Title
stringlengths 3
331
| text
stringlengths 14
9.14k
⌀ |
|---|---|
Transdifferentiation of fibroblasts to neural stem cells | The developmental process is carefully controlled by transcriptional and epigenetic changes that occur as a zygote transforms into an adult organism. This process can be reversed by the overexpression of transcription factors Oct4, Sox2, Klf4, and c-Myc, which reprogram a differentiated cell!s nucleus to one that is transcriptionally and epigenetically indistinguishable from an embryonic stem (ES) cell. However, it is still unclear if transcription factors can completely convert the nucleus of a differentiated cell into that of a distantly related somatic cell type with complete transcriptional and epigenetic reprogramming maintained in the absence of exogenous factor expression. To test this idea, we generated doxycyline (dox)-inducible vectors encoding neural stem cell-expressed factors. We found that stable, self-maintaining NSC-like cells could be induced under defined growth conditions. These cells were characterized in the absence of exogenous factor induction and were shown to be transcriptionally, epigenetically, and functionally similar to endogenous embryonic cortical NSCs. Additionally, a cellular system was created for reproducible generation of doxindependent iNSCs without additional factor transduction. Our results show that a transcriptionally and epigenetically reprogrammed somatic nucleus can be stabilized in vitro and provides a tool to study the mechanism of somatic cell conversion. |
Single-dose growth factor treatments to enhance cell recruitment and neotissue integration in an augmented microfracture cartilage repair model | Focal cartilage defects caused by joint injury have a limited capacity to self-repair, and if left untreated, can lead to the early onset of osteoarthritis. The current gold standard of care, microfracture surgery, induces an endogenous repair response, but typically results in poorly integrated fibrocartilage, rather than native hyaline cartilage. The objective of this thesis was to test the hypothesis that a self-assembling peptide hydrogel functionalized with chemotactic and pro-anabolic growth factors and placed into the defect during surgery could induce migration of endogenous progenitor cells into the repair tissue. Since these progenitors are naturally accessed during microfracture surgery, clinical translation of this approach could ultimately steer repair to a more hyaline-like response. Poor cartilage repair is believed to be the result of an insufficient number of progenitor cells at the defect site. We hypothesized that the addition of a single dose combination of chemotactic growth factors (such as platelet derived growth factor-BB (PDGF-BB), transforming growth factor-P 1 (TGF-[beta]1), and heparin-binding IGF-l (HB-IGF-1)) premixed into a hydrogel scaffold could stimulate bone-marrow progenitor cell migration into the hydrogel. A novel 3D gel-to-gel migration assay, using (KLDL)₃ self-assembling peptide gels, demonstrated that the combination of PDGF-BB and TGF-[beta]1 induced significant migration into the gel compared to growth-factor free controls. Importantly, these growth factors were retained in the hydrogel and exhibited a slow release over 1-2 weeks. We also hypothesized that a brief enzymatic pre-treatment of the defect site could release proteoglycans from the walls of the surrounding native cartilage in a controlled manner, and thereby create space for newly synthesized repair tissue to anchor and integrate with this adjacent host cartilage. We used an in vitro model in which a cylindrical annulus of native cartilage was pre-treated with trypsin over a 2-minute period and then filled with a chondrocyte-seeded (KLDL) ³ hydrogel ftnctionalized with pro-anabolic HB-IGF-I that had been premixed into the gel. (This procedure was deemed to be clinically tractable by collaborating equine surgeons now using this approach in parallel animal studies.) Trypsin pre-treatment depleted proteoglycan content of adjacent cartilage in a controlled manner, and HB-IGF-l was found to be delivered to the surrounding cartilage from the peptide gel. HB-IGF-I was found to stimulate matrix biosynthesis both in the surrounding cartilage and the chondrocyte-seeded KLD scaffold, and to enhance mechanical integration. We further explored the uptake and diffusive transport properties of HB-IGF-l into cartilage motivated by the need to understand the pharmacokinetics of delivery from the repair construct to surrounding cartilage. The positively charged heparin-binding domain of HB-IGF-l results in high uptake into cartilage, making it an effective method of delivering the pro-anabolic attributes of IGF-1, which in its native form would be rapidly cleared from the joint. The observed high and rapid uptake of HB-IGF-l into cartilage will enable characterization of dosing for HB-IGF-l delivery to cartilage by either intra-articular injection or from implanted hydrogel scaffolds. In summary our results show that of (KLDL)₃ peptide hydrogel scaffolds can foster growth-factor induced progenitor cell migration to increase progenitor cell recruitment into a cartilage defect. Thus, the use of a peptide gel premixed with PDGF-BB and HB-IGF-l to enhance progenitor migration into the scaffold, combined with a trypsin pre-treatment to help promote subsequent integration, is a promising strategy towards improving integrative repair. The combination of these approaches is currently being tested in an in vivo rabbit model. |
Process and politics in interactive musical works | As everyday musical experiences move further into software platforms, an interest among musicians in taking fuller advantage of computational media produces a strand of interactive, software-based musical works I call open mediational music. This phenomenon stands apart from other types of creative work centered on music and interaction by valorizing the listener's responsibility for instantiating musical works. It also advances an agenda of openness with respect to interactivity as a principle of new media. I center four case studies on a set of interactive musical works that exemplify this phenomenon: Reflective by Reiko Yamada, Thicket by Morgan Packard and Joshue Ott, Jazz. Computer by Yotam Mann and Sarah Rothberg, and Baggage Allowance by Pamela Z. Each of these works takes shape out of unique motivations and in different forms and settings. Collectively, they advance a notion of platforms as objects of critical awareness and propose listening as a model for mindful participation in algorithmic environments. Illuminating the distinct claims that sound and software hold on one another as creative domains, open mediational music invites listeners to rehearse a conscientious engagement with the sites and conditions of computationally mediated cultural encounter. |
Ensuring sustainable employment in the US manufacturing industry by leveraging demand proximity | This thesis investigates the theme of "manufacturing offshoring" that became a political issue during the 2004 U.S. presidential election. As during previous elections, employment became a key focus on the home front. Whereas the 1992 presidential election was marked by the debate around the loss of manufacturing jobs due to NAFTA, the 2004 election focused on both manufacturing and services jobs lost due to offshoring to low cost countries. For the first time, well paying jobs, such as IT programming, were outsourced to emerging countries like India or China. Offshoring of "white collar" jobs became the focus of academic, consultant, and journalist discussions in U.S., whereas offshoring of manufacturing activities generated comparatively less interest. For decades offshoring of manufacturing activities to low cost countries was used by American companies to either reduce production cost or to avoid high tariffs on exports. Offshoring strategy was historically applied in labor intensive industries, such as the apparel and electronics sectors. On the other hand, the influence of offshoring on high-tech industries was assumed to be limited. However, with the fast technological development of China and India, this paradigm might change quickly. |
A mapping system for an autonomous helicopter | Acknowledgments: This thesis was prepared at The Charles Stark Draper Laboratory, Inc., under Internal Research & Development No. 18598. Publication of this thesis does not constitute approval by Draper or the sponsoring agency of the findings or conclusions contained herein. It is published for the exchange and stimulation of ideas. The work that is described in this document was by no means done all by myself. There are many people that I would like to thank for their support of my education here at MIT and Draper Laboratory during the past few years. I would like to thank the members of the helicopter team, who have greatly influenced my development as an engineer: Paul Debitetto, Christian Trott, Bob Butler, Long Phan, Mike Piedmonte, and Anthony Lorusso. Special thanks goes to Paul Debitetto for his thoughtful critiquing and leadership. Long Phan was my close partner in the design of the scanning laser rangefinder, and this thesis would not have been possible without his inspiration and expertise. During my work on the mapping system, a number of Draper staff took time out of their own busy schedules to help me. I am especially indebted to Chris Sanders, Chris Smith, John Plump. Linda Leonard, John Danis, and Dave Hauger for their patience and helpfulness over the last year. I would also like to thank my MIT advisor, Seth Teller, for helping me to graduate this year. Other Draper fellows and students in the autonomous vehicle lab have also provided me with both technical expertise and encouragement. I would like to thank Mohan Gurunathan, Jonah Peskin, and Bill Kaliardos for their advice and moral support. Finally, I would like to thank my entire family for years of encouragement and support. It is to my mother, father, and sister that I dedicate this thesis. Rusty Sammon May, 1999. |
Power extraction from an oscillating water column along a coast | For reasons of wave climate, geography, construction, maintenance, energy storage and transmission, some devices for extracting energy from sea waves will likely be installed on the coast. We study here the specific case where an Oscillating Water Column (OWC) is attached to the tip of a long breakwater. A three-dimensional numerical model of a skeletal geometry of the the Foz do Douro breakwater is developed in order to determine the response inside the OWC pneumatic chamber to incident waves and assess the possible effects of the breakwater geometry. The model uses the hybrid element method and linear water wave theory. Then, a more analytical approach for a simplified geometry is presented. Making use of an exact solution for the scattering by a solid cylinder connected to a wedge, we solve for the linearized problems of radiation and scattering for a hollow cylinder with an open bottom. Power-takeoff by Wells turbines above an air chamber is modeled by including the compressibility of air. It is shown for the case of a circular OWC attached to a thin breakwater, that the incidence angle affects only the waves in and outside the column but not the power extraction which depends only on the averaged water-surface displacement inside. Optimization by controlling the turbine characteristics is examined for a wide range of wavelengths. Finally, the same approach is used to solve the case of an OWC positioned along a straight coast line. It is found that in this configuration, the extracted power does depend on the incidence angle. It is also shown that the average efficiency is doubled compared to the thin breakwater geometry. |
Parallel pulsed jets for precise underwater propulsion | A significant limitation for underwater robots is their ability to maneuver in tight spaces or for complex tracking tasks. Next generation vehicles require thrusters that can overcome this problem and efficiently provide high maneuverability at low speeds. Recently, thruster design has begun to draw inspiration from nature's swimmers, applying the principles of pulsed jet propulsion to robotic thrusters. Although most developments have focused on single jet actuators, nature provides some indications that multi-jet systems can provide propulsive benefits -- marine invertebrates called sales connect into chains of individual animals that each eject short jets to collaboratively move the entire chain efficiently around the ocean. However, despite the promise of multi-jet propulsion, there are no existing models or empirical data that explain the physics of multi-jet propulsion. As a result, there are no physically motivated rules to guide the design of man-made multi-jet thrusters. In this thesis, I experimentally investigate how interactions between neighboring jets in a multi-jet thruster will affect the system's propulsive performance. I use high-speed fluorescence imaging to investigate the mutual influence of two pulsed jets under conditions relevant to low-speed maneuvering in a vehicle (Re ~ 350). Using a new force estimation technique developed in this thesis, I analyze the video data to evaluate how thrust and efficiency are affected by the jet spacing. This analysis reveals that, compared to non-interacting jets, the efficiency and thrust generated by the pair of interacting jets can fall by nearly 10% as the jets are brought into close proximity. Based on this data, I develop a model of vortex interactions to explain the thrust and efficiency drop. The data and model described in this thesis contribute new insights to understand vortex formation in pulsed jets, and these results can be used to guide the design of multi-jet underwater propulsion syste |
Using sentiment and social network analyses to predict opening-movie box-office success | In this thesis, we explore notions of collective intelligence in the form of web metrics, social network analysis and sentiment analysis to predict the box-office income of movies. Successful prediction techniques would be advantageous for those in the movie industry to gauge their likely return and adjust pre- and post-release marketing efforts. Additionally, the approaches in this thesis may also be applied to other markets for prediction as well. We explore several modeling approaches to predict performance on the Hollywood Stock Exchange (HSX) prediction market as well as overall gross income. Some models use only a single movie's data to predict its future success, while other models build from the data of all the movies together. The most successful model presented in this thesis improves on HSX and provides high correlations/low predictive error on both HSX delist prices as well as the final gross income of the movies. We also provide insights for future work to build on this thesis to potentially uncover movies that perform exceptionally poorly or exceptionally well. |
Bi-modal hemispherical sensors for three axis force and contact angle measurement | Humans and animals demonstrate a unique ability to interact with the physical environment around them with coordination and control, including moving quickly across rugged terrain or deftly handling small objects. Much of this success is related to our ability to accurately perceive the world around us through a sense of touch. In order to better perform dynamic, physical interactions, such as locomotion or manipulation, robots need to be able to accurately measure contact locations and forces. However, many existing sensors do not satisfy the stringent requirements or do not supply sufficient information for robotic locomotion and manipulation. This thesis builds on work by a previous PhD student, Meng Yee (Michael) Chuah, of the MIT Biomimetic Robotics Laboratory to develop stress field based force sensors for use in robotic applications. The concept of stress field based force sensing consists of pressure sensors embedded within a rubber hemisphere. |
Establishing Transit-Oriented Development (TOD) on the ground : case-based analysis of implementing TOD in China | While the study of Transit-Oriented Development (TOD) in China has been under way since the 1990s, effective examples of TOD in practice remain scarce. This research conducted 41 semi-structured interviews and applied the Policy-Program-Implementation Process (PPIP) model to identify the challenges to TOD implementation-which are found throughout the entire process, from general acceptance, to urban planning capacity and regulations, to government administration and the real estate market. Misunderstanding and incomplete understanding prevent people from fully adopting the idea of TOD. In planning and design, main challenges are in the form of conflicts between current planning regulations and TOD planning techniques. Regarding the public sector, institutional coordination will not become a major barrier once a strong leadership in the municipality is in support of TOD; but there are still barriers in the administrative regulations and financing mechanisms. In the real estate market, great challenges come from developers who hesitate to take the risk to introduce the new type of TOD product. In order to establish successful TODs on the ground in Chinese cities, this research recommends the following solutions. First, effective TOD training programs need to be carried out. Second, the planning codes should be reviewed and revised to establish the legitimacy of TOD. Third, economic incentives and ongoing revenue resources for the government are needed to reassure real estate developers and also provide sustainable economic solutions for the government. Fourth, institutional structure needs to be developed in both administrative and development processes to reduce the institutional barriers and internalize the economic benefits in the TOD implementation. |
Optical breakdown acoustics : transduction and sensing underwater | In the sea, infrastructures such as ships, pipelines, and wind turbines are exposed to harsh conditions that can wear down the structures through wave loading and corrosion. Because of these wear mechanisms, maritime structures require regular inspections to identify early signs of damage or fatigue. Currently, inspections are performed visually or with contact acoustic transducers, often by a human diver. However, these methods are slow and costly, and can be hindered by surface irregularities like biofouling. Therefore, new sensing techniques are needed to meet the rising demand for offshore infrastructure monitoring. In this thesis, I develop optical breakdown as an acoustic source for non-contact measurements of underwater structures. Optical breakdown occurs when a high-power laser is focused to a small spot, causing nonlinear interactions between the light and water. A compact plasma forms at the focus and expands explosively, radiating a loud, broadband pressure wave. |
Clustered Regularly Interspaced Short Palindromic Repeats-mediated interrogation of small cell lung cancer | Small cell lung cancer (SCLC) is a highly aggressive neuroendocrine lung carcinoma that remains among the most lethal of solid tumor malignancies. Despite decades of research, treatment outcomes for SCLC remain very poor, highlighting the need for novel approaches to target the disease. Recent genomic sequencing studies have identified multiple recurrently altered genes in human SCLC tumors, many of which remain to be functionally validated. Genetically engineered mouse models (GEMMs) of SCLC have been developed that recapitulate many key features of human SCLC. These models have been used extensively to investigate various aspects of SCLC biology, including tumor initiation, progression and metastasis. The development of the CRISPR-Cas9 system has greatly facilitated genome editing in mammalian cells, leading to its widespread adoption for various applications in cancer biology. We have utilized this system in two complementary ways to investigate the molecular mechanisms involved in SCLC initiation, progression and maintenance. Firstly, we have adapted the CRISPR-Cas9 system for use in GEMMs of SCLC, to enable rapid modeling and functional validation of candidate tumor suppressor genes in vivo. Using this system, we have demonstrated that p107, a member of the retinoblastoma family that is mutated in a significant fraction of human SCLC tumors, is a functional tumor suppressor in SCLC. Notably, loss of p107 in SCLC tumors resulted in significant phenotypic differences compared with loss of its close relative, p130. We also demonstrated that CRISPR-induced mutations can be used to infer lineage relationships between primary and metastatic tumors in the same animal. Secondly, we have performed a CRISPR-based genetic screen, utilizing a custom sgRNA library targeting the druggable genome, to identify novel SCLC-specific genetic vulnerabilities. We found that SCLC cells displayed enhanced sensitivity towards disruption of several key metabolic pathways, including the de novo pyrimidine biosynthesis pathway. Pharmacological inhibition of Dhodh, a key enzyme in this pathway, reduced the viability of SCLC cells in vitro and strongly suppressed SCLC tumor growth in vivo, validating this pathway as a promising therapeutic target in SCLC. Taken together, the work presented here demonstrates the utility of the CRISPR-Cas9 system for performing functional interrogation of SCLC. |
