be an array formatted as [<drug_1>, <drug_2>, ..., <drug_n>, <category>] where <category> is one of the three categories listed above and <drug_1>, ..., <drug_n> are drug names copied verbatim from the input sentence. Different combinations may involve different numbers of drugs; in other words, the `n` in <drug_n> may vary.\n\nAs an example, the following output describes two extracted drug combinations; one between two drugs, and one between three drugs:\n\n[[\"artesunate\", \"artemether\", \"COMB\"], [\"Nifedipine\", \"Labetalol\", \"Prazosin\", \"POS\"]]\n\nThe paragraph and specific sentence are below. You should only extract relationships expressed in the specific sentence; the paragraph is provided for context.\n\nIn your output, return only the JSON array and no other text. If no drug combinations appear, return an empty array `[]`.\n\nparagraph: Curcumin Rescues Doxorubicin Responsiveness via Regulating Aurora a Signaling Network in Breast Cancer Cells. Insensitivity towards anthracycline drugs like doxorubicin poses a significant challenge in the treatment of breast cancer. Among several factors, Aurora A (a mitotic serine threonine kinase) plays crucial roles in acquiring non-responsiveness towards doxorubicin. However, the mechanisms underlying need to be elucidated. The present study was therefore designed to evaluate the underlying mechanisms of Aurora A mediated doxorubicin insensitivity in MCF-7Dox/R, an isolated resistant-subline of MCF-7 (breast adenocarcinoma cell line). Effect of curcumin , a natural phytochemical in restoring doxorubicin sensitivity by targeting Aurora A was assessed furthermore . ### methods A doxorubicin resistant subline (MCF-7Dox/R) was isolated from the parental MCF-7 cells by treating the cell with gradual step-wise increasing concentration of the drug. Expressions of Aurora A and its target proteins (Akt, IκBα and NFκB) were assessed in both parental and MCF-7Dox/R cells. Both the cell lines were pretreated with curcumin prior to doxorubicin treatment. Cellular proliferation rate was measured using BrdU (5-bromo-2'-deoxyuridine) assay kit. Intracellular doxorubicin accumulation was estimated spectrofluorimetrically. Cellular uptake of curcumin (spectrophotometric and spectrofluorimetric method) and its nuclear localization was confirmed by confocal microscopic study. Protein expressions were determined by western blot analysis. Localization of Aurora A was ascertained by immunofluorescence assay. To explore the possible outcome of impact of curcumin on Aurora A, cell-cycle distribution and apoptosis were performed subsequently. ### results Higher expressions of Aurora A in MCF-7Dox/R cells led to phosphorylation of Akt as well as IκBα. Phosphorylated IκBα preceded release of NFκB. Phospho-Akt, NFκB consequentially decreased doxorubicin accumulation by enhancing the expressions of ABCG2 and Pgp1 respectively. curcumin by regulating Aurora A and its target molecules sensitized resistant subline towards doxorubicin mediated G2/M-arrest and apoptosis. ### conclusion Molecular targeting of Aurora A by curcumin restores chemosensitivity by increasing the efficacy of doxorubicin in breast cancer..\n\nsentence: Effect of curcumin , a natural phytochemical in restoring doxorubicin sensitivity by targeting Aurora A was assessed furthermore .", "role": "user" }, { "content": "[[\"curcumin\", \"doxorubicin\", \"POS\"]]", "role": "assistant" } ]
science.genia_ner
science.genia_ner.781
[ { "content": "You will be shown an abstract from a biomedical research paper. Given this abstract, your task is to extract all unique entities of the following types: [\"RNA\", \"CellType\", \"CellLine\", \"Protein\", \"DNA\"].\n\nPlease return the output as a JSON object of the format: {\"DNA\" : [\"TAL1\", ...], \"RNA\": [\"mRNAs\", ...], \"CellLine\": [\"T cells\", ...], \"CellType\": [\"monocytes\", ...], \"Protein\": [\"IL-2\", ...]}. The keys should be entity types and values should be lists of extracted entities belonging to the corresponding type. If you cannot find entities belonging to a specific type, the value should be [].\n\nOnly output the JSON object and do not include any additional text.