Potential for meeting the European Union new passenger car CO₂ emissions targets | In 2009, the European Parliament agreed to limit the CO2 emissions from new passenger cars sold in the European Union to an average of 130g/km by 2015. Further, a probable longer-term CO2 emissions target of 95g/km is specified for 2020. This thesis attempts to assess the feasibility of meeting these targets in a representative European Union by developing and evaluating Optimistic and Realistic scenarios of varied powertrain sales mix, vehicle weight reduction levels, and Emphasis on Reduction of Fuel Consumption (ERFC) using a European New Passenger Cars CO2 Emissions Model. Further, this thesis develops custom fleet models for select member states to understand the impact of the developed scenarios on reduction of fuel use and on the diesel to gasoline fuel use ratio. The thesis finds that while the European Union is poised to meet the 2015 target in an Optimistic scenario, it will find it difficult to do so in a Realistic scenario. Moreover, the 2020 target would not be achieved in either of the two scenarios. Further, the diesel to gasoline fuel use ratio will continue to rise through year 2020 for the studied countries, potentially reaching as high as 3 in the case of France and at least as high as 0.71 in the case of Germany. Finally, an increase in ERFC and introduction of PHEVs would most help reduce fuel use in all studied countries. In France and Italy, a reduction of Diesel car sales would additionally be significantly useful in reducing the fuel use. Whereas, in Germany and UK, a higher number of Turbocharged Gasoline cars would be another significant option to reduce fuel use. |
How network structure impacts socially reinforced diffusion? | Social scientists have long studied adoption choices that depend on the number of prior adopters. What is the effect of network structure on such adoption dynamics? The emerging consensus holds that when agents require a high reinforcement threshold for adoption, clustered networks are better conduits of social contagion than random ones. Using models with deterministic thresholds this argument formalizes the idea that transmission will get 'stuck' should the number of neighboring adopters fall below a threshold. In this paper, we explore the effect of stochastic thresholds on the diffusion races between random and clustered networks. We show that even low probabilities of adoption upon a single contact would tilt the balance in favor of random networks, a tendency that is reinforced with the size of the network. Moreover, if repeated signals from the same adopter can reinforce a message, random networks are further promoted. However, we also show that clustered networks can still be preferred over random networks if adopters become 'inactive' - i.e. they stop sending messages - with high probability. These findings refocus our theoretical understanding of how network structure moderates social influence, and raises new questions on contagion phenomena that benefit from clustered networks. |
Full wavefield inversion methods for monitoring time-lapse subsurface velocity changes | Quantitative measurements of seismic velocity changes from time-lapse seismic experiments provide dynamic information about the subsurface that improves the understanding of the geology and reservoir properties. In this thesis, we propose to achieve the quantitative analysis using full wavefield inversion methods which are robust in complex geology. We developed several methodologies in both the data domain and image domain to handle different time-lapse seismic acquisitions. In the data domain, we implemented double-difference waveform inversion (DDWI), and investigated its robustness and feasibility with realistic acquisition non-repeatabilities. Well-repeated time-lapse surveys from Valhall in the North Sea are used to compare DDWI and conventional time-lapse full waveform inversion (FWI) schemes. An FWI approach that uses the baseline and monitor datasets in an alternating manner is proposed to handle time-lapse surveys without restrictions on geometry repeatability, and to provide an uncertainty analysis on the time-lapse changes. In the image domain, we propose time-lapse image domain wavefield tomography (IDWT) that inverts for P- and S-wave velocity changes by matching baseline and monitor images produced with small offset reflection surveys. This method is robust to survey geometry non-repeatabilities and baseline velocity errors. A low velocity zone caused by local CO2 injections in SACROC, West Texas is found by IDWT with time-lapse walkaway vertical seismic profile surveys. The methods in this thesis combined, allow for an integrated velocity inversion to achieve high-resolution subsurface monitoring with various types of acquisitions in complex geology. |
A computational memory and processing model for prosody | This thesis links processing in working memory to prosody in speech, and links different working memory capacities to different prosodic styles. It provides a causal account of prosodic differences and an architecture for reproducing them in synthesized speech. The implemented system mediates text-based information through a model of attention and working memory. The main simulation parameter of the memory model quantifies recall. Changing its value changes what counts as given and new information in a text, and therefore determines the intonation with which the text is uttered. Other aspects of search and storage in the memory model are mapped to the remainder of the continuous and categorical features of pitch and timing, producing prosody in three different styles: for small recall values, the exaggerated and sing-song melodies of children's speech; for mid-range values, an adult expressive style; for the largest values, the prosody of a speaker who is familiar with the text, and at times sounds bored or irritated. In addition, because the storage procedure is stochastic, the prosody from simulation to simulation varies, even for identical control parameters. As with with human speech, no two renditions are alike. Informal feedback indicates that the stylistic differences are recognizable and that the prosody is improved over current offerings. A comparison with natural data shows clear and predictable trends although not at significance. However, a comparison within the natural data also did not produce results at significance. One practical contribution of this work is a text mark-up schema consisting of relational annotations to grammatical structures. Another is the product - varied and plausible prosody in synthesized speech. The main theoretical contribution is to show that resource-bound cognitive activity has prosodic correlates, thus providing a rationale for the individual and stylistic differences in melody and rhythm that are ubiquitous in human speech. |
ZEB1 is a central mediator of the Epithelial-Mesenchymal Transition | Carcinomas are solid tumors arising from epithelial tissue, and account for the majority of cancer deaths in the United States. In most occurrences of carcinoma, it is the metastases that kill, not the primary tumor. The Epithelial-Mesenchymal Transition (EMT) provides a model by which tightly associated epithelial cancer cells can disseminate to distant sites. Many factors are known to trigger the EMT, but the extent to which the observed phenotypes represent a common process is unknown. There is also little appreciation of the extent to which EMT-inducing factors interact with one another or act on common or redundant pathways. In this study, I sought a common gene expression signature of the EMT by comparing five mesenchymal cell lines independently derived from the same parental epithelial line using different EMT-inducing factors. The resultant EMT core signature strongly suggested a common pathway is involved. Bioinformatics analysis revealed the transcription factor ZEBI to be a possible mediator of this common pathway. ZEB1 was found to be both sufficient to induce EMT and necessary for maintaining the mesenchymal phenotype in the same cells. ZEBI and miR-200 were known to reciprocally regulate each other, but their relative importance to the EMT phenotype had never been directly tested. I found that ZEB1 induced EMT regardless of miR-200c levels, thereby excluding the model in which miR-200c downregulation is a necessary step for the EMT. I also show evidence that EMT induced by the transcription factor Snail works at least in part through ZEB1. |
A low-cost automated micromanipulator | The automated micromanipulator, a device widely used in the life sciences, allows precise three-dimensional positioning of tools and equipment with resolutions at or well below one micrometer, providing cellular-level movement resolution in a highly-controlled manner. The cost of a state-of-the- art electromechanical micromanipulator can be upwards of ten thousand dollars for a complete system, and can be a major, if not limiting, expense for labs, startups, and even hobbyists in these fields. The objective of this project is to build a proof-of-concept micromanipulation device that provides similar levels of performance, but at a price in the range of several hundred dollars. We intend to achieve this by utilizing equipment made recently and inexpensively available at low-costs because of the hobbyist 3D-printing market, as well as advances in piezoelectrics and imaging technologies from the last five years. By accomplishing this, the barrier of entry for work requiring sub-micron measurement will be lowered significantly. |
Identifying and modeling unwanted traffic on the Internet | Accurate models of Internet traffic are important for successful testing of devices that provide network security. However, with the growth of the Internet. it has become increasingly difficult to develop and maintain accurate traffic models. While much internet traffic is legitimate, productive communications between users and services, a significant portion of Internet traffic is the result of unwanted messages sent to IP addresses without regard as to whether there is an active host at that address. In an effort to analyze unwanted traffic, tools were developed that generate statistics and plots on captured unwanted traffic to unused IP addresses. These tools were used on a four-day period of traffic received on an inactive IPv4 class A network address space. Each class B subnet in this address space received an average of 7 million packets corresponding to 21 packets per second. Analyses were performed on a range of class B and C subnets with the intent of discovering the types of variability that are characteristic of unwanted traffic. Traffic volume over time, number of scans, destinations ports, and traffic sources varied substantially across class B and C subnets. |
Satellite-based estimates of net radiation and modeling the role of topography and vegetation on inter-annual hydro-climatology | The Fourth Assessment Report of the Intergovernmental Panel on Climate Change acknowledged that the lack of relevant observations in various regions of the world is a crucial gap in understanding and modeling impacts of climate change related to hydrologic cycle. The Surface Radiation Budget (SRB) is an important component in the study of land surface processes. Existing SRB retrieval algorithms generally suffer from two major shortcomings: difficulty in dealing with cloudy sky conditions and reliance on study-site specific ancillary ground data. In this work, a framework of estimating net radiation from the MODerateresolution Imaging Spectroradiometer (MODIS) data is presented that is applicable under all-sky conditions, while solely relying on satellite data. The results from the proposed methodology are compared against several ground measurements within the United States for the entire 2006. Finally, monthly radiation maps for the Continental United States are produced. Modeling, similar to observations, is critical to the Earth Sciences and the second part of this work focuses on the impact of incorporating vegetation dynamics and topography in modeling hydro-climatology over large river basins. Land and atmosphere are coupled with each other through the exchange of heat, momentum and water at the boundary. This work involves coupling of a physically-based, fully distributed ecohydrology model with a numerical atmospheric model, using high performance computing. The ability of the ecohydrology model (in an offline mode) to accurately resolve hydro-climatic signatures and vegetation dynamics is first examined. The ecohydrology model is applied in a highly instrumented catchment, Walnut Gulch Experimental Watershed (WGEW) in Arizona, for a period of 11-years (1997-2007). The ecohydrology model is able to capture the behavior of several key hydrologic variables and vegetation dynamics within the WGEW. A series of three synthetic experiments are conducted with a coupled land-atmosphere model. The anomalies of various simulated quantities between the synthetic experiments are examined within the rainfall-soil moisture feedback hypothesis proposed by Elathir [1998]. The results from the experiments highlight the need to explicitly account for vegetation dynamics and topography within a numerical atmospheric model. The thesis concludes with a discussion of contributions, and future directions for this work. |
On deploying succinct zero-knowledge proofs | Zero-knowledge proofs, introduced by Goldwasser, Micali, and Rackoff, are a fundamental building block in theoretical cryptography with numerous applications. Still, the impact of zero-knowledge proofs for building secure systems in practice has been modest at best. Part of this can be explained by the economics of deploying new technology in the wild: often introducing a trusted third party in lieu of a proof system achieves users' security goals with lower anticipated cost. The goal of this thesis is to lower the cost of using zero-knowledge proofs in real-world systems. This cost has two major components: the cost incurred by the proof system itself, and the price paid to instantiate the security model the proof system relies on. Working with my collaborators, I have contributed to reducing both of these costs: -- Cost of the security model. For many practical scenarios it is crucial that proofs be non-interactive and succinct. In the standard model, non-interactive zero-knowledge (NIZK) proofs do not exist for languages outside BPP (even with just computational soundness). However, if the security model includes a trusted party, available for a one-time setup phase, then NIZKs exist for all languages in N P. Soundness of the NIZK depends on this trusted setup: if public parameters are not correctly generated, or if the trusted party's secret internal randomness is revealed, an attacker could convince the verifier of false N P statements without being detected. We show how public parameters for a class of NIZKs can be generated by a concretely-efficient multi-party protocol, such that if at least one of the parties is honest, then the result is secure and can be subsequently used for generating and verifying numerous proofs without any further trust. -- Cost of the proof system. We have designed and built an open-source cryptographic library, called libsnark, that provides efficient implementations of state-of-the-art zero-knowledge proof constructions. Our library is the fastest and most comprehensive suite of zero-knowledge proofs currently available. Working in tandem, these contributions have achieved industrial impact, and are the main efficiency enablers for Zerocash, a privacy-preserving payment system. |
Towards a more ethical animal model in biomedical research | Since the early twentieth century, mice have emerged as the standard mammalian model organism for biomedical research. When pain relief is provided during experimentation, it typically comes in the form of transient and sometimes ineffective analgesics or anesthesia. This thesis proposes an alternative to the current method of research in the form of an engineered mouse model in which pain sensing can be ablated before an experiment. An ERT2-inducible Cre recombinase under the Wntl promoter was designed to be combined with a floxed Nav1.7 ion channel mouse model. When a 4- hydrotamoxifan class small molecule is fed to the mouse, Cre recombinase expression in the peripheral nervous system will disrupt function of the ion channel involved in inflammatory and mechanosensory pain. Additional designs for floxed Nav1.6 ion channel and Nax ion-like channel were made to explore disruption of peripheral cancer-induced neuropathic pain. In parallel with mouse model development, a survey was conducted to understand the potential for adoption of this new animal model by researchers. The survey was sent to IACUC members questioning if this model was needed, as well as how it may be regulated under the existing protocol approval framework. Results indicated that there is a both a need and desire for further refinement strategies within animal research, and that this inducible painfree mouse model could be categorized as alternative analgesic upon sufficient characterization and peer-reviewed publications. Additional input was provided that will shape testing done on the generated animals to assure that this model can mitigate animal suffering while still recapitulating important biological processes investigated in biomedical research. |
Automated creation of Wikipedia articles | This thesis describes an automatic approach for producing Wikipedia articles. The wealth of information present on the Internet is currently untapped for many topics of secondary concern. Creating articles requires a great deal of time spent collecting information and editing. This thesis presents a solution. The proposed algorithm creates a new article by querying the Internet, selecting relevant excerpts from the search results, and synthesizing the best excerpts into a coherent document. This work builds on previous work in document summarization, web question answering, and Integer Linear Programming. At the core of our approach is a method for using existing human-authored Wikipedia articles to learn a content selection mechanism. Articles in the same category often present similar types of information; we can leverage this to create content templates for new articles. Once a template has been created, we use classification and clustering techniques to select a single best excerpt for each section. Finally, we use Integer Linear Programming techniques to eliminate any redundancy over the complete article. We evaluate our system for both individual sections and complete articles, using both human and automatic evaluation methods. The results indicate that articles created by our system are close to human-authored Wikipedia entries in quality of content selection. We show that both human and automatic evaluation metrics are in agreement; therefore, automatic methods are a reasonable evaluation tool for this task. We also empirically demonstrate that explicit modeling of content structure is essential for improving the quality of an automatically-produced article. |
The development of steady State and lifetime fluorescence instruments for real time in situ aquatic chemistry measurements | The development of three optical instruments for the chemical exploration and characterization of natural waters is described. The first instrument (called LEDIF) employs a novel flowcell, 6 UV LEDs as excitation sources, a wideband lamp, and a spectrometer to measure steady state chemical fluorescence and absorbance. The instrument is packaged aboard an autonomous underwater vehicle (AUV) and demonstrates the ability to map chemical concentrations in three dimensions. The second instrument repackages the sensor components to study dissolved organic matter (DOM) in tropical Southeast Asia peatland rainforests. This instrument is optimized for low power consumption over long deployments to remote locations. Two field trials in Pontianak Indonesia with durations of two and six weeks captured peatland river fluorescence measurements at 20 minute intervals. The results show changes in DOM linked to tidally induced water level fluctuations and provide insight into the complex biogeochemical dynamics of the system. The third instrument increases the chemical sensitivity and specificity of LEDIF with the addition of fluorescence lifetime sensing capabilities. The development of this sensor for AUV deployment required the engineering of a compact, low power, high speed (GHz) data acquisition circuit board. The resulting circuit digitizes data at a rate of 1 gigasample/second and performs user customizable digital signal processing. This board is used along with a 266 nm Q-switch laser, fast photomultiplier tube (PMT), and computer controlled monochromator to build a small fluorescence lifetime instrument. The instrument is tested with solutions of low concentration pyrene to demonstrate its ability to identify small, long-lived fluorescence signals in the presence of large background fluorescence. Results indicate a pyrene limit of detection below environmentally relevant levels. The final overall instrument dimensions allows it to be packaged for future AUV deployments |