\n\nAbstract:\n\nT helper differentiation proceeds through Stat1-dependent, Stat4-dependent and Stat4-independent phases. Much of our focus in understanding Th1 / Th2 development has been on the signals delivered by IL-12 and IL-4 as final determinants of terminal T cell differentiation. Because extinction of IL-12 signaling in early Th2 development could potentially be important in imprinting a more permanent Th2 phenotype on a population of T cells, we have also examined various parameters regulating the IL-12 signaling pathway. Whereas IL-4 appears to repress functional IL-12 signaling through inhibition of IL-12R beta 2 expression, IFN-gamma in the mouse, and IFN-alpha in the human appear to induce IL-12R beta 2 expression and promote IL-12 responsiveness. We propose that Th1 development can be considered in two stages, capacitance and development. Capacitance would simply involve expression of IL-12R beta 1 and beta 2 subunits, regulated by TCR, IL-4 and IFNs. The second stage, development, we propose is the true IL-12 induced developmental stage, involving expression of Stat4 inducible proteins. In the human, this may also occur via IFN-alpha, which is able to activate Stat4. It is perhaps possible that all of Stat4 actions on Th1 development may be exert directly by Stat4 at the IFN-gamma gene, however we suggest that, more likely, Stat4 may act to induce Th1 development through the induction of other non-cytokine genes, whose stable expression maintains the transcriptional state of a Th1 cell.", "role": "user" }, { "content": "{\"CellLine\": [], \"CellType\": [\"T cells\", \"Th1 cell\"], \"DNA\": [\"IFN - gamma gene\", \"non - cytokine genes\"], \"Protein\": [\"IFNs\", \"IFN - alpha\", \"IL - 12R beta 2\", \"IL - 12R beta 1 and beta 2 subunits\", \"Stat4 inducible proteins\", \"IL - 4\", \"IFN - gamma\", \"IL - 12\", \"Stat4\"], \"RNA\": []}", "role": "assistant" } ]
science.healthver_entailment
science.healthver_entailment.495
[ { "content": "You will be shown a claim about public health and the abstract of a biomedical research paper. Each sentence from the abstract will be on a separate line. Your task is to return a JSON object with two fields:\n\n- \"verdict\": The fact-checking verdict. If the information in the abstract supports the claim, write \"SUPPORT\". If the abstract contradicts the claim, write \"CONTRADICT\". If the abstract does not provide enough information to arrive at a verdict, write \"NEI\" (for \"not enough information\").\n- \"evidence\": An array of sentences providing evidence for the verdict. Please copy all relevant sentences verbatim from the abstract. If the verdict was \"NEI\", then return an empty array.\n\nFor instance, if the model were given the claim \"wearing masks can prevent the spread of COVID\", the output might be:\n{\n \"verdict\": \"SUPPORT\",\n \"evidence\": [\"Our findings indicate that mass mask-wearing reduces the transmission rate for COVID-19.\"]\n}\n\nClaim: But there is a lot of competition, with more than 200 coronavirus vaccine candidates under investigation around the world. Two dozen are undergoing testing in humans, and six are in phase 3 \n\nAbstract:\nCOVID-19 is an emerging infectious disease and was recently declared as a pandemic by WHO.\nCurrently, there is no vaccine or therapeutic available for this disease.\nDrug repositioning represents the only feasible option to address this global challenge and a panel of 48 FDA-approved drugs that have been pre-selected by an assay of SARS-CoV was screened to identify potential antiviral drug candidates against SARS-CoV-2 infection.\nWe found a total of 24 drugs which exhibited antiviral efficacy (0.1 μM < IC50 < 10 μM) against SARS-CoV-2.\nIn particular, two FDA-approved drugs - niclosamide and ciclesonide – were notable in some respects.\nThese drugs will be tested in an appropriate animal model for their antiviral activities.\nIn near future, these already FDA-approved drugs could be further developed following clinical trials in order to provide additional therapeutic options for patients with COVID-19.", "role": "user" }, { "content": "{\"verdict\": \"CONTRADICT\", \"evidence\": [\"Currently, there is no vaccine or therapeutic available for this disease.\"]}", "role": "assistant" } ]
science.scitldr_aic
science.scitldr_aic.1947
[ { "content": "Below are the Abstract, Introduction, and Conclusion sections of a computer science research paper. Please summarize the main contribution of the work in a single sentence. Your response should include the summary sentence and no additional text.\n\nPaper text:\n\nWe introduce a new and rigorously-formulated PAC-Bayes few-shot meta-learning algorithm that implicitly learns a model prior distribution of interest.\nOur proposed method extends the PAC-Bayes framework from a single task setting to the few-shot meta-learning setting to upper-bound generalisation errors on unseen tasks.\nWe also propose a generative-based approach to model the shared prior and task-specific posterior more expressively compared to the usual diagonal Gaussian assumption.\nWe show that the models trained with our proposed meta-learning algorithm are well calibrated and accurate, with state-of-the-art calibration and classification results on mini-ImageNet benchmark, and competitive results in a multi-modal task-distribution regression.