Why aren't we looking for lessons about producer-buyer relationships and niche markets in non-export cases? : farmers and wholesalers in El Salvador's central wholesale market | Two trends in the development field slow resolving small farmers' difficult entry into marketplaces. The first is that in seeking to enhance small farmer competitiveness, analysts and policy makers tend to look towards production rather than marketing improvements, missing growth linkages to other sectors in which farmers become embedded through their marketing activities. The second is that where marketing is considered, there tends to be a disproportionate focus on non-traditional crops grown for specialty export markets. This paper analyzes how small farmers from the Las Pilas region of El Salvador competed against imports to become potato, cabbage and tomato suppliers to wholesalers in the national wholesale market, the Tiendona. Enduring relationships with buyers in the wholesale market created opportunities and pressures for Las Pilas farmers to market and produce a diversity of crops over the course of 25 years and stimulate linked sectors of their local economy. This type of hands-on, tacit learning about marketing contrasts with 1) supply-driven assistance programs aimed at improving production and; 2) being informed about marketing opportunities by a broker. The analysis suggests that farmer learning through marketing is as important to success in production as learning about production directly. The Tiendona is flooded by Guatemalan and Honduran tomatoes. Long-standing sourcing arrangements between Salvadoran wholesalers and Guatemalan and Honduran growers were altered by small wholesalers' collective actions pressuring the marketplace administration to cede vending space to them. Las Pilas small farmers were then able to enter the marketplace as suppliers to the small wholesalers. The Tiendona's public, contested nature contrasts with export marketing channels where there are few public pressure points to influence marketplace policies and buyers' sourcing decisions. Power asymmetries and coordination problems between producers and buyers can dissuade production. The case examined here finds robust marketing relationships between small wholesalers and small farmers driving crop diversification and production and marketing success. |
Unsupervised discovery of human behavior and dialogue patterns in data from an online game | A content authoring bottleneck in AI, coupled with improving technology, has lead to increasing efforts in using large datasets to power Al systems directly. This idea is being used to create Al agents in video games, using logs of human-played games as the dataset. This new approach to AI brings its own challenges, particularly the need to annotate the datasets used. This thesis explores annotating the behavior in human-played games automatically, namely: how can we generate a list of events, with examples, describing the behavior in thousands of games. First dialogue is clustered semantically to simplify the game logs. Next, sequential pattern mining is used to find action-dialogue sequences that correspond to higher-level events. Finally, these sequences are grouped according to their event. The system can not yet replace human annotation, but the results are promising and can already help to significantly reduce the amount of human effort needed. |
Cell-seeded type II collagen scaffolds for articular cartilage tissue engineering | Defects in articular cartilage exhibit little spontaneous healing response, in part due to the limited number of chondrocytes available to infiltrate the defect and the absence of a provisional fibrin scaffold to accommodate cell migration into the lesion. One variable related to tissue engineering strategies employing cell-seeded scaffolds to treat such defects is the amount of cartilage formed in the construct prior to implantation. The objectives of this thesis were to evaluate effects of scaffold cross-link density and bioreactor culture environment on chondrogenesis in cell-seeded type II collagen scaffolds in vitro, and to begin to test effects of implant compositional maturity (viz. glycosaminoglycan, GAG, content) on chondral defect repair. Scaffold cross-link density, a determinant of cell-mediated scaffold contraction and degradation, affected chondrogenesis; scaffolds of low cross-link density that experienced contraction exhibited greater cartilaginous tissue formation compared to highly cross-linked scaffolds that resisted contraction. In addition to tissue-level effects on histogenesis, cross-link density was found to direct phenotypic differentiation at the cellular level. When employing marrow-derived stem cells as an alternative to chondrocytes, scaffolds with lower cross-link densities (and thus less resistance to contraction and degradation) favored chondrocytic differentiation. |
Stadium development and urban renewal : a look at Washington, DC | This thesis investigates the factors, related to urban stadium development, that act as a catalyst for subsequent local urban renewal. Over the recent decades there has been substantial debate related to stadium or arena development. "The stadium debate intersects with cultural studies, economics, law, urban studies, civic planning, sports administration, mass communications, and sociology."' The center of this debate is over the economic and social "net benefit" to a city that undertakes a stadium development. Many argue that the economic and social costs created by urban stadium development outweigh the public good, especially in the case of publicly funded or subsidized stadiums. This thesis concentrates on the renewal of the surrounding real estate markets rather than broader economic renewal. When this thesis refers to "urban renewal" it is meant in the context of the renewal of the physical infrastructure and real property. The thesis examines the range of costs and benefits resulting from stadium induced urban real estate renewal. The benefits analyzed are derived from the changes in the local real estate markets that may be connected to the arena or stadium construction. The subject case study illustrates some of the broader economic benefits related to urban real estate renewal. Washington, DC provides a recent example of urban arena development that led to significant local investment in the development of the surrounding area. Construction of the Verizon Center led to development of residential, office, and retail product in the immediate area. The case study explores the factors (specific to the site, team owners, local developers, and city officials) that create a fertile environment for urban real estate renewal. |
Automotive component product development enhancement through multi-attribute system design optimization in an integrated concurrent engineering framework | Automotive industry is facing a tough period. Production overcapacity and high fixed costs constrain companies' profits and challenge the very same existence of some corporations. Strangulated by the reduced cash availability and petrified by the organizational and products' complexity, companies find themselves more and more inadequate to stay in synch with the pace and the rate of change of consumers' and regulations' demands. To boost profits, nearly everyone pursue cost cutting. However, aggressive cost cutting as the sole approach to fattening margins results invariably in a reduction of operational capabilities which is likely to result in a decline in sales volume that leads to further cost reductions in a continuous death spiral. Long-term profitable growth requires, instead, a continuous flow of innovative products and processes. The focus should be, therefore, shifted from cost reduction to increased throughput. Automotive companies need to change their business model, morphing into new organizational entities based on systems thinking and change, which are agile and can swiftly adapt to the new business environment. The advancement of technology and the relentless increase in computing power will provide the necessary means for this radical transformation. This transformation cannot happen if the Product Development Process (PDP) does not break the iron gate of cycle time-product cost-development expenses-reduced product performance that constrains it. A new approach to PD should be applied to the early phases, where the leverage is higher, and should be targeted to dramatic reduction of the time taken to perform design iterations, which, by taking 50-70% of the total development time, are a burden of today's practice. Multi-disciplinary Design |
Portable optical ground stations for satellite communication | Small satellite technical capabilities continue to grow and launch opportunities are rapidly expanding. Several commercial constellations of small satellites for Earth observation and communications are making their way onto orbit, increasing the need for high bandwidth data downlink. Laser communications (lasercom) has the potential to achieve high data rates with a reduction in power and size compared to radio frequency (RF) communications, while simultaneously avoiding the significant regulatory burden of RF spectrum allocation. Lasercom benefits from high carrier frequencies and narrow beamwidths, but the resulting challenge is to precisely point these beams between transmit and receive terminals. Arcsecond to sub-arcsecond pointing is required from both the space terminal and the ground station. While existing lasercom ground stations have primarily utilized professional telescopes at observatory-class facilities, making optical ground stations more affordable and transportable is a key enabler for expanding lasercom to small satellites and new applications, as well as establishing networks to mitigate the effects of weather. We describe the development of the Massachusetts Institute of Technology Portable Telescope for Lasercom (MIT-PorTeL) utilizing an amateur telescope augmented with an externally mounted receiver assembly. The ground station has a 28 cm aperture and utilizes a star tracker for automated calibration. The ground station reduces mass by at least 10x and cost by at least 100 x over existing optical ground stations. We present a ground station architecture that enables deployment in less than one hour and that is capable of tracking satellites in low-Earth orbit. We describe the receiver assembly and fine pointing system that enables arcseconds-level pointing accuracy. Finally, we present results from testing the ground station on the roof of an MIT building tracking a star and tracking the International Space Station. |
Beichuan Middle School architecture design | This Thesis is comprised of two interrelated parts: the first part attempts to incorporate Design and Methodology into the framework of Knowledge Organization, tries to refine/ redefine the workflow, and establishes a working model which represents/guides the process from the formation of ideas to the solutions of fabrication/construction using BIM. The second part further explore the meaning/connotation of Knowledge Organization, reconfigure the model and accordingly adjust the methods being used, considering the specific context and conditions at the earthquake area of Beichuan County in China. |
United States nuclear posture and foreign policy | This dissertation explains how and why US nuclear posture has changed from the late 1940s to the present. It argues that presidents reliably pursue aggressive nuclear postures to advance their ambitious foreign and security policy goals. In the course of advancing this main argument, it makes five additional contributions. First, it overturns the conventional or folk wisdom that Mutual Assured Destruction (or MAD) characterized US Cold War nuclear posture. In fact, the desire to escape MAD-not maintain it-was a major driver of aggressive US posture. Second, it upends the standard political science argument that US nuclear posture became aggressive as a result of military service rivalries or bureaucratic pathologies within the Pentagon. When it comes to nuclear posture, presidents carry far more weight than bureaucrats. Third, it fills an important gap in the existing literature. Barrels of ink have been spilled on US nuclear weapons policy and related topics. |
Industrial/statecraft : infrastructure and the making of industrial capitalism in India, ca. 1940 | In theories of development, public infrastructure serves as one of the myriad mediums through which the state seeks validation. In the modern period, infrastructure has often served as the symbol of state led progress. Infrastructure is thus a project of endorsement and justification of state's intervention. It is superfluous to say that infrastructure is a contested terrain within which the political economy of development unfolds. However, this thesis argues that it is through this iteration of infrastructure's intricate ways of creation and functioning that private capital begins to accumulate in post-colonial India. The project looks at the discourse of industrial development and planning in late and post-colonial India, investigating the manner in which infrastructure appears as a trope not only for state's validation but also for aggregation of the Indian industrialist class. |
Exploring electron and phonon transport at the nanoscale for thermoelectric energy conversion | Thermoelectric materials are capable of solid-state direct heat to electricity energy conversion and are ideal for waste heat recovery applications due to their simplicity, reliability, and lack of environmentally harmful working fluids. Recently, nanostructured thermoelectrics have demonstrated remarkably enhanced energy conversion efficiencies, primarily due to a reduction in lattice thermal conductivity. Despite these advances, much remains unknown about heat transport in these materials, and further efficiency improvements will require a detailed understanding of how the heat carriers, electrons and phonons, are affected by nanostructures. To elucidate these processes, in this thesis we investigate nanoscale transport using both modeling and experiment. The first portion of the thesis studies how electrons and phonons are affected by grain boundaries in nanocomposite thermoelectric materials, where the grain sizes are smaller than mean free paths (MFPs). We use the Boltzmann transport equation (BTE) and a new grain boundary scattering model to understand how thermoelectric properties are affected in nanocomposites, as well as to identify strategies which could lead to more efficient materials. The second portion of the thesis focuses on determining how to more directly measure heat carrier properties like frequency-dependent MFPs. Knowledge of phonon MFPs is crucial to understanding and engineering nanoscale transport, yet MFPs are largely unknown even for bulk materials and few experimental techniques exist to measure them. We show that performing macroscopic measurements cannot reveal the MFPs; instead, we must study transport at the scales of the MFPs, in the quasi- ballistic transport regime. To investigate transport at these small length scales, we first numerically solve the frequency-dependent phonon BTE, which is valid even in the absence of local thermal equilibrium, unlike heat diffusion theory. Next, we introduce a novel thermal conductivity spectroscopy technique which can measure MFP distributions over a wide range of length scales and materials using observations of quasi-ballistic heat transfer in a pump-probe experiment. By observing the changes in thermal resistance as a heated area size is systematically varied, the thermal conductivity contributions from different MFP phonons can be determined. We present the first experimental measurements of the MFP distribution in silicon at cryogenic temperatures. Finally, we develop a modification of this technique which permits us to study transport at scales much smaller than the diffraction limit of approximately one micron. It is important to access these length scales as many technologically relevant materials like thermoelectrics have MFPs in the deep submicron regime. To beat the diffraction limit, we use electron-beam lithography to pattern metallic nano dot arrays with diameters in the hundreds of nanometers range. Because the effective length scale for heat transfer is the dot diameter rather than the optical beam diameter, we are able to study nanoscale heat transfer while still achieving ultrafast time resolution. We demonstrate the modified technique by measuring the MFP distribution in sapphire. Considering the crucial importance of the knowledge of MFPs to understanding and engineering nanoscale transport, we expect these newly developed techniques to be useful for a variety of energy applications, particularly for thermoelectrics, as well as for gaining a fundamental understanding of nanoscale heat transport. |
Energy aware network coding in wireless networks | Energy is one of the most important considerations in designing reliable low-power wireless communication networks. We focus on the problem of energy aware network coding. In particular, we investigate practical energy efficient network code design for wireless body area networks (WBAN). We first consider converge-cast in a star-shaped topology, in which a central base station (BS), or hub, manages and communicates directly with a set of nodes. We then consider a wireless-relay channel, in which a relay node assists in the transmission of data from a source to a destination. This wireless relay channel can be seen as a simplified extended star network, where nodes have relay capabilities. The objective is to investigate the use of network coding in these scenarios, with the goal of achieving reliability under low-energy and lower-power constraints. More specifically, in a star network, we propose a simple network layer protocol, study the mean energy to complete uploads of given packets from the nodes to the BS using a Markov chain model, and show through numerical examples that when reception energy is taken into account, the incorporation of network coding offers reductions in energy use. The amount of achievable gains depends on the number of nodes in the network, the degree of asymmetry in channel conditions experienced by different nodes, and the relative difference between transmitting and receiving power at the nodes. We also demonstrate the compatibility of the proposed scheme with the IEEE 802.15.6 WBAN standard by describing ways of incorporating network coding into systems compliant to the standard. For a wireless relay channel, we explore the strategic use of network coding according to both throughput and energy metrics. In the relay channel, a single source communicates to a single sink through the aid of a half-duplex relay. The fluid flow model is used to describe the case where both the source and the relay are coding, and Markov chain models are proposed to describe packet evolution if only the source or only the relay is coding. Although we do not attempt to explicitly categorize the optimal network coding strategies in the relay channel under different system parameters, we provide a framework for deciding whether and where to code, taking into account of throughput maximization and energy depletion constraints. |
Brill-Noether-type theorems with a movable ramification point | The classical Brill-Noether theorems count the dimension of the family of maps from a general curve of genus g to non-degenerate curves of degree d in projective space Pr. These theorems can be extended to include ramification conditions at fixed general points. This thesis deals with the problem of imposing a ramification condition at an unspecified point. We solve the problem completely in dimension 1, prove a closed-form existence criterion and a finiteness result in dimension 2, and provide an existence test and bound the dimension of the family in the general case. |
Design and characterization of acoustic pulse shape memory alloy actuators | Single crystal Ni-Mn-Ga ferromagnetic shape memory alloys (FSMAs) are active materials which produce strain when a magnetic field is applied. The large saturation strain (6%) of Ni-Mn-Ga, and material energy density comparable to piezoelectric ceramics make Ni- Mn-Ga an interesting active material. However, their usefulness is limited by the bulky electromagnet required to produce a magnetic field. In this thesis, a novel actuation method is developed for shape memory alloys in their martensitic phase, whereby asymmetric acoustic pulses are used to drive twin boundary motion. Experimental actuators were developed using a combination of Ni-Mn-Ga FSMA single crystals and a piezoelectric stack actuator. In bi-directional actuation without load, strains of over 3% were achieved using repeated pulses (at 100 Hz) over a 30 s interval, while 1% strain was achieved in under 1 s. The maximum strains achieved are comparable to the strains achieved using bi-directional magnetic actuation, although the time required for actuation is longer. No-load actuation also showed a nearly linear relationship between the magnitude of the asymmetric stress pulse and the strain achieved during actuation, and a positive correlation between pulse repetition rate and output strain rate, up to at least 100 Hz. Acoustic actuation against a spring load showed a maximum output energy density for the actuator of about 1000 J/m³, with a peak-to-peak stress and strain of 100 kPa and 2%, respectively. |