\nOne unique ability of humans is to be able to quickly learn new tasks with only a few training examples.\nThis is due to the fact that humans tend to exploit prior experience to facilitate the learning of new tasks.\nSuch exploitation is markedly different from conventional machine learning approaches, where no prior knowledge (e.g. training from scratch with random initialisation) (Glorot & Bengio, 2010) , or weak prior knowledge (e.g., fine tuning from pre-trained models) (Rosenstein et al., 2005) are used when encountering an unseen task for training.\nThis motivates the development of novel learning algorithms that can effectively encode the knowledge learnt from training tasks, and exploit that knowledge to quickly adapt to future tasks (Lake et al., 2015) .\nPrior knowledge can be helpful for future learning only if all tasks are assumed to be distributed according to a latent task distribution.\nLearning this latent distribution is, therefore, useful for solving an unseen task, even if the task contains a limited number of training samples.\nMany approaches have been proposed and developed to achieve this goal, namely: multi-task learning (Caruana, 1997) , domain adaptation (Bridle & Cox, 1991; Ben-David et al., 2010) and meta-learning (Schmidhuber, 1987; Thrun & Pratt, 1998) .\nAmong these, meta-learning has flourished as one of the most effective methods due to its ability to leverage the knowledge learnt from many training tasks to quickly adapt to unseen tasks.\nRecent advances in meta-learning have produced state-of-the-art results in many benchmarks of few-shot learning data sets (Santoro et al., 2016; Ravi & Larochelle, 2017; Munkhdalai & Yu, 2017; Snell et al., 2017; Finn et al., 2017; Rusu et al., 2019) .\nLearning from a few examples is often difficult and easily leads to over-fitting, especially when no model uncertainty is taken into account.\nThis issue has been addressed by several recent Bayesian meta-learning approaches that incorporate model uncertainty into prediction, notably LLAMA that is based on Laplace method (Grant et al., 2018) , or PLATIPUS (Finn et al., 2017) , Amortised Meta-learner (Ravi & Beatson, 2019) and VERSA (Gordon et al., 2019 ) that use variational inference (VI).\nHowever, these works have not thoroughly investigated the generalisation errors for unseen samples, resulting in limited theoretical generalisation guarantees.\nMoreover, most of these papers are based on variational functions that may not represent well the richness of the underlying distributions.\nFor instance, a common choice for the variational function relies on the diagonal Gaussian distribution, which can potentially worsen the prediction accuracy given its limited representability.\nIn this paper, we address the two problems listed above with the following technical novelties:\n(i) derivation of a rigorous upper-bound for the generalisation errors of few-shot meta-learning using PAC-Bayes framework, and\n(ii) proposal of a novel variational Bayesian learning based on implicit\nThe few-shot meta-learning problem is modelled using a hierarchical model that learns a prior p(w i ; θ) using a few data points s\nij )}.\nShaded nodes denote observed variables, while white nodes denote hidden variables.\ngenerative models to facilitate the learning of unseen tasks.\nOur evaluation shows that the models trained with our proposed meta-learning algorithm is at the same time well calibrated and accurate, with competitive results in terms of Expected Calibration Error (ECE) and Maximum Calibration Error (MCE), while outperforming state-of-the-art methods in a few-shot classification benchmark (mini-ImageNet).\nWe introduce and formulate a new Bayesian algorithm for few-shot meta-learning.\nThe proposed algorithm, SImBa, is based on PAC-Bayes framework which theoretically guarantees prediction generalisation on unseen tasks.\nIn addition, the proposed method employs a generative approach that implicitly models the shared prior p(w i ; θ) and task-specific posterior q(w i ; λ i ), resulting in more expressive variational approximation compared to the usual diagonal Gaussian methods, such as PLATIPUS or Amortised Meta-learner (Ravi & Beatson, 2019) .\nThe uncertainty, in the form of the learnt implicit distributions, can introduce more variability into the decision made by the model, resulting in well-calibrated and highly-accurate prediction.\nThe algorithm can be combined with different base models that are trainable with gradient-based optimisation, and is applicable in regression and classification.\nWe demonstrate that the algorithm can make reasonable predictions about unseen data in a multi-modal 5-shot learning regression problem, and achieve state-of-the-art calibration and classification results with on few-shot 5-way tasks on mini-ImageNet data set.", "role": "user" }, { "content": "Bayesian meta-learning using PAC-Bayes framework and implicit prior distributions", "role": "assistant" } ]