From the beginning screening of lithium diffusion rates in transition metal oxide cathodes for lithium ion batteries | A screening metric for diffusion limitations in lithium ion battery cathodes is derived using transition state theory and common materials properties. The metric relies on net activation barrier for lithium diffusion. Several cathode materials are screened using this approach: [beta]'-LiFePO4, hexagonal LiMnBO3, monoclinic LiMnBO3, Li 3Mn(CO3)(PO4), and Li9V3 (P2O7)3(PO4) 2. The activation barriers for the materials are determined using a combined approach. First, an empirical potential model is used to identify the lithium diffusion topology. Second, density functional theory is used to determine migration barriers. The accuracy of the empirical potential diffusion topologies, the density functional theory migration barriers, and the overall screening metric are compared against experimental evidence to validate the methodology. The accuracy of the empirical potential model is also evaluated against the density functional theory migration barriers. |
Algorithmic issues in queueing systems and combinatorial counting problems | (cont.) However, these randomized algorithms can never provide proven upper or lower bounds on the number of objects they are counting, but can only give probabilistic estimates. We propose a set of deterministic algorithms for counting such objects for three classes of counting problems. They are interesting both because they give an alternative approach to solving these problems, and because unlike MCMC algorithms, they provide provable bounds on the number of objects. The algorithms we propose are for special cases of counting the number of matchings, colorings, or perfect matchings (permanent), of a graph. |
Evaluation of kickback energies in abrasive chain saws and cut-off saws: a theoretical and experimental study | Abrasive chain saws and cut-off saws used in construction work are not currently kickback regulated even though fatal kickbacks occur with them. Kickback regulation has been in place for wood chain saws since the 1980's, but it is inadequate for regulating abrasive power saws as they have a different cutting mechanic than wood chain saws. This thesis investigates the different kickback modes analytically and experimentally in order to develop tools for characterizing and predicting them. Simple physics models, numerical integrations, and two test rigs were designed to study them. First the different mechanisms which could cause kickback were considered (kickback modes) and simple physics models were made to describe them. From these, it was found that a pinch based kickback was potentially far more dangerous than a tip-engagement or snag type kickback for abrasive saws. |
Hip function characterization in the sagittal plane with varying gait speed | The function of the human hip joint during the stance phase of walking can be characterized with a configuration of simple mechanical elements. This combination of elements is capable of providing general hip behavior in the sagittal plane. Data was collected from two healthy, young subjects who walked at slow, normal and fast gait speeds. The hip can be modeled with a torque actuator and two independent, linear torsional springs, which are activated at different times during the stance phase of gait. The activation times consistently identify gait cycle events across all three gait speeds. The first spring operates during the single limb stance of the gait cycle. The second spring is actuated during second double support, in the pre-swing phase. The springs effectively reduce the amount of work an unaccompanied torque actuator would have to exert in order to reproduce the hip gait pattern. |
The design and control of a thermal management system for a photovoltaic reverse osmosis system | Reverse osmosis (RO) is a well-known process for desalinating seawater and brackish groundwater. Desalination is energy-intensive, so using photovoltaic (PV) panels to power the process is an attractive environmentally friendly concept, especially for community-scale systems. Increasing the system efficiency will lower the total cost of water produced, making the systems more economically competitive for a greater number of geographic locations. The thermal behaviors of PV panels and RO systems are complementary and can be exploited to improve photovoltaic reverse osmosis (PVRO) system efficiency. For a given level of solar radiation, a PV panel produces more electrical power at lower panel temperatures. For a given applied pressure, the flow of clean water across an RO membrane increases with increasing temperature. By using the RO feed water to cool the PV panels and warming the water in the process, more electrical power can be produced and higher flow rates of clean water across the RO membrane can be achieved, increasing total daily water production. The ability to cool the PV panels permits the use of low-cost, flat-plate concentrating mirrors, which further increase electrical power and clean water production. This thesis develops a thermal management system to improve the performance of a small-scale PVRO. Preliminary case studies show that the thermal exploitation concept is feasible and that a 50% increase in the total daily clean water production of a PVRO system is achievable, with an active thermal controller. A thermal controller is proposed that optimizes the PVRO system performance by minimizing the temperature of the solar panel and maximizing the temperature of the RO feed water. The control system uses a solar panel-mounted heat exchanger, circulator pump and servo valves to maximize water production while operating within the temperature limits of both the solar panel and the reverse osmosis membrane. Preliminary controller simulations show that it can successfully manage the temperatures of both the solar panel and RO feed water. The thermal management concept was experimentally validated on a small-scale, 300 L/day PVRO system. A 57% increase in clean water production was achieved using thermal management and solar concentrating mirrors, which agrees well with simulated performance predictions. |
Citrobacter rodentium induced liver changes in C57BL/6 mice : animal model of acute inflammatory stress and injury | The activation of inflammatory responses, while critical for host defense, contributes to hepatic injury in numerous acute and chronic liver disease states as well as drug-induced liver injury (DILI). The interactions that mediate susceptibility to liver injury and disease, however, are still poorly understood, underscored by the complexity of immune interactions and the diverse cellular composition and functions of the liver. Using Citrobacter rodentium, a well characterized rodent-specific enteric pathogen as a source of extrahepatic inflammatory stress; host liver responses, metabolic dysregulation, and susceptibility to injury in C57BL/6 mice were investigated. For the first time, we show altered liver pathology during the early course of C. rodentium infection, characterized by periportal necrosis indicative of thrombic ischemic injury, correlating with distinct circulating and tissue specific cytokine/chemokine profiles. Using Acetaminophen (APAP), a widely used analgesic and well-characterized hepatotoxin, we evaluated liver responses in isolation and in the context of host inflammation to gain insight into the role of live bacterial infection in altering liver metabolism and susceptibility to DILI. We combined systemic and tissue-specific cytokine/chemokine levels, clinical serum chemistries, and histopathological assessments of hepatic and enteric inflammation and necrosis to measure molecularlevel responses to treatment and their physiological effect. Using principal components analysis (PCA), clustering, partial least squares regression (PLSR), and a combination mutual-information-correlation network, enabled detection and visualization of both linear and nonlinear dependencies between molecules and physiological states across tissues and timepoints. C. rodentium-induced inflammatory stress was finally investigated for its potential in altering drug pharmacokinetics (PK) of substrates varying in their metabolic biotransformation and clearance mechanisms. Infection resulted in increased systemic oral exposure (AUC) of clinically relevant xenobiotics such as verapamil, propranolol, and digoxin. Functionally, these changes were not found dependent on CYP-mediated biotransformation of parent compounds; rather, they appear driven more by proposed gut barrier compromise. In conclusion, gastrointestinal infection with C. rodentium alters systemic and hepatocytes specific responses, not previously appreciated from this enteric pathogen, making it a useful model for studying host-pathogen interactions under acute hepatic inflammatory stress and injury. |
Micro active network after massive urban expansion | Urban form and transportation systems are closely related; each influences the other in different ways. This thesis explores this relationship, with a specific focus on systems for mega-cities in China, to pursue a new micro scale active mobility network supporting current shifts toward transit-oriented development, as a criticism for the past massive urban expansion in automobile development mode, and as an example for sustainable urban life in relation to transit development mode. The test field of this thesis is Shenzhen, a city that developed in an automobile-dependent mode and now aims to shift towards a transit-oriented development mode. Through city form analysis and transit-oriented city mapping studies, several typical urban forms of the city and the transportation characters of metro system are highlighted. Four typical urban form areas, which are connected by metro system, are selected as examples in the city for further design guideline initiative. Active mobility network is not only functioned to solve the last mile problem, connect citizens from public transit to destinations, but also a platform providing various new street life, for its flexibility, low-cost and low energy consumption. Contrasting to the wide and monotonous impression of roads in Chinese mega cities, such a network advocates for separate lanes with dedicated urban activities and speeds and a flexible surface that can adapt to changes in demand. The strategies of the four selected sites are four different solutions for active mobility networks related to internal resolving, density, service and recreation, with generations of various physical space and forms in different urban scenarios. These four typologies can be treated as typical examples for active mobility in Shenzhen, as important urban intervention operations, and references to other similar areas. |
Insights into formation of semiconductor nanocrystals : from first principles calculations to kinetic models of nucleation and growth | Over the past decade the synthesis of colloidal semiconductor nanocrystals of diverse shapes and sizes has sparked tremendous interest in both the industrial and scientific communities. Much of the work thus far has been done by extensive trial-and-error optimization of the chemistry to produce the desired nanocrystalline product. However, despite a tremendous effort in developing adaptable chemistries, the underlying mechanisms leading to nucleation and crystal growth in these systems are still not well understood. This thesis aims to address this challenge by utilizing first principles calculations and mathematical modeling to study the formation of cadmium selenide nanocrystals, the most frequently studied and best characterized nanocrystal system. In the first part of this thesis we investigate the elementary reaction steps that occur in the organic medium during early stages of particle nucleation. In particular, using density functional theory calculations, we probe the mechanism of formation of active growth species and small molecular clusters. We further explore the effect of ligand stabilization on cluster formation. In the second part, we explore reactions occurring on various surfaces of CdSe at later stages of crystal growth using periodic density functional theory calculations. Homoepitaxy and heteroepitaxy reactions on several relaxed and reconstructed wurtzite CdSe surfaces are investigated. Furthermore, the effect of ligand binding on crystal growth is examined using several model ligands. We show that ligands exhibit a range of affinities and selectivities for different facets of CdSe. We relate our findings to experimental observations, in particular, nanocrystal morphology and shape anisotropy. Finally, utilizing experimental and computational insights, we develop a mathematical model that explains both nucleation and growth in the formation of nanocrystals. |
Adaptively reusing limited service hospitality properties as workforce housing | This thesis examines the potential for reinventing an existing asset class -- limited service hospitality properties -- as new workforce housing properties. Due to the cyclical demand profile, modest price points, and relatively fragmented ownership of limited service hotels in the United States, these properties often escape the attention of highly sophisticated investors and developers. Furthermore, as relatively low margin businesses, many of the properties are visually unappealing and have significant deferred maintenance needs. Coupled with this dynamic is the fact that the United States is facing a nationwide housing crisis and suffers from a structural shortage of affordably priced housing alternatives. This thesis examines the economics of limited-service hospitality assets; the growth of workforce housing as a residential asset class; and the potential for redeveloping legacy hospitality properties into new workforce housing. In order to supplement the aforementioned macro and secular analyses, this thesis also includes a case study of the redevelopment of an antiquated motel in Oceanside, California. |
Synthetic strategies to improve the cytotoxicity of platinum-based cancer therapeutics | The purpose of this thesis is to explore the conjugation of biomolecules to platinum(IV) compounds. Ever since the serendipitous discovery that cisplatin has cytotoxic properties in the 1970's, research has focused on both understanding the mode of action and making new potential drugs that have more desirable properties than cisplatin. Oxidizing cisplatin to platinum(IV) allows for the tethering of amines that can be used to target cancer cells. The first chapter is a brief introduction on the scope of platinum compounds that have been made. It also provides general background on the proposed mode of action. The second chapter adapts the knowledge of making cisplatin derived platinum(IV) complexes that are able to conjugate biomolecules, and adapts it to a second generation platinum drug, carboplatin. Furthermore, a method to make mono- or bis- substituted conjugates is devised. Chapter three describes the tethering of a series of estrogen linkers to oxidized derivatives of carboplatin, in hopes of seeing increased toxicity in ER(+) cells. This work mimics previous work in our lab which dealt with the tethering of estrogen to oxidized derivatives of cisplatin. Finally, in chapter four platinum-folate conjugates are described. Because cancer cells grossly overexpress folate receptors, tethering folic acid to platinum(IV) is desirable as it will allow for increased uptake in cancer cells. |
A nonlocal dense granular flow model implemented in the material point method | A nonlocal model for dense granular flow is implemented in the material point method (MPM), an extension of the finite element method (FEM) for solid mechanics. The nonlocal model used has shown great predictive capability for dense flows when implemented in the finite element framework, but limitations of FEM prevent application of the model to truly large scale, inhomogeneous deformations. We show that these FEM results may be replicated in the MPM framework through the solution of a vertical chute flow, and allows for future work utilizing the strengths of MPM for larger and more complex flows. |
Analytical modeling of CO₂ migration in saline aquifers for geological CO₂ storage | Injection of carbon dioxide into geological formations for long-term storage is widely regarded as a promising tool for reducing global atmospheric CO₂ emissions. Given the environmental and health risks associated with leakage of CO₂ from such a storage site, it is critical to ensure that injected CO₂ remain trapped underground for the foreseeable future. Careful site selection and effective injection methods are the two primary means of addressing this concern, and an accurate understanding of the subsurface spreading and migration of the CO₂ plume during and after injection is essential for both purposes. It is well known that some CO₂ will be trapped in the pore space of the aquifer rock as the plume migrates and spreads; this phenomenon, known as capillary trapping, is an ideal mechanism for geological CO₂ storage because the trapped gas is immobile and distributed over a large area, greatly decreasing the risk of leakage and enhancing the effectiveness of slower, chemical trapping mechanisms. Here, we present an analytical model for the post-injection spreading of a plume of CO₂ in a saline aquifer, both with and without capillary trapping. We solve the governing equation both analytically and numerically, and a comparison of the results for two different initial plume shapes demonstrates the importance of accounting for the true initial plume shape when capillary-trapping effects are considered. We nd that the plume volume converges to a self-similar, power-law trend at late times for any initial shape, but that the plume volume at the onset of this late-time behavior depends strongly on the initial shape even for weakly trapping systems. |
Field studies of psychological health delivery systems in the U.S. Military | Large-scale health enterprises comprise multiple organizations that provide programs and services for patients. Despite the interconnectedness of these systems there is a lack of empirical research documenting how these organizations work collectively - or integrate - and how this integration impacts enterprise performance measured through quality, efficiency, and access. In the case of psychological healthcare, patients often require a number of services that span multiple departments and programs within an enterprise, increasing the complexity of maintaining a continuum of care for these patients. This paper, which is part of a larger effort to examine psychological healthcare in the U.S. Military Health System, presents a series of qualitative observations and analyses of the integration of psychological health-related organizations at two large health enterprises within the military. These qualitative inquiries take a multilevel approach for examining integration within these enterprises and address the following areas of interest: 1) the mechanisms for integration; 2) the objects of integration; 3) the dimensions of integration; 4) the contextual factors that influence integration; and 5) the impacts of integration on enterprise performance. Using semi-structured interviews, qualitative data was collected and then examined using content analysis to identify the most frequent themes for each area of interest. This data was used to validate and refine a comprehensive framework for integration that was developed to pull together multiple, distinct strands of the integration literature. This data was also used to demonstrate the relationship between different dimensions of enterprise performance and to identify areas where, in the process of optimizing enterprise performance, there is a trade-off between these dimensions. The preliminary, qualitative results of this research are intended to provide a conceptual foundation and framework for future analytic studies. |
A modal analysis of acoustic propagation in the changing Arctic environment | This work takes an in depth look at acoustic propagation through double duct sound speed profiles in the Arctic. While the traditional Arctic sound speed profile has a single surface duct, some portions of the Arctic have a sound speed profile which includes a second, lower duct. These double duct systems are seen through out the Beaufort Sea, dating back to 2004, in data made available by the Ice-Tethered Profiler program at the Woods Hole Oceanographic Institutes. The acoustic propagation through the double duct system is analyzed using normal mode analysis, through the Kraken normal mode code. A simulated lower duct is introduced in order to isolate only those modes which travel within the lower duct. Propagation through the lower duct is compared to propagation in traditional Arctic sound speed profiles, and for certain ducts distinct increases in propagation strength are shown. |
Combining adaptive and designed statistical experimentation : process improvement, data classification, experimental optimization and model building | Research interest in the use of adaptive experimentation has returned recently. This historic technique adapts and learns from each experimental run but requires quick runs and large effects. The basis of this renewed interest is to improve experimental response and it is supported by fast, deterministic computer experiments and better post-experiment data analysis. The unifying concept of this thesis is to present and evaluate new ways of using adaptive experimentation combined with the traditional statistical experiment. The first application uses an adaptive experiment as a preliminary step to a more traditional experimental design. This provides experimental redundancy as well as greater model robustness. The number of extra runs is minimal because some are common and yet both methods provide estimates of the best setting. The second use of adaptive experimentation is in evolutionary operation. During regular system operation small, nearly unnoticeable, variable changes can be used to improve production dynamically. If these small changes follow an adaptive procedure there is high likelihood of improvement and integrating into the larger process development. Outside of the experimentation framework the adaptive procedure is shown to combine with other procedures and yield benefit. Two examples used here are an unconstrained numerical optimization procedure as well as classification parameter selection. The final area of new application is to create models that are a combination of an adaptive experiment with a traditional statistical experiment. |
Spicket River Greenway Project, Lawrence, MA : teaching & learning design with the community | This research deals with three different issues. The first two, a site and a community, with their own possibilities and constrains, are going to be the protagonists of this adventure. The third one, the architect-planner, trying to play the role of a facilitator in the process, will translate their needs, rules and solutions into a comprehensive plan that should be suitable for the other two. Interpreting the rules of design, as a tool, the architect here would like to perform his role as a medieval alchemist, acting and waiting in expectation, trying not to interfere to allow the process to flow, but aware that transformations also involves oneself. The main source for this research is the teaching-learning experience at the Young Architects Program, with whom the author is going to share the discovery of the site and the envisioning of a better future. His personal goal is to explore the intimate relationship between ideal form, originated in the minds and images of each one of us, compared with the possibilities for realization after the filter of social, political and economic forces. This will be a teaching and learning experience to confront utopia and reality, within a given urban design problem. The experience of the first four months has been recorded in seven chapters and six journals, that in the form of a diary, carry the observations and reflections of the first approach to the case studied, as the result of the everyday contact with the space and the people. |
Toward commercializable microphase-separating copolymer electrolytes for rechargeable lithium batteries | Microphase-separating copolymers have been shown to possess the electrical properties of a polymer liquid and the mechanical properties of a solid. In the past, these materials had to be produced via anionic methods that are not easily scalable. To circumvent this problem, two approaches were adopted. In one, atom transfer radical polymerization (ATRP) was used to create the block copolymer, poly(oligo(oxyethylene) methacrylate)-b-poly(butyl methacrylate) [POEM-b-PBMA], which when doped with LiCF₃SO₃ (lithium triflate) exhibited good conductivity at room temperature ([sigma] [approximately] 3 x 10⁻⁶ S/cm at 23CÌ). The other approach used free radical techniques to prepare the graft copolymer, POEM-g-PDMS, starting with commercially available high molecular weight poly(dimethyl siloxane) [PDMS] monomethacrylate-terminated macromonomer. When doped with lithium triflate this material exhibited a conductivity of [sigma] [approximately] 8 x 10⁻⁶ S/cm at 23° C.This material was found to be stable up to [approximately] 300° C and flow resistant below an applied pressure of [approximately] 60 psi. In an effort to achieve single-ion conduction by Li⁺, the anionic charge was incorporated into the polymer backbone as a carboxylate. To prevent undesirable ion-pairing interactions two approaches were adopted: (1) tethering the anionic charge outside the conducting block of a block copolymer; (2) reacting the Lewis acid BF3 with the carboxylate to delocalize the negative charge. Both approaches produced materials with T[sub]Li⁺ = 1 while retaining conductivity ([sigma] <[or equal to] 7 x 10⁻⁶ S/cm) nearly identical to that observed in salt-doped systems. The self-doped electrolytes, however, demonstrated the ability to deliver stable currents at high drain rates, suggesting that immobilizing the anion improves rate capability. 100% solid-state cells consisting of lithium anodes, thin-film, fully-dense, vanadium oxide cathodes and microphase-separating copolymer electrolytes were operated for hundreds of cycles at discharge rates as high as [approximately] 1.5 C. Such batteries are conservatively projected to possess energy densities of 173 Wh/kg or 337 Wh/L. A design exercise optimizing a multilayer, thin-film battery configuration with the materials studied in this investigation speculated that specific energies exceeding 400 Wh/kg or 700 Wh/L may be attainable. The use of inexpensive materials and scalable processes in a simple design is expected to keep the cost of production low. |
Virtual urban traffic network simulator | In this project, I designed and implemented a virtual urban traffic network simulator. The simulator serves as a testbed for human-subject experiments to determine driver behavior in road networks and also as a platform for testing route-planning algorithms. The simulator was implemented using the C4 game engine and OpenGL. The simulator is capable of producing both 3- and 2-dimensional visualizations of a traffic network. In this thesis, I describe the key components of the simulator, the necessary inputs, and the expected outputs. I verify operation of the simulator through observation of the actual system outputs. |
3-pronged approach addressing capacity concerns in advanced turbine airfoil coatings | Following the tragedies of 9/11 in the United States and the ensuing down years for the aerospace industry, there are now signs of a comeback. This increase in demand is causing stress on the operations of many aerospace companies, who reduced operations and cut costs during the slowdown. This thesis proposes a three-pronged approach to undertake when looking to increase the capacity of an operation. The first step is to start changing the mindset on the shop floor to value all production pieces. The second step is to utilize all the analytical tools (such as Statistical Process Control) available to reduce set-up time and improve the targeted improvement activities of the operation. The third step is to properly evaluate the costs of purchasing new equipment and comparing that with potential alternative technologies. The proper utilization of these steps requires a thorough understanding of the organization's culture and how these change initiatives impact the relevant stakeholders. Without this organizational understanding, the chance of successfully meeting the increased customer demand is reduced. A case study was performed using this methodology at the Turbine Module Center at Pratt and Whitney. |
Structural studies of large architectural nucleoporins and coat proteins | The Nuclear Pore Complex (NPC) is a ~50 MDa protein complex that forms the sole conduit for macromolecular transport across the nuclear envelope. It assembles from ~30 proteins, termed nucleoporins or nups, symmetrically arranged about a central 8-fold axis. Some nucleoporins also contribute to other structures or perform diverse other functions. A subset forms a stably-associated core scaffold for the NPC, organized into two large subcomplexes, the Y-complex and the Nic96-complex. Studies using electron microscopy and X-ray crystallography have begun to elucidate the architecture of the NPC scaffold. To better understand the NPC and related proteins, two studies were performed: 1. The crystal structure of the C-terminal a-helical domain of Nup133 in complex with the C terminus of Nup107 revealed (with a prior structure of the N-terminal β-propeller domain of Nup133) the complete structure of Nup133 and its connection to the Y-complex. This contributes to a nearly complete molecular model of the Y-complex. Nup107-Nup133 forms the distal half of the stalk of the Y-complex. Sequence similarity between Nup133 and another nucleoporin, Nup170, was detected, and structural similarity proven by solving the structure of the a-helical domain of Nup170. Nup170 is a member of the Nic96-complex. Similarity between Nup133 and Nup170 suggests that the Y-complex and the Nic96-complex employ the same architectural principles. 2. Four nucleoporins contain an a-helical domain structurally related to the COPII coat protein Sec31. This domain is called the ancestral coatomer element 1 (ACE1). Sec13-Sec31 and Nup84-Nup145C-Sec13 complexes form analogous edge elements for the COPII coat and the NPC. A sequence-based search for other ACE1s identified the COPII accessory protein Sec16. Sec16 and Sec13 were shown to form a 2:2 heterotetramer. A crystal structure of Sec13-Sec16 revealed similarities to the Sec13-Sec31 edge element. Together with other structural and in vivo data, this result suggests that Sec13-Sec16 is a template for the Sec13-Sec31 coat. These studies demonstrate that duplication of multiple classes of architectural proteins occurred in the evolution of the NPC and COPII coat, and support the hypothesis that these systems evolved from a common, ancestral membrane-coating complex. |
Characterization of nanostructured hexagonal boron nitride patterned via high-resolution ion beam lithography | The forefront of polariton research in two-dimensional (2D) materials focuses on pushing the limits of patterning 2D materials into nanoresonators and other nanophotonic structures that manipulate highly confined polaritons for technologically relevant near-IR and mid-IR applications. Furthermore, tuning the properties of hexagonal boron nitride, graphene, and other 2D materials in-plane and stacking them into heterostructures has the potential to create hybrid optical, electronic, thermal, and mechanical properties with a wealth of new functions. To fully tailor these novel properties, controlled nanoscale patterning of these and other van der Waals materials is essential. Moreover, it becomes imperative to understand how patterning and geometries modify the properties within each layer or introduce defects that affect the interfaces of layered 2D heterostructures. Herein, we demonstrate high-resolution patterning of h-BN via both helium and neon ion beams and pattern a h-BN grating with a 35 nm pitch and 20 nm feature size. We study varying degrees of nanostructuring and defects via Raman spectroscopy, photo-thermal microscopy, and scattering-type scanning near-field optical microscopy and observe complimentary information about the phonon modes and the absorption and scattering of light from such nanostructures. Specifically, we observe geometry and layer dependent photo-thermal expansion of h-BN nanostructures that are mediated by phonons. This work demonstrates a thorough understanding of directly patterned 2D materials with ion beams and demonstrates that far-field and near-field measurements are essential in understanding how the nanostructuring of 2D materials can tune their properties. |
Quantitative analysis of the Epidermal Growth Factor Receptor-vIII mutant receptor signaling networks in GBM | Glioblastoma multiforme (GBM) is the most aggressive adult brain tumor and remains incurable despite multimodal intensive treatment regimens. EGFRvIII is a truncated extracellular mutant of the EGF receptor (EGFR) that is commonly found in GBMs and confers tumorigenic behavior. Although much work has been done over the past decade to elucidate pathways involved in EGFRvIII receptor signaling, the global map of the signaling networks that it activates remains incomplete, making it difficult to assess downstream components involved in EGFRvill-mediated transformation. To gain a molecular understanding of the mechanisms by which EGFRvIII acts, we have employed a mass spectrometry-based phosphoproteomic approach to quantitatively map cellular signaling events activated by this receptor. Using this approach, we have determined the major downstream pathways activated as a function of titrated EGFRvIII receptor levels. This analysis highlighted several aspects of EGFRvIII tumor biology, including crosstalk between EGFRvIII and other receptor tyrosine kinases. Specifically, we have identified the c-Met receptor as a co-target in the treatment of EGFRvIII positive GBM cells, and have shown that an EGFR and c-Met combination inhibitor strategy may be applicable in overcoming the poor efficacy of EGFR kinase inhibitor monotherapy in GBM patients. We then went on to investigate the mechanisms by which signaling networks are regulated in response to site-specific tyrosine mutations on EGFRvIII. This analysis has revealed a receptor compensation mechanism that is capable of restoring network architecture, upon the loss of a major tyrosine phosphorylation site on EGFRvIII. |
Effect of directed port air flow on liquid fuel transport in a port fuel injected spark ignition engine | With highly efficient modem catalysts, startup HC emissions have become a significant portion of the trip total. Liquid fuel is a major source of HC emissions during the cold start and fast idle period. Thus the control of liquid fuel, particularly during startup, is required for future engine designs. The effect of a turbulence plate, or charge motion control valve, on the port liquid fuel mass is examined. A purging process was used to examine the effect of a charge motion control plate on the port fuel film mass. It was found that the charge motion plate can provide substantial reductions in both the total and downstream intake fuel film mass. These reductions are the result of the increased intake turbulence that results from the charge motion plate. This increased turbulence provides enhanced fuel - air mixing and increases port film strip atomization due to the increased viscous shear. The effect of different film locations was postulated and backed experimentally. The downstream and valve films have the most immediate effect on the fuel delivery. Large upstream films, which can take thousands of cycles to develop, influence the fuel delivery on a much longer time scale. |
A client side tool for contextual Web search | This thesis describes the design and development of an application that uses information relevant to the context of a web search for the purpose of improving the search results obtained using standard search engines. The representation of the contextual information is based on a Vector Space Model and is obtained from a set of documents that have been identified as relevant to the context of the search. Two algorithms have been developed for using this contextual representation to re-rank the search results obtained using search engines. In the first algorithm, re-ranking is done based on a comparison of every search result with all the contextual documents. In the second algorithm, only a subset of the contextual documents that relate to the search query is used to measure the relevance of the search results. This subset is identified by mapping the search query onto the Vector Space representation of the contextual documents. A software application was developed using the .NET framework with C# as the implementation language. The software has functionality to enable users to identify contextual documents and perform searches either using a standard search engine or using the above-mentioned algorithms. The software implementation details, and preliminary results regarding the efficiency of the proposed algorithms have been presented. |
Excitation and de-excitation mechanichisms and luminescence enhancement in Er-doped silicon | Er-doped silicon (Si:Er) is a promising light emitting material for silicon microphotonics. A study of Si:Er excitation/de-excitation mechanisms and luminescence enchancement is presented in this thesis. A model based on impurity Auger and nonradiative nmltiphonon transitions (NRl\·IPT) is shown to describe the temperature quenching of the photoluminescence (PL) intensity from 4K to 300K This model asserts that the nonradiative Auger process is mainly responsible for the temperature quenching below lOOK, and NRMPT backtransfer process is mainly responsible for the temperature quenching above lOOK. Junction photocufrei1t · spectmscopy (JPCS) measurements confirmed the existence of a backtransfer mechanism that grows with temperature in accordance to the model. In order to circumvent the onset of nonradiative transitions at higher temperatures, spontaneous emission enhancement in nrnltilayer Si/Si02 microcavities was explored as a means to increase the PL intensity. Because multilayer microcavity structures cannot be constructed using single crystal silicon, Er-doped polysilicon (poly-Si:Er) was developed as a light emitting material for these microcavities. The poly-Si:Er material exhibited a luminescence very similar to that of Er in single crystal silicon. By crystallizing poly-Si:Er from amorphous material and performing a post-anneal hydrogenation, a reasonably high PL intensity, which was limited by the excitation power, was attained. Microacavities with poly-Si:Er were fabricated and measured for the first time. Cavity quality factors of -60-300 were measured, and an Er enhancement of -20x was observed. A -lOx enhancement of a small background emission from the polysilicon was also observed. The observed enhancement factors match well with computed enhancement factors derived from electric field intensity distribution within the microcavity structure. Exploratory work in optical gain from Si:Er waveguides and vertically coupled ring resonntors was conducted. A fiber coupling technique for low temperature waveguide transmission experiments was developed for the gain experiments. The transmission spectrum of a 3-cm long waveguide was measured at temperatures down to 125K. Because the temperature could not be lowered without debonding the fiber, a net gain could not be observed in this particular waveguide. The application of stimulated emission in Si:Er devices is analyzed and discussed. |
Responsive polymers for dynamic modulation of bio-macromolecular transport properties | Responsive self-assembling polymers are used in wide range of applications in the food, pharmaceutical, agricultural, electronic and environmental industries, as well as in the biomedical field. The proper design of such polymers is critical for the particular applications being considered. In this thesis, different matrices that can be modulated dynamically by the application of appropriate stimuli were designed and used for two applications: electrophoretic separation and gene transfection. Light represents an attractive trigger to change the properties of a polymer solution because it enables structural transitions to be induced under isothermal conditions without the addition of other chemical species to the solution, and is externally reversible and hence amenable to device design and automation. Amphiphilic copolymers with azobenzene moieties are of interest because the azobenzene can undergo reversible trans-cis photoisomerization leading to conformational isomers with significantly dissimilar dipole moments and hydrophobicities and thus different propensities to aggregate into nanoscale structures in aqueous media. Copolymers of 4methacryloyloxyazobenzene (MOAB) and N,N-dimethylacrylamide aggregate strongly in aqueous solutions with concentration-dependent aggregate size distributions and welldefined boundaries between the dilute and semi-dilute regimes. The copolymers are strongly surface active, an uncommon observation for random copolymers, and exhibit pronounced photoviscosity effects at higher concentrations. Trans-to-cis isomerization under UV light leads to partial dissociation of the azobenzene aggregates that form physical crosslinks, thereby significantly affecting the polymer solution rheology, with a consequent ten-fold loss of viscoelasticity upon irradiation, especially in concentrated polymer solutions. Photo-responsive poly(N,N-dimethylacrylamide-co-methacryloyloxyazobenzene) (MOAB-DMA) and temperature-responsive Pluronic F127 (PF127) copolymers were blended to obtain mixed micellar systems that were responsive to both stimuli. |