science.scientific_lay_summarisation_plos_single_doc_summ
science.scientific_lay_summarisation_plos_single_doc_summ.12316
[ { "content": "You will be presented with the abstract, introduction, and discussion section from a biomedical research article. Your task is to create a summary that a layperson can understand, capturing the essence of the research article. Keep essential scientific terms, but ensure that the language remains clear and the concepts are explained in an uncomplicated manner.\n\nTitle: High-Resolution Functional Mapping of the Venezuelan Equine Encephalitis Virus Genome by Insertional Mutagenesis and Massively Parallel Sequencing\nArticle:\nAbstract:\nWe have developed a high-resolution genomic mapping technique that combines transposon-mediated insertional mutagenesis with either capillary electrophoresis or massively parallel sequencing to identify functionally important regions of the Venezuelan equine encephalitis virus( VEEV) genome. We initially used a capillary electrophoresis method to gain insight into the role of the VEEV nonstructural protein 3( nsP3) in viral replication. We identified several regions in nsP3 that are intolerant to small( 15 bp) insertions, and thus are presumably functionally important. We also identified nine separate regions in nsP3 that will tolerate small insertions at low temperatures( 30°C), but not at higher temperatures( 37°C, and 40°C). Because we found this method to be extremely effective at identifying temperature sensitive( ts) mutations, but limited by capillary electrophoresis capacity, we replaced the capillary electrophoresis with massively parallel sequencing and used the improved method to generate a functional map of the entire VEEV genome. We identified several hundred potential ts mutations throughout the genome and we validated several of the mutations in nsP2, nsP3, E3, E2, E1 and capsid using single-cycle growth curve experiments with virus generated through reverse genetics. We further demonstrated that two of the nsP3 ts mutants were attenuated for virulence in mice but could elicit protective immunity against challenge with wild-type VEEV. The recombinant ts mutants will be valuable tools for further studies of VEEV replication and virulence. Moreover, the method that we developed is applicable for generating such tools for any virus with a robust reverse genetics system.\nIntroduction:\nVenezuelan equine encephalitis virus( VEEV) is a New World Alphavirus endemic to regions of South America. Normally maintained in a rodent reservoir, VEEV can be transmitted by mosquitoes to horses and humans where it can cause debilitating and potentially fatal encephalitis. There are currently no vaccines for VEE licensed for use in humans. Alphaviruses contain an approximately 11–12 kb single-strand, capped and polyadenylated positive-sense RNA genome. The 5′ two-thirds of the genome encode the non-structural proteins; nsP1, nsP2, nsP3, and nsP4, which are involved in genome replication and transcription. The 3′ one-third of the genome encodes the structural proteins; capsid, E3, E2, 6K, and E1. Much of what is known about the functions of Alphavirus non-structural proteins has been elucidated through molecular and classical genetics studies of two prototypical alphaviruses: Sindbis virus( SINV), and Semliki Forest virus( SFV)( reviewed in[1],[2],[3],[4]). The non-structural proteins are initially translated as two polyproteins, P123 and P1234. In SINV, and several other alphaviruses including VEEV, the major non-structural polyprotein, P123, is produced by translation termination at an opal codon at the end of nsP3. Occasional read-through of the opal termination codon produces P1234. Cleavage of P1234 in cis by a protease activity that resides in nsP2 generates a complex of P123 and nsP4 that can initiate minus-strand RNA synthesis. P123 is cleaved into nsP1, nsP2, and nsP3, and these fully cleaved forms generate a complex with cellular proteins, replicate the full-length viral genome from minus-strand templates, and transcribe the subgenomic RNA encoding the viral structural proteins. The enzymatic activities and functional roles of nsP1, nsP2, and nsP4 have been partially characterized. nsP1 has methyltransferase and guanylyltransferase activity[5],[6],[7],[8],[9], required for capping RNA, and is necessary for synthesis of minus-strand RNA[10]. nsP2 has multiple functions in viral replication. It has RNA helicase activity[11] and NTPase activity[12],[13], and the C-terminus of nsP2 functions as a cysteine protease that cleaves