Foundational platform for mammalian synthetic biology | The emergent field of synthetic biology is different from many other biological engineering efforts, in that its roots, design principles, and forward engineering perspective have been adopted from electrical engineering and computer science. Synthetic biology is uniquely poised to make great contributions to numerous fields such as bio-fuel, energy production, agriculture and eco-remediation, national defense, and biomedical and tissue engineering. Considerable progress has been made in engineering novel genetic circuits in many different organisms. However, not much progress has been made toward developing a formal methodology to engineer complex genetic systems in mammalian cells. One of the most promising areas of research is the study of embryonic and adult stem cells. Synthetic biology has the potential to greatly impact the progression and development of research in this area of study. A critical impediment to the development of stem cell engineering is the innate complexity, little to no characterization of parts, and limited compositional predictive capabilities. In this thesis, I discuss the strategies used for constructing and optimizing the performance of signaling pathways, the development of a large mammalian genetic part and circuit library, and the characterization and implementation of novel genetic parts and components aimed at developing a foundation for mammalian synthetic biology. I have designed and tested several orthogonal strategies aimed at cell-cell communication in mammalian cells. I have designed a characterization framework for the complete and proper characterization of genetic parts that allows for modular predictive composition of genetic circuits. With this characterization framework I have generated a small library of characterized parts and composite circuits that have well defined input-output relationships that can be used in novel genetic architectures. I also aided in the development of novel analysis and computational tools necessary for accurate predictive composition of these novel circuits. This work collectively provides a foundation for engineering complex intracellular transcriptional networks and intercellular signaling systems in mammalian cells. |
Building practical systems that compute on encrypted data | Theft of confidential data is prevalent. In most applications, confidential data is stored at servers. Thus, existing systems naturally try to prevent adversaries from compromising these servers. However, experience has shown that adversaries still find a way to break in and steal the data. This dissertation shows how to protect data confidentiality even when attackers get access to all the data stored on servers. We achieve this protection through a new approach to building secure systems: building practical systems that compute on encrypted data, without access to the decryption key. In this setting, we designed and built a database system (CryptDB), a web application platform (Mylar), and two mobile systems, as well as developed new cryptographic schemes for them. We showed that these systems support a wide range of applications with low overhead. The work in this thesis has already had impact: Google uses CryptDB's design for their new Encrypted BigQuery service, and a medical application of Boston's Newton-Wellesley hospital is secured with Mylar. |
Islamic automata in the absence of wonder | This thesis looks at the interpretive difficulties posed by the Islamic automata, or hiyal manuscript--an ingenious genre of medieval illustrated manuscripts that describes and depicts mechanical devices such as water clocks, trick vessels, and automata. I choose to focus on the ways in which the automata manuscript has been viewed by scholars, rather than providing a history of the manuscripts themselves, precisely because this latter effort is complicated by a scholarly anxiety with what, exactly, Islamic automata manuscripts are, how they were used, or if (and how) they are valuable. This anxiety reveals not only a deeply subjective discontent with our totalizing "bourgeois" notion of technology - one that claims that we progress only by perfecting our implements - but also points to an inability to overcome this discontent. The way that this discontent is revealed through automata is that this "bourgeois" notion is not only totalizing, but also European. Automata scholarship thus allows us to see how European technology itself can be totalizing. The thesis reviews interpretive trends of this literature: The art historical origins of automata scholarship; mid century scholarship that touted the functional principles of the devices, and today's framework, which places automata in a linear technological evolution towards robotics, cybernetics, and advancement of human self-reproduction. Automata scholarship throughout has maintained a sterile distance from the historical context of the automata production. To close this gap, I argue, the ideological character of the Islamic automata manuscript must be revealed and its problematical relationship to technology disenchanted at every step. |
Coated particle fuel using advanced fuel performance modeling techniques | Modifying material properties provides another approach to optimize coated particle fuel used in pebble bed reactors. In this study, the MIT fuel performance model (TIMCOAT) was applied after benchmarking against the experiment results. The optimization study focuses on the fracture toughness of silicon carbide and Bacon anisotropy factor (BAF) of pyrocarbon. The variations on the silicon carbide toughness show that higher fracture toughness leads to a lower fuel failure probability, as expected. However, the results from the BAF variations reveal that a higher BAF lowers a fuel failure probability. This quite contradicts the generally believed notion that a higher BAF would increase fuel failures. In addition to the fuel design optimization, the failure characteristics of coated particle fuel are explained and the key factors influencing such characteristics are identified. |
Chemomechanics of self-oscillating gels | Biological materials such as cardiac and skin tissue exhibit the unique capacity to transduce mechanical stimuli into propagating electrical and chemical signals throughout the body. Few synthetic materials have been engineered to produce communicative chemical signals in response to mechanical input, though such synthetic material analogues could enable devices that mimic biological tissues and pressure sensitive processes whereby molecular mechanoreceptors enable rapid and localized transmission of chemical signals. In this thesis, self-oscillating polymer gels comprising N-isopropylacrylamideco- Ru(bpy) 3 are synthesized in order to elucidate chemical and mechanical (chemomechanical) coupling in synthetic, stimuli-responsive materials, and to design mechanically induced, oscillatory signaling systems. N-isopropylacrylamide-co-Ru(bpy) 3 gels represent a unique class of polymeric materials known as BZ gels, that are capable of undergoing the Belousov- Zhabotinsky (BZ) self-oscillating reaction. When submerged in stagnant solution containing chemical reactants, and in the absence of continuously applied external perturbation, the BZ gels exhibit sustained, colorful oscillations due to the changing oxidation state of Ru(bpy)3 transition metal complex. By measuring temperature profiles of the BZ gel, we showed that the swelling behavior and hydrophobicity of the gel depend on the oxidation state of covalently bound Ru(bpy) 3 . Using timelapse microscopy, we recorded the BZ oscillations and tracked the far from equilibrium chemical behavior exhibited by the gels. At constant system temperature, the BZ reaction induced cyclic changes in the osmotic pressure of the gel, resulting in periodic gel swelling and shrinking. Such volumetric changes, driven by the BZ reaction, are largest (22 %) when the edge length of the gel is relatively short (0.6 mm), and pattern formation is dominated by slow kinetics. Therefore, by quantifying the chemomechanical behavior of BZ gels, we demonstrated that the gels convert chemical oscillations into mechanical actuation. Next, we sought to design novel stimuli-responsive behavior in BZ gels by devising methods for mechanically triggering oscillations in quiescent gels. When sufficient macroscopic compressive stress was applied to submerged, non-oscillating gels, BZ oscillations were triggered and persisted until the stress was removed. To our knowledge, BZ gels represent the first synthetic hydrogel capable of producing oscillations in response to mechanical stimuli. To establish the conditions conducive to mechanical triggering, we quantified the chemical regimes for which BZ gels spontaneously oscillate or fail to oscillate. In doing so, we demonstrated that such regimes are governed by the ratio of inhibitor to activator species, which are both intermediate species that are produced throughout the reaction, providing negative and positive chemical feedback, respectively. Mechanically triggerable conditions corresponded to an intermediate ratio of reactant to inhibitor species, such that mechanical compression enabled transitions near the boundary dividing the non-oscillatory and oscillatory regimes. By varying the crosslinking density of the material, we also showed that both the required stress and strain for inducing oscillations in BZ gels increased with decreasing polymer volume fraction. Application of macroscopic, compressive stress to BZ gels caused a decrease in overall gel volume and an increase in the concentration of Ru(bpy) 3 , and oscillations were triggered at a critical concentration of Ru(bpy)3. In demonstrating that BZ gels can sense mechanical pressure and respond by transducing such energy into chemical oscillations, we have opened up new avenues of research based on mechanical sensing in BZ gels. Finally, we explore the mechanisms of synthetic "communication" in which discrete BZ gels sense mechanical stress and transmit chemical signals to neighboring gels. Specifically, we designed arrays of closely spaced gels (0.2 mm gap distance) that communicate via diffusion of activator species. We demonstrated that mechanical triggering can induce wave directionality to a set of gels that would otherwise exhibit disorder, and can propagate signals that change directions by migrating around bends without decay in signal amplitude. By introducing a node to a set of BZ gels, we showed that the BZ signal can split without attenuation, effectively doubling the system output. Last, we quantified the collision of two mechanically induced signals to show that wave collision occurs without amplification, and results in signal extinction. Taken altogether, these studies of signal propagation in BZ gels demonstrate that the underlying mechanism of BZ gel communication is governed by the diffusion of activator species. In addition to demonstrating for the first time a synthetic hydrogel that is capable of generating oscillations in response to mechanical compression, we have shown that BZ gels can propagate mechanically induced signals over long ranges and complex trajectories. Our results can be used to facilitate understanding of complex biological phenomena involving chemomechanical coupling and mechanotransduction, or to design advanced, functional materials that act as pulsating chemical or pressure sensors. |
Design and prototyping of a low-cost portable mechanical ventilator | This paper describes the design and prototyping of a low-cost portable mechanical ventilator for use in mass casualty cases and resource-poor environments. The ventilator delivers breaths by compressing a conventional bag-valve mask (BVM) with a pivoting cam arm, eliminating the need for a human operator for the BVM. An initial prototype was built out of acrylic, measuring 11.25 x 6.7 x 8 inches (285 x 170 x 200 mm) and weighing 9 lbs (4.1 kg). It is driven by a stepper motor powered by a 14.8 VDC battery and features an adjustable tidal volume of up to 900 mL, adjustable breaths per minute (bpm) of 5-30, and inhalation to exhalation time ratio (i:e ratio) options of 1:2, 1:3 and 1:4. Tidal volume, breaths per minute and i:e ratio are set via user-friendly knobs, and the settings are displayed on an LCD screen. The prototype also features an assist-control mode and an alarm to indicate over-pressurization of the system. Future iterations of the device will be fully calibrated to medical standards and include all desired ventilator features. Future iterations will be further optimised for low power-consumption and will be designed for manufacture and assembly. With a prototyping cost of only $420, the bulk-manufacturing price for the ventilator is estimated to be less than $100. Through this prototype, the strategy of cam-actuated BVM compression is proven to be a viable option to achieve low-cost, low-power portable ventilator technology that provides essential ventilator features at a fraction of the cost of existing technology. Keywords: Ventilator, Bag Valve Mask (BVM), Low-Cost, Low-Power, Portable and Automatic. |
Wirelessly addressable chemiresistors : carbon nanotube-based chemical sensors and dosimeters | In Chapter 1, we develop a brief tutorial on chemiresistive sensors and our rationale for choosing nanowires (NWs), specifically carbon nanotubes (CNTs), as the basis for chemiresistor research. Due to their high surface area, interwire junctions, and restricted conduction pathways, NWs give intrinsically high sensitivity and new mechanisms to transduce the binding or action of analytes. We explore principles for understanding electrical transport and transduction mechanisms in NW sensors. Next, we offer the reader a review of chemical sensor device performance parameters. Then, we summarize NW assembly and different device platform architectures. Subsequently, we discuss CNT functionalization strategies. Finally, we propose future developments in NW sensing to address selectivity, sensor drift, sensitivity, response analysis, and emerging applications. In Chapter 2, we describe a rapid, solvent-free, two-step procedure for the fabrication of selective gas and vapor sensors from carbon nanotubes and graphite on the surface of paper that overcomes challenges associated with solvent-assisted chemical functionalization and integration of these materials into devices. The first step generates solid composites from carbon nanotubes (or graphite) and small molecules (chosen to interact with specific types of gases and vapors) by mechanical mixing and subsequent compression into a form similar to a conventional pencil "lead." The second step uses mechanical abrasion ("drawing") of these solid composites on the surface of paper to generate functional devices. The use of diverse composites yields sensing arrays capable of detecting and differentiating gases and vapors at part-per-million concentrations. In Chapter 3, we describe an adaption of near-field communication (NFC) technology toward portable, wireless, non-line-of-sight gas phase chemical sensing. The result is a Chemically Actuated Resonant Device (CARD). We demonstrate the ability to convert inexpensive commercial NFC tags into chemical sensors that detect and discriminate analytes at part-per-thousand and part-per-million concentrations. This effort merges rational design of conductive nanostructured materials for selective chemical sensing with portable and widely distributed NFC technology to deliver a new method of acquiring chemical information about an NFC tag's local environment. We introduce a concept for distributed chemical sensing by the growing number of people that carry NFC-enabled smartphones, tablets, and other smart devices. We highlight second-generation NFC-based chemical sensor circuit designs that generalize turn-on/turn-off functionality. Finally, we devise a new Chemically Actuated Resonant Device (CARD) architecture that simplifies device fabrication to a single step and unambiguously relates change in chemiresistance to a wireless CARD readout. In Chapter 4, we describe a single-use wearable hazard badge that dosimetrically detects diethylchlorophosphate (DCP), a model organophosphorous cholinesterase inhibitor simulant. To provide selective and readily manufacturable sensor elements, we developed an ionic liquid-mediated single walled carbon nanotube based chemidosimetric scheme capable of detecting DCP across a broad dynamic range with limits of detection of 28 parts-per-billion (ppb). Furthermore, we have established a relationship between CARD readout and exposure dose, which can be generalized for any dosimeter developed with this device architecture. The device's practical utility is demonstrated wherein an 8-hour workday time weighted average equivalent exposure of 10 ppb DCP effects an irreversible change in smartphone readout. Finally, we show that the device is selective for DCP by screening against 8 commonly encountered interferent gases and vapors. In Chapter 5, we seek to develop an alkene dosimeter by exploiting known [4+2] cycloaddition chemistries. Specifically, di-substituted- 1,2,4,5-tetrazines are employed as a model inverse-electron demand Diels-Alder agent for the dosimetric detection of ethylene. Possible device architectures and fabrication strategies are explored, including colorimetric and chemiresistive implementations. A new solid/liquid chemiresistor fabrication method is demonstrated, and future directions for improving the sensitivity of diene-type alkene dosimeters are discussed. |
Optimization of acoustic feature extraction from dysarthric speech | Dysarthria is a motor speech disorder characterized by weak or uncoordinated movements of the speech musculature. While unfamiliar listeners struggle to understand speakers with severe dysarthria, familiar listeners are often able to comprehend with high accuracy. This observation implies that although the speech produced by an individual with dysarthria may appear distorted and unintelligible to the untrained listener, there must be a set of consistent acoustic cues that the familiar communication partner is able to interpret. While dysarthric speech has been characterized both acoustically and perceptually, most accounts tend to compare dysarthric productions to those of healthy controls rather than identify the set of reliable and consistently controlled segmental cues. This work aimed to elucidate possible recognition strategies used by familiar listeners by optimizing a model of human speech recognition, Stevens' Lexical Access from Features (LAFF) framework, for ten individual speakers with dysarthria (SWDs). The LAFF model is rooted in distinctive feature theory, with acoustic landmarks indicating changes in the manner of articulation. The acoustic correlates manifested around landmarks provide the identity to articulator-free (manner) and articulator-bound (place) features. |
Prefabrication and the individual | This thesis attempts to address the perceived failure of the manufactured housing industry to adequately speak to issues of universality and specificity. The universal is investigated and redefined according to two specific circumstances: two communities, two Lots and two houses, in two very different places. Hillside sites were selected in Santo Domingo, Dominican Republic, and in Los Angeles, California. A set of principles (the universal) was derived through the analysis of conditions on the two sites (the specific), and a new prefabricated construction system was developed in response. The universal acts as a platform for the individual; the prefabricated components are combined with sitebuilt elements, both systems having their role to play in the evolution of life on a hillside. |
Electrochemistry and Electrokinetics in Microchannels | The main body of this work considers the design and development of a microfluidic, continuous electrochemical sensor capable of measuring accurate potential differences. The key challenge in creating such a device is the implementation of a miniaturized reference electrode and salt bridge. The purpose of a salt bridge is to allow ionic conduction between the reference and working electrodes while maintaining a physical separation between the two systems. Macro reference electrode and salt bridge techniques are difficult to implement on a micro scale. Instead of attempting to conform one of these techniques to function in a micro system, new methods were developed that take advantage of the conditions in a continuous microfluidic device. In particular, laminar flow and slow relative diffusion times allow for a reference electrode that does not require a physical salt bridge. Ionic conduction is maintained between neighboring reference and analyte streams while slow mixing effectively separates the two systems. Several different device designs were investigated focusing on the prevention of reference electrode contamination. If the reference electrode is chemically contaminated it will no longer behave as expected and can not be used as a reference point. Contamination at the reference electrode was evaluated while varying flow rates and the geometry of the microfluidic device. Mathematical models were simulated in order to understand the mass transport in each device design. Based on these simulations, dimensionless groups were found that defined the dominant physics in each system. These dimensionless numbers were then validated experimentally and numerically over a range of device parameters. Subsequently, operation criteria were developed to ensure that the reference electrode remains stable and uncontaminated. By creating a stable reference electrode on chip, any homogeneous electrochemical system that was previously studied on the macro scale can now be studied continuously in a microfluidic device. A secondary portion of this work investigates the role of surface charge with respect to electrodynamics in a microchannel. As the surface area to volume ratio increases, the concentration of charge at a channel wall may begin to approach the electrolyte concentration in the bulk solution. This phenomenon is studied numerically, with and without convection, in particular as it relates to a possible mechanism for overlimiting current. Additionally, a potential de-ionization device is theorized based on this mechanism along with scaling arguments that can be used to aid device design. |