the non-structural polyproteins P123 and P1234[14],[15],[16],[17]. nsP2 has been found to enter cell nuclei and to be an inhibitor of transcription of cellular messenger and ribosomal RNAs including those involved in innate immune responses[18],[19],[20] nsP4 is the RNA dependent RNA polymerase[21],[22],[23],[24]. nsP3 is the least characterized of the Alphavirus nonstructural proteins. Studies with SINV showed that nsP3 is essential for both minus-strand and plus-strand synthesis, but the precise role that it plays in these activities is unknown[25],[26],[27]. The protein consists of two domains, an N-terminal macrodomain that is highly conserved among alphaviruses and a hypervariable, phosphorylated C-terminal domain. Macro domains( also called X domains) are found in proteins from bacteria, archea, and eukaryotes, as well as several viruses including alphaviruses, coronaviruses, hepatitis E virus, and rubella virus[28],[29],[30],[31]. They bind ADP-ribose and poly( ADP-ribose) and exhibit ADP-ribose 1″-phosphatase activity. The crystal structure of the VEEV macrodomain was recently solved, revealing a conserved adenosine binding pocket[32]. nsP3 is the only Alphavirus non-structural protein that is phosphorylated[33], with most or all of the phosphorylation on serine and threonine residues in the C-terminus[34],[35]. The studies that we report here were initiated to gain insight into the function of VEEV nsP3. Toward this goal, we used transposon mutagenesis, reverse genetics, and fragment analysis by capillary electrophoresis to identify regions of the nsP3 gene that are important for replication and that result in temperature sensitive( ts) mutations. Although this work demonstrated the utility of using insertional mutagenesis for identifying such regions, the method was limited by the inherently low capacity of fragment analysis by capillary electrophoresis. Consequently, we developed a novel high-resolution functional mapping technique that couples transposon insertional mutagenesis with tag sequencing on a next-generation sequencing platform. We used transposon mutagenesis to construct a cDNA library with small DNA fragments randomly inserted throughout the VEEV genome and then produced replication-competent virus through reverse genetics. Comparing transposon insertion sites in the resultant viruses to those in the starting library, we were able to produce a functional map of the entire genome of VEEV, and to identify several hundred potential ts mutations, including those we originally identified with the capillary electrophoresis method. We further validated the mutations in cell culture assays of recombinant viruses generated through reverse genetics and tested two of the nsP3 ts mutants in mice to see if they could act as attenuated vaccines. These studies provide both new information about the association of nsP3 gene regions with virulence, as well as a rapid and effective method for identifying and creating new ts mutants for replication studies.\nDiscussion:\nThe analysis of ts mutants has been used extensively to study Alphavirus replication, and has helped to identify the activities and interactions of many viral proteins[10],[21],[37],[38],[39],[40],[41],[42],[43],[44]. We originally planned to use functional mapping to gain insight into the role( s) that nsP3 plays during viral replication simply by mapping domains that would and would not tolerate short insertions. However, we found that by performing functional mapping on virus pools that had been produced at different temperatures, we could identify large numbers of ts mutants, which could be used for more in depth studies of viral replication. Our mapping of nsP3 identified several interesting features. Not surprisingly, most of the conserved N-terminus of the protein was intolerant to short insertions while most of the variable C-terminus would tolerate insertions. Two regions in the C-terminus that would not tolerate insertions spanned nucleotides 5628–5666 and nucleotides 5684–5702. The latter region might be expected to be intolerant to insertions, as that region encodes the extreme C terminus of nsP3, including the cleavage site between nsP3 and nsP4. Disruptions in this region could prevent cleavage of nsP3 from nsP4, which is an essential step in viral replication[45]. The other site is more enigmatic. The region from 5628–5666 has no obvious homology to known regions of importance, and will require additional study to determine why insertions are not tolerated. One recent study with SFV indicated that the extreme C-terminus of nsP3 contains a degradation signal and a signal important for its' cellular localization[46]. The regions we have identified here might play a similar role in VEEV replication. In addition to these two intolerant regions, we also identified nine separate regions in the 5′ end of nsP3 that are ts for insertions. Several of these regions mapped near ts regions identified in SINV[21],[25],[26],[47], but ts regions 1, 6, 9, and 10 were in regions that were not previously identified. Construction of these ts mutants through reverse genetics will be required to more fully characterize their roles in viral replication. To generate a functional map of nsP3, we used fragment analysis by capillary electrophoresis to generate functional maps in a manner similar to one reported earlier[48],[49]. While this technique worked relatively well for analyzing short stretches of DNA or RNA( i. e., up to 2 kb), the technique had limitations that made it cumbersome for analysis of larger gene regions. For example, we could only analyze individual amplicons of approximately 600–800 bp due to the inherent loss of size resolution in larger fragments. We also needed to include a sequencing ladder in each electrophoretic analysis to obtain accurate fragment size information. Because our mapping fragments had short transposon-derived sequences appended to the 3′ end, fragment sizing with the sequencing ladder was not exact. We theorized that functional mapping by using massively parallel sequencing might overcome some of these issues. In principle, the large number of sequencing reads obtained would allow us to measure the relative frequencies of insertions at each nucleotide position in the genome. Massively parallel sequencing would also provide an exact sequence readout of the locations of inserts and generate highly accurate maps. Lastly, mapping by massively parallel sequencing would enable large stretches of DNA or RNA to be examined in a single sequencing run. For our studies, we used a Roche GS-FLX system, which was adequate for mapping the approximately 11. 5 kb VEEV genome( ∼100 K–200 K reads per sample.) Viruses with larger genomes could be mapped by using other sequencing platforms such as the Illumina Genome Analyzer or Applied Biosystems SOLiD sequencer. These platforms produce hundreds of millions of sequencing reads, and would permit the analysis of a much larger number of samples per sequencing run. Therefore, they are probably the optimal systems for this type of analysis. Although the length of sequencing reads obtained on these systems are shorter than those obtained on the GS-FLX, the 35+ bp reads that they generate should be enough to map insertion sites onto the relatively simple genomes of viruses. In our analysis of VEEV, we found that the first 20 nts of sequence was sufficient for mapping each read back to the genome. Functional mapping of the VEEV genome by using the massively parallel sequencing method that we developed revealed several hundred sites, spread throughout the genome in every gene, where the number of insertions detected at 30°C was greater than 10 times higher than the number of sequences detected at 40°C. All 10 sites that we chose for further analysis gave rise to a virus with a ts phenotype, thus it is likely that most or all of the other sites could also be reverse engineered to generate novel ts mutant viruses that could be used for functional studies of VEEV proteins. Many of the sites fell outside of the known functional domains of VEEV proteins, such as the helicase and peptidase domains of nsP2 and the RNA-dependent RNA polymerase domain of nsP4, and may represent additional domains with unknown functions in these proteins. Elucidating the mechanisms of attenuation for these mutants will be important for identifying additional functions required for viral replication. In addition to providing new ts mutants for studying viral replication, this method can also be used to map virulence properties. For example, we demonstrated one use of functional mapping data by constructing ts mutants that acted as attenuated vaccines in a mouse model of lethal VEEV infection. None of the mice vaccinated with these viruses showed any signs of disease, and most were protected from challenge with wild-type VEEV. Of the three mice that did succumb to challenge, all were in groups that were inoculated with only 10 pfu of ts mutant virus; thus, it is possible that these mice were not infected in the first place. This is supported by the observation that one mouse in each group did not develop neutralizing antibodies after infection with the ts mutants. The utility of this technique for designing ts vaccines lies not only in the ability to detect large numbers of ts mutants, but also in the ability to multiplex those mutations to generate viruses with a potentially more stable ts phenotype. In our study, we demonstrated that we could combine two mutants in our double ts mutant and