Low power display driver circuit for organic light-emitting diode displays | Organic light-emitting diode (OLED) devices offer a very promising alternative to existing flat panel display technologies, such as liquid crystal displays (LCD) that currently dominate the market. OLED displays offer very attractive characteristics, including higher luminous, larger viewing angle, and low-power consumption, over the established LCD technology. The ability to integrate OLED devices on a silicon microchip is one of the most favorable characteristics of this new technology. The primary goal of this research project is to implement a low-power display driver circuit for an OLED microdisplay. The implementation will be chosen based on the outcome of a feasibility study aimed at investigating the various options available for addressing the display and the design requirements imposed by the operation of the OLED. There are three primary design options to be considered: 1 ). Passive Matrix Addressing with sequentially addressed rows/columns, 2). Active Matrix Addressing with sequentially addressed rows/columns and dynamic storage at each pixel, and 3). Active Matrix Addressing with sequentially addressed rows/columns and static storage at each pixel. Each implementation is compared in terms of the overall power consumed in driving the high capacitance row and column lines in the display matrix. |
Visualizations for mental health topic models | Crisis Text Line supports people with mental health issues through texting. Unfortunately, support is limited by the number of counselors and the time each counselor has for clients, as well as the cognitive load on counselors from managing multiple conversations simultaneously. We conducted a contextual inquiry with crisis counselors to find contributing problems in their work flow. We believe topic modeling can provide automatic summaries of conversation text to augment note-taking and transcript-reading. Four simple and familiar visualizations were developed to present the model data: 1) a list of conversation topics, 2) a donut chart of topic percentages, 3) a line chart of topic trends, and 4) a scatter plot of specific topic points in the text. Our hypothesis is that these visualizations will help counselors spend more time on clients without overloading the counselors. The visualizations were evaluated through a user study to determine their effectiveness against a control interface. |
Multiscale structural investigation of bamboo under compressive Loading | Bamboo has been widely utilized as a load bearing material in building construction since ancient times by taking advantage of its excellent mechanical performance under loading as well as its low density and rapid growth. Application of bamboo to engineering, architecture and infrastructure requires in-depth understanding of the relationship between its morphology and mechanics, including how this regularly spaced segmental structure is optimized for its load bearing capability. However, our current knowledge about the linkage between the hierarchical structure and mechanical performance of bamboo is quite limited and we have little idea about the contribution of its regular spaced segment to its excellent mechanical behavior under various kinds of loading conditions. Here, we have implemented representative volume elements to our finite element analysis to study the mechanical response of the entire bamboo under compressive force and systematically investigated how the bamboo's meso-scale and macro-scale structural features (e.g., gradient fiber distribution, periodic nodes, etc.) contribute to its strength in compression. We find that column buckling is a critical failure mode that leads to overall collapse of the structure, which can be disastrous. We observe that the gradient fiber distribution pattern along the bamboo thickness direction significantly contributes to its strength. We find that the occurrence of fiber deviation at the node region reduces the strength of bamboo. Nevertheless, our results show that structural features such as external ridge and internal diaphragm play the role of reinforcement while the effect is more significant for bamboo than other plants with similar node appearance. The work reported in this thesis provides structural insights into the outstanding mechanics of bamboo, which could offer guidance for engineers to evaluate the material mechanics according to its structure and design high-performance structures with bamboo accordingly. |
Opportunities and challenges of mergers and acquisitions in India | The Indian economy has witnessed a major transformation since the government of India introduced the liberalization policies in 1991 .Since then M&A activity in India has picked up pace as foreign companies began to enter and expand their footprint in India. In developed economies especially the US and western Europe, M&A has been prevalent for a long time and is used as a major tool in corporate restructuring. India will continue to present itself as an attractive investment destination due to its high economic growth rate, growing middle class, favorable demographic divide and exceptional management talent. However M&A in India is a challenging proposition primarily due to the nature of its business holdings. Majority of companies in India are promoter held and their motivation of doing deals is different from that of firms run by professional management. Further the regulatory framework is still evolving and needs to be carefully analyzed to prevent post deal issues. In order to achieve the intended results companies must choose an appropriate sector, find relatively small but growing firms and acquire majority stake in those firms. It is imperative to build and manage relationship with the promoter family long before the initiation of deal activity. This would enable firms to be successful in their endeavors, mitigate potential risks and achieve desired synergies. |
Novel palladium precatalysts and their application in cross- coupling reactions and copper-catalyzed enantioselective ring formation | Chapters 1 - 3. A series of easily prepared, phosphine-ligated palladium precatalysts based on the 2-aminobiphenyl scaffold have been prepared. The role of the precatalyst-associated labile halide (or pseudohalide) in the formation and stability of the palladacycle has been examined. It was found that replacing the chloride in the previous version of the precatalyst with a mesylate leads to a new class of precatalysts with improved solution stability and that are readily prepared from a wider range of phosphine ligands, including the bulky, electron-rich di-tert-butylphosphino biaryl ligands. Additionally, N-methyl- and N-phenyl analogues have been prepared. These efficacy of these precatalysts were examined in a broad range of C-C, C-N, and C-O bond-forming reactions. Chapter 4. The intramolecular hydroalkylation of di- and trisubstituted alkenes bearing a pendant alkyl bromide to form stereodefined (hetero)carbocycles is reported. The system is highly regio- and stereoselective and employs a Cu-DTBM-SEGPHOS catalyst and (dimethoxy)methylsilane as the stoichiometric reductant. This intramolecular hydroalkylation reaction provides facile access to a multitude of ring systems and its utility is further demonstrated in the enantioselective synthesis of paroxetine. |
A practical application of simulation for production planning in a flexible pharmaceutical manufacturing environment | In pursuit of Novartis Pharmaceutical's vision to "Make Quality Medicine-On Time, Every Time" Novartis Ringaskiddy Limited (NRL), is pursuing Class A Manufacturing (MRP II) certification. Achieving Class A certification requires production plans be established and met with great accuracy. The Multi Synthesis Production Unit (MSPU) at NRL sets rolling eighteen month production commitments on a semi-annual basis. During planning a rudimentary tool is used to allocate production campaigns for over 40 different Active Pharmaceutical Ingredients (API) and intermediate products across roughly 200 pieces of equipment. This manual tool is time intensive to operate, prone to errors, and requires extensive knowledge of the facility. To address this problem, a simulation based model was developed using the software package, SchedulePro". The aim of this project was to create a system to support more effective and reliable production planning. The approach requires accuracy in modeling, a process to deploy the tool and a connection to people who use and benefit from the process. In order to evaluate the proposed model and understand potential benefits to Novartis, five case studies are used to compare the proposed planning process with traditional methods. First, when planning a future production year the proposed planning process demonstrates a reduction in required time as well as an increase in accuracy. Second, when evaluating the response to a change in demand a user with little knowledge of the plant can attain comparable response time as an experienced user of traditional methods. Third, when faced with an unplanned equipment failure the user is able to explore alternative production plans which minimize disturbance to the established production plan. Fourth, through the evaluation of alternative resource allocation plans the user can determine the lowest cost approach. Finally, when applied to a product launch evaluation the model is shown to reduce the number of planning cycles, focusing specifically on the ability of the site to support a product launch in a previously allocated year. |
Agency in computer-generated formal representations of sales work | This research builds on the literature on information technology and organizations to suggest an alternative to the current understanding of the production of computer-generated formal representations of work. This literature sees computer-generated formal representations of work as automatic outcomes of information technology that managers use to scrutinize employees. My ethnography of a desk-based sales unit suggests that managers have incentives to forfeit surveillance and instead apply their efforts to use information technology to build a facade of compliance with prescribed goals and prescribed rules, roles, and procedures. I show that such a facade requires continuous maintenance work and that it is employees, not managers, that have to engage in this work. Specifically, I show that employees need to engage in unprescribed work to earn the right to use formal information systems to represent work that they have not actually carried out. I explain how employees improvise a shadow information system to coordinate their unprescribed work across time. |
Intersystem crossing in acetylene : a mechanistic study | The triplet electronic states of acetylene and the intersystem crossing process that populates these states were studied using a second-generation, supersonic jet molecular beam machine with a multispectral capability. Surface Electron Ejection (SEELEM), Ultraviolet Laser Induced Fluorescence (UV-LIF), and Infrared Laser Induced Fluorescence (IR-LIF) spectra were acquired simultaneously following excitation in the vicinity of the A1Au [upsilon]3 = 3 [left arrow]X1[sum of]+g [upsilon] = 0 transition of acetylene. The A1Au [upsilon]3 = 3 level of acetylene is known to be coupled strongly to a background of near-degenerate triplet states. Simultaneous analysis of the complementary SEELEM and UV-LIF data using a set of recently developed analytical tools, yielded mechanistic insights into the non-radiative relaxation (i.e. intersystem crossing) process that distributes the optically accessible bright state over the background of dark states. The potential of the IR-LIF signal as an indirect probe of the triplet states of acetylene has also been demonstrated. Mechanistic insights into the SEELEM detection process itself were gained through a comparison of SEELEM data obtained separately in the first-generation and second-generation molecular beam machines. The two versions of the apparatus differed in their respective operating pressures. The operating background pressure has been found to be an important external factor in the SEELEM process. The much higher sensitivity (a factor of 103) of low-work function SEELEM surfaces such as Cs ([Phi]=2.1 eV) compared to that of other metals such as Au ([Phi]=5.1 eV) is observed only under low operating background pressures (4x 10-7 Torr). |
A system for bridging the ideological divide by establishing a moral framework for news consumption | Society is becoming increasingly ideologically divided. People exist in filter bubbles online, where they are exposed to primarily homogeneous perspectives and news sources. However, diversity in news consumption is important: it creates more informed societies, healthier democracies, and more solid understandings of one's own beliefs. In this thesis, we propose Pano - a system for bridging the ideological divide by surfacing moral framing in news. Pano provides a way to educate consumers of online news to think in a moral framework of shared human values, challenging their understandings of views held by those different from them through collaborative highlighting and annotation of moral framing in text. We demonstrate the effectiveness of Pano in a 10-day field study and find positive changes toward improved empathy and ability to re-frame arguments in the moral foundations of the audiences, showing promise toward the ultimate goal of bridging the ideological divide. |
Co-targeting among microRNAs is widespread and enriched in the brain | MicroRNAs (miRNAs) play roles in diverse developmental processes and cellular differentiation. Distinct miRNAs have hundreds to thousands of conserved binding sites in mRNAs, but typically exert only modest repression on a single site. Co-targeting of individual mRNAs by multiple different miRNAs could be commonly used to achieve stronger and more complex patterns of repression. Comparing target sets of different miRNAs, we identified hundreds of pairs of miRNAs that share more mRNA targets than expected (often ~2-fold or more) relative to stringent controls. For one co-targeting pair, miR-138 and miR-137, we validated functional overlap in neuronal differentiation. Clustering of the pairing relationships revealed a group of 9 predominantly brain-enriched miRNAs that share many targets. In reporter assays, subsets of these miRNAs together repressed gene expression by 5- to 10-fold or more, sometimes exhibiting cooperative repression. Our results uncover an unexpected pattern in which certain combinations of miRNAs can collaborate to strongly repress particular targets, and suggest important developmental roles. |
Interplay between electronic structure and catalytic activity in transition metal oxide model system | The efficiency of many energy storage and conversion technologies, such as hydrogen fuel cells, rechargeable metal-air batteries, and hydrogen production from water splitting, is limited by the slow kinetics of the oxygen electrochemical reactions. Transition-metal oxides can exhibit high catalytic activity for oxygen electrochemical reactions, which can be used to improve efficiency and cost of these devices. Identifying a catalyst "design principle" that links material properties to the catalytic activity can accelerate the development of highly active, abundant transition metal oxide catalysts fore more efficient, cost-effective energy storage and conversion system. In this thesis, we demonstrate that the oxygen electrocatalytic activity for perovskite transition metal oxide catalysts primarily correlates to the a* orbital ("eg") occupation. We further find that the extent of B-site transition metal-oxygen covalency can serve as a secondary activity descriptor. We hypothesize that this correlation reflects the critical influences of the a* orbital and transition metal-oxygen covalency on the ability of the surface to displace and stabilize oxygen-species on surface transition metals. We further propose that this ability to stabilize oxygen-species reflect as the rate-limiting steps of the oxygen electrochemical reactions on the perovskite oxide surfaces, and thus highlight the importance of electronic structure in controlling the oxide catalytic activity. |
Women in politics : a cross-national demand and supply analysis | It is striking that the sharp increase in the number of countries moving towards self-governance and democracy has not been accompanied by more equal political representation of women. What is equally puzzling is the contrast in the share of women in positions of political authority observed between countries, with many developed nations having fewer women legislators than a number of lesser-developed countries. Why are there so few women in most parliaments and why is there such variation across countries? To understand gender-based inequality in political authority, we look at the various stages of candidacy and identify potential bottlenecks to women participation and election into public office. There are three stages which one must pass through successfully to become a legislator. The first is becoming eligible and a part of the pool from which politicians are drawn, then being selected as a candidate and finally being elected to office. Potential barriers to entry for women in the legislative process may exist at any or all of these three stages. Each of these candidacy stages is discussed through a cross-national analysis and a case study of India. The cross-national data is for 175 countries at three points in time: 1975, 1985 and 1995. The Indian case study looks at women in parliament from the first general elections in 1951-1952 and focuses most on the 1996 parliamentary data. We argue that the key factor limiting the recruitment of women into politics is women's sparse representation in the pool from which politicians are recruited. Just as in thecase of men, women are drawn from an elite pool based on their occupational achievements. |
Energy system development in Africa : the case of grid and off-grid power in Kenya | This research used a combination of a grounded theory approach and system dynamics to study the electric power system in Kenya and to model the feedback at work in the development of the system. The ethnographic study revealed the challenges faced by consumers in choosing between grid and off-grid power options. Examination of this challenge leads to the hypothesis that competition between the grid and off-grid markets is contributing to the low growth in power consumption and that there is the potential for off-grid to become the dominant option in the future. This theory guided the construction of a system dynamics model focusing on consumers' decision-making and their interaction with the operation of the system. I then used the model to explore the dynamics of the system through scenario testing. There were two key outcomes from the model. The first showed that given the parameters chosen in most cases there is a clearly dominant option, although it changes over time. This finding points to the second key outcome the model, which is that there are realistic scenarios under which off-grid generation will become the dominant supply source. This shift could be induced by either reduced overhead on photovoltaic panels or high fuel prices. The outcomes from this research have implications for future electricity planning in Kenya and elsewhere in Africa. In particular, there is a need to decouple the system from external prices or account for the extreme uncertainty in fuel prices. Given the potential shift to large-scale off grid power generation, energy planners also need to look at options for managing a decentralized power system architecture and consider how to build in options for future reintegration if a large-scale centralized generation source comes online. |
Participation art online | Online participatory media holds the promise of activating otherwise passive audiences by providing spaces that encourage creative collaboration among diverse participants. The thesis traces the history of participation in artistic movements and early networked communication to contextualize a series of projects at the intersection of performance and participation online. Projects include WikiPhone, in which multiple participants collaborate on soundtracks in real-time, modifying existing online videos; OpenBrand, a system that allows participants to rewrite advertisements; Emma On Relationships, a video blog inviting participants to call in for love advice; and several other projects, exploring aspects of creativity and collaboration. Commonalities within these systems are examined in order to define design principles governing the creation of participatory media, and to explore the potential of these systems to effect social and political change. |