still recover viable ts virus. It might be possible to combine more than two mutations, with each additional mutation reducing the reversion potential of the final attenuated virus. Finally, although we only presented functional mapping data relating to ts phenotypes, functional mapping of other phenotypes is also possible using this technique. For example, comparison of functional maps generated from RNA isolated from infected cells to vRNAs isolated from the supernatant of infected cells might identify mutants defective for particle formation. As an additional example, we have performed preliminary functional mapping of VEEV propagated in mosquito cells( C6/36) to identify insertions that confer a species-specific replication phenotype, and have identified several insertions in the VEEV genome that appear to allow virus to replicate in Vero cells but not in C6/36 cells, and vice versa. Similar studies could be performed to examine animal-, organ-, or cell-specific replication characteristics of viruses. This method should provide a powerful and new means to generate tools for studying a myriad of characteristics of any virus with a robust reverse genetics system.", "role": "user" }, { "content": "Venezuelan equine encephalitis virus ( VEEV ) is a New World Alphavirus that was first identified in Venezuela in 1938 . VEEV normally circulates in rodent populations , but during outbreaks it can jump to horses and humans where it can cause debilitating and potentially fatal disease . There are currently no vaccines or antiviral agents against VEEV licensed for use in humans . In this study , we describe a technique that we have developed that allows for the rapid identification of viral mutants that can be useful for studying the basic biology of viral replication . These mutants can also be used to generate vaccines that protect against infection with wild-type virus . We demonstrate the utility of this technique by identifying over 200 mutations spread throughout VEEV genome that make the virus unable to replicate efficiently at higher temperatures ( 37°C or 40°C . ) Furthermore , we show that two of the mutant viruses work as vaccines , and protect mice against lethal infection with VEEV . This technique can be applied to studying other viruses , and may allow for the rapid identification of numerous vaccine candidates .", "role": "assistant" } ]
science.covidfact_entailment
science.covidfact_entailment.142
[ { "content": "You will be shown a claim related to the COVID-19 pandemic, and the abstract of a biomedical research paper. Each sentence from the abstract will be on a separate line. Your task is to return a JSON object with two fields:\n\n- \"verdict\": The fact-checking verdict. If the information in the abstract supports the claim, write \"SUPPORT\". If the abstract contradicts the claim, write \"CONTRADICT\".\n- \"evidence\": An array of sentences providing evidence for the verdict. Please copy all relevant sentences verbatim from the abstract.\n\nFor instance, if the model were given the claim \"the COVID vaccine is safe for healthy adults\", the output might be:\n{\n \"verdict\": \"SUPPORT\",\n \"evidence\": [\"Our findings indicate that the risks of severe COVID vaccine side effects are low for healthy adults.\"]\n}\n\nClaim: Sars-cov-2 antibody responses correlate with resolution of rnaemia but are over-lived in patients with mild illness\n\nAbstract:\nSARS-CoV-2-specific antibodies, particularly those preventing viral spike receptor binding domain (RBD) interaction with host angiotensin-converting enzyme 2 (ACE2) receptor, could offer protective immunity, and may affect clinical outcomes of COVID-19 patients.\nWe analyzed 625 serial plasma samples from 40 hospitalized COVID-19 patients and 170 SARS-CoV-2-infected outpatients and asymptomatic individuals.\nSeverely ill patients developed significantly higher SARS-CoV-2-specific antibody responses than outpatients and asymptomatic individuals.\nThe development of plasma antibodies was correlated with decreases in viral RNAemia, consistent with potential humoral immune clearance of virus.\nUsing a novel competition ELISA, we detected antibodies blocking RBD-ACE2 interactions in 68% of inpatients and 40% of outpatients tested.\nCross-reactive antibodies recognizing SARS-CoV RBD were found almost exclusively in hospitalized patients.\nOutpatient and asymptomatic individuals’ serological responses to SARS-CoV-2 decreased within 2 months, suggesting that humoral protection may be short-lived.", "role": "user" }, { "content": "{\"verdict\": \"CONTRADICT\", \"evidence\": [\"Severely ill patients developed significantly higher SARS-CoV-2-specific antibody responses than outpatients and asymptomatic individuals.\"]}", "role": "assistant" } ]