Multiphase flow in porous media with phase transitions : from CO₂ sequestration to gas hydrate systems | Ongoing efforts to mitigate climate change include the understanding of natural and engineered processes that can impact the global carbon budget and the fate of greenhouse gases (GHG). Among engineered systems, one promising tool to reduce atmospheric emissions of anthropogenic carbon dioxide (CO₂) is geologic sequestration of CO₂ , which entails the injection of CO₂ into deep geologic formations, like saline aquifers, for long-term storage. Among natural contributors, methane hydrates, an ice-like substance commonly found in seafloor sediments and permafrost, hold large amounts of the world's mobile carbon and are subject to an increased risk of dissociation due to rising temperatures. The dissociation of methane hydrates releases methane gas-a more potent GHG than CO₂-and potentially contributes to a positive feedback in terms of climatic change. In this Thesis, we explore fundamental mechanisms controlling the physics of geologic CO₂ sequestration and natural gas hydrate systems, with an emphasis on the interplay between multiphase flow-the simultaneous motion of several fluid phases and phase transitions-the creation or destruction of fluid or solid phases due to thermodynamically driven reactions. We first study the fate of CO₂ in saline aquifers in the presence of CO₂ -brine-carbonate geochemical reactions. We use high-resolution simulations to examine the interplay between the density-driven convective mixing and the rock dissolution reactions. We find that dissolution of carbonate rock initiates in regions of locally high mixing, but that the geochemical reaction shuts down significantly earlier than shutdown of convective mixing. This early shutdown reflects the important role that chemical speciation plays in this hydrodynamics-reaction coupled process. We then study hydrodynamic and thermodynamic processes pertaining to a gas hydrate system under changing temperature and pressure conditions. The framework for our analysis is that of phase-field modeling of binary mixtures far from equilibrium, and show that: (1) the interplay between phase separation and hydrodynamic instability can arrest the Ostwald ripening process characteristic of nonflowing mixtures; (2) partial miscibility exerts a powerful control on the degree of viscous fingering in a gas-liquid system, whereby fluid dissolution hinders fingering while fluid exsolution enhances fingering. We employ this theoretical phase-field modeling approach to explain observations of bubble expansion coupled with gas dissolution and hydrate formation in controlled laboratory experiments. Unraveling this coupling informs our understanding of the fate of hydrate-crusted methane bubbles in the ocean water column and the migration of gas pockets in hydrate-bearing sediments. |
Self-assembled thin films with tunable release kinetics for biomedical applications | The appropriate drug delivery approach is critical to therapeutic success without causing serious harm. Systemic delivery is a common method for delivering drugs but high dosages are needed to maintain elevated plasma levels, which can increase the risk of adverse drug reactions. Locally delivering these drugs offers an interesting alternative by significantly reducing the necessary dosages for therapeutic effect. Unfortunately, controlled delivery strategies can be limited with regards to the types of drugs as well as their release kinetics, especially if more than one drug is involved. Furthermore, for implantable devices biocompatibility and biodegradability are important considerations for minimizing undesirable side effects. This thesis utilizes the layer-by-layer (LbL) approach for the construction of biodegradable thin films that are capable of controlling the kinetics of localized drug delivery. Typically, controlling the release profiles of more than one drug is difficult, especially from a single film or device. To treat especially recalcitrant diseases, it is sometimes necessary to utilize complex or multimodal drug release behavior for positive therapeutic outcome. We first examined the use a naturally-derived, hydrolytically-degradable polyanion, poly(P-L-malic acid) (PMLA), for use in LbL films to control the rate of film erosion and hence protein elution which was tuned from minutes to weeks depending on the film's architecture. We further adapted the film architecture with subtle chemical modifications for sequential protein release without significant overlap and staged release where a rapid hemostat release coincided with sustained antibiotic release for more than 24 hours. We also examined a biodegradable thin film formulation capable of long-term (14+ months) elution of an active small molecule drug. For rapid hemostasis, it is sometimes difficult to achieve without use of sensitive biologics that require constant refrigeration and so we utilized films composed of self-assembling peptide nanofibers, which we found to be extremely robust and rapidly capable of forming nanofiber based clots despite prolonged incubation in elevated temperatures (2 months at 60°C). Overall this work expands on the controlled release repertoire by investigating fundamental aspects as well as translational applications. |
Energy consumption metrics of MIT buildings | With world energy demand on the rise and greenhouse gas levels breaking new records each year, lowering energy consumption and improving energy efficiency has become vital. MIT, in a mission to help improve the global energy system, launched the MIT Energy Initiative in 2006, and is aggressively trying to improve campus-wide energy consumption. MIT has also teamed up with the NSTAR utility company and pledged to reduce electricity generation by 15% by 2013 to serve as a model for others to follow. This thesis presents a measurement of the current performance of the most recent energy practices and seeks to provide a direction for future improvements. Following energy consumption over the course of the past decade, energy performance for the MIT campus and its buildings could be observed. It was determined that the majority of efforts to improve energy efficiency have been successful. Improvements to the campus' largest consuming laboratory buildings have significantly reduced consumption, and new energy projects have provided millions of dollars in yearly savings. Behavioral projects have not yet shown widespread success across campus but have had a drastic reduction effect in student residences where energy consumption has decreased 23% since 2006. Many buildings, which are not the largest energy consumers on campus, are continuing to increase consumption and need to radically improve for MIT to decrease overall consumption. |
Development of a process for continuous creation of lean value in product development organizations | Ideas and methodologies of lean product development were developed into tools and processes that help product development organizations improve their performances. The definition of waste in product development processes was re-examined and developed into a frugal set to cover all types of waste in product development processes through preliminary case studies. Value stream mapping (VSM) was optimized for measuring the waste indicators in product development processes. Typical causes for low product development project performances were organized into a root-cause analysis diagram. Three case studies in product development companies were performed. The tools were tested and improved through intensive interviews with both project managers and engineers. VSM was effective for identifying and measuring waste indicators. The root-cause analysis diagram was effective for quickly identifying root causes for low product development project performances. Synchronized uses of these tools made it possible to measure each root cause's impact on project performances. The result of measurements revealed both problems shared by all the projects and the ones specific to the projects, indicating that the tools and processes developed in this research can provide suggestions for continuous improvement of product development processes. Some waste indicators were more prevalent than the others, implying that the number of waste indicators to be considered can be reduced. Inventory of information was prevalent in all the projects, and the analyses of it implied that Today's product development processes are as premature as those of manufacturing several decades ago. |
Rethinking the application-database interface | Applications that interact with database management systems (DBMSs) are ubiquitous in our daily lives. Such database applications are usually hosted on an application server and perform many small accesses over the network to a DBMS hosted on a database server to retrieve data for processing. For decades, the database and programming systems research communities have worked on optimizing such applications from different perspectives: database researchers have built highly efficient DBMSs, and programming systems researchers have developed specialized compilers and runtime systems for hosting applications. However, there has been relative little work that examines the interface between these two software layers to improve application performance. In this thesis, I show how making use of application semantics and optimizing across these layers of the software stack can help us improve the performance of database applications. In particular, I describe three projects that optimize database applications by looking at both the programming system and the DBMS in tandem. By carefully revisiting the interface between the DBMS and the application, and by applying a mix of declarative database optimization and modern program analysis and synthesis techniques, we show that multiple orders of magnitude speedups are possible in real-world applications. I conclude by highlighting future work in the area, and propose a vision towards automatically generating application-specific data stores. |
A method for using polyethylene passive samplers to measure polycyclic aromatic hydrocarbon chemical activity in sediments | In order to aid in the determination of the hazards posed by hydrophobic organic compounds (HOCs) in sediment beds, a method for the use of polyethylene (PE) sheets as passive sampling devices for measuring chemical activities was explored. A model which depends on a concentration gradient and two mass transfer limiting zones in series was used. Internal tracer chemicals within the polyethylene devices (PEDs) were used to calibrate the mass transfer model which can have different mass transfer coefficients depending on the site and target chemicals being investigated. The model allowed for the measurement of HOC chemical activities by measuring the change of mass of tracer and target chemical within the PED, and knowing the PE-water partitioning coefficient, ..., and the liquid solubility, ..., of the target chemical. The method was tested using PEDs impregnated with dlO-phenanthrene and dlO-pyrene. First, PEDs were used to measure known concentrations of phenanthrene and fluoranthene in stirred seawaters. Seeing that the PEDs performed well, returning results which were within 25% of the known chemical activities, PEDs were then tested for measuring phenanthrene, fluoranthene, and pyrene in Boston Harbor sediments. |
Generalized approach to minimal uncertainty products | A general technique to construct quantum states that saturate uncertainty products using variational methods is developed. Such a method allows one to numerically compute uncertainties in cases where the Robertson-Schrodinger (RS) uncertainty approach fails. To demonstrate the limitations of the RS approach, the ([Delta]x2 )([Delta]p) relation is examined using both the variational and direct method. |
A framework for achieving lifecycle value in product development | Best lifecycle value is a concept rooted in value analysis, lifecycle consideration, and systems engineering which has evolved to support a holistic perspective of system development and program management. This thesis determines factors enabling consideration and achievement of lifecycle value by examining product development work. The scope of this effort focuses on aerospace programs to characterize lifecycle value for complex systems. Although different systems may define lifecycle value differently, there are common elements of the concept that have been identified. Based on four in-depth case studies and existing models, a theoretical framework for lifecycle value creation has been developed. The structure for this framework consists of three somewhat sequential and iterative processes: value identification, value proposition, and value delivery. Results from the case studies reported in the form of best practices have been related to this theoretical framework in six categories of value attributes. These attributes (holistic perspective, organizational factors, requirements and metrics, tools and methods, enterprise relationships, and leadership and management) apply to the entire lifecycle value framework. The combination of the framework and the practices from the case studies form a lifecycle value creation model, suggesting a lifecycle value approach encompasses appropriate and successful strategies for product development, system design, and program management. |
Digital pre-compensation for faulty digital-to-analog converters | In some contexts, DACs fail in such a way that specific samples are dropped. The dropped samples lead to distortion in the analog reconstruction. We refer to this as the "missing pixel" problem. Under certain conditions, it may be possible to compensate for the dropped sample by pre-processing the digital signal, thereby reducing the analog reconstruction error. We develop three such compensation strategies in this thesis. The first strategy uses constrained minimization to calculate the optimal finite-length compensation signal. We develop a closed-form solution using the method of Lagrange multipliers. Next, we develop an approximation to the optimal solution using discrete prolate spheroidal sequences. We show that the optimal solution is a linear combination of the discrete prolates. The last compensation technique we develop is an iterative solution in class of projection-onto-convex-sets. We develop the algorithm and prove that it converges to the optimal solution found using constrained minimization. Each of the three strategies are analyzed and results from numerical simulations are presented. |
Learning hierarchical motif embeddings for protein engineering | This thesis lays the foundation for an integrated machine learning framework for the evolutionary analysis, search and design of proteins, based on a hierarchical decomposition of proteins into a set of functional motif embeddings. We introduce, CoMET - Convolutional Motif Embeddings Tool, a machine learning framework that allows the automated extraction of nonlinear motif representations from large sets of protein sequences. At the core of CoMET, lies a Deep Convolutional Neural Network, trained to learn a basis set of motif embeddings by minimizing any desired objective function. CoMET is successfully trained to extract all known motifs across Transcription Factors and CRISPR Associated proteins, without requiring any prior knowledge about the nature of the motifs or their distribution. We demonstrate that motif embeddings can model efficiently inter- and intra- family relationships. Furthermore, we provide novel protein meta-family clusters, formed by taking into account a hierarchical conserved motif phylogeny for each protein instead of a single ultra-conserved region. Lastly, we investigate the generative ability of CoMET and develop computational methods that allow the directed evolution of proteins towards altered or novel functions. We trained a highly accurate predictive model on the DNA recognition code of the Type II restriction enzymes. Based on the promising prediction results, we used the trained models to generate de novo restriction enzymes and paved the way towards the computational design of a restriction enzyme that will cut a given arbitrary DNA sequence with high precision. |
Google as a bookmarking tool | In this thesis, a scheme is presented that harnesses the power of Google to locate previously visited webpages. Instead of bookmarking webpages by their URL, this process is based on addressing pages by their Google URL, which is a short sequence of words that, when queried in Google, locates the desired webpage as the top search result. Good GoogleURLs are memorable and easily communicable, even by speech, and avoid the portability and scalability problems of traditional bookmarks. The GoogleURL Generator-a tool to automatically generate GoogleURLs-is presented, and is complemented with a Google Reordering tool that automatically reorders Google search results based on user browsing history. The goal of this thesis is to employ GoogleURLs and Google Reordering to make Google's "I'm Feeling Lucky" button more lucky. |
Roles for the polycomb group protein BMI1 in lung adenocarcinoma progression and maintenance | The B lymphoma Mo-MLV insertion region 1 homolog (Bmil) protein is implicated as an oncogene in a variety of human cancers. During normal development, Bmil acts as part of a transcriptional repressive complex that regulates processes such as stem cell self-renewal, cell-fate commitment, and proliferation. During tumorigenesis, many cancers co-opt these core Bmil functions. A subset of these malignancies demonstrates an increased dependence on Bmil, revealing a window of vulnerability that could be exploited for the therapeutic benefit of patients. With the advent of cancer genomics, we have begun to appreciate the complex molecular determinants that inform these therapeutic windows. Indeed, Bmil is now understood to play unique roles depending on the underlying genetic contexts of tissue type, developmental stage, and of individual tumors. With this in mind, in this thesis I used mouse models of oncogenic Kras driven lung cancer to explore the potential dependence of lung adenocarcinoma, among the deadliest of cancers, on Bmil for tumor initiation, progression, and maintenance. Specifically, I demonstrate that Bmil is dispensable for tumor initiation but mediates the transition to advanced disease and thereby impacts the overall survival of tumor bearing animals. I show that Bmil is required to sustain the proliferative capacity of Kras driven lung adenomas. In part, it does so by enforcing efficient progression through the cell cycle independent of a canonical target, the pl9ARF-p53 tumor suppressive axis. This creates a large potential therapeutic window, as half of all lung adenocarcinoma patients display p53 mutations. My gene expression analyses further establishes a critical role for Bmil in the repression of developmental regulators, which may contribute to the atypical differentiation state of tumors as they adapt to the loss of Bmil. Finally, using both mouse models and cell lines, I present evidence that a subset of advanced tumors are sensitive to the loss of Bmil. Overall, this thesis advances our understanding of the genetic dependencies of Bmil in lung cancers and reveals novel mechanisms that potentially can be exploited to combat this deadly of disease. |
Essays in corporate finance | This dissertation presents three essays in Corporate Finance. In the first essay, I study managerial incentives in internal capital markets. In particular, I develop a two-tiered agency model to study division managers' incentives within internal capital markets. Division managers try to influence the external capital market's assessment of the firm and the internal capital market's assessment of their divisions in order to increase their level of funding. I show that, as the number of divisions increases, the external capital market's assessment of the firm becomes a public good for division managers, and the internal capital market replaces the external capital market in the provision of managerial incentives. I also show that, while diversified firms have an advantage in allocating resources, this may come at the expense of managerial incentives. Based on the analysis, the paper relates the value of diversification to characteristics of the firm, the industry, the external capital market, and the internal capital market. In the second essay, I propose a model of entrepreneurship in which investors decide whether to become venture capitalists or to form firms and entrepreneurs decide whether to join a firm or seek financing in the venture capital market. The venture capital market allows better matching between investors and entrepreneurs, but this comes at the cost of adverse selection. The model suggests that as a sector matures, innovation takes place first within firms, then in ventures backed by venture capitalists backed ventures, and finally within firms again. |