Update README.md

Removed example from README

README.md

@@ -86,268 +86,6 @@ a corpus of pre-annotated text for other tasks (e.g. figure caption to graphic,
 
 ## Dataset Structure
 
-### Data Instances
-```json
-{
-    "accession_id": "PMC3222825",
-    "pmid": "22038649",
-    "introduction": [
-        "<title>Introduction</title>",
-        "<p>RNA is increasingly viewed not only as an intermediary between DNA and protein, but also as a key regulator of gene expression and catalysis (Boisvert et al.  ##REF## ; Korostelev et al.  ##REF## ; Steitz  ##REF## ; Breaker  ##REF## ; Wahl et al.  ##REF## ; Newman and Nagai  ##REF## ). Methods to obtain very high-resolution structures and site specific dynamics of RNA molecules are therefore critical. Nuclear magnetic resonance (NMR) spectroscopy has emerged as an effective method for solving RNA 3D structures and for probing their motions. However, RNA molecules that are bigger than 40 nucleotides are difficult to characterize using NMR spectroscopy because of severe chemical shift overlap and rapid signal decay. Selective labeling techniques, first introduced to extend the utility of NMR spectroscopy for protein analysis, are increasingly being applied to RNA as well (Dayie  ##REF## ; Lu et al.  ##REF## ).</p>",
-        "<p>Of the three primary methods for incorporating selective isotopic labels into nucleic acids, biomass production using <italic>E</italic>. <italic>coli</italic> variants (Johnson et al.  ##REF## ; Johnson and Hoogstraten  ##REF## ; Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ) appears to be more general and cost-effective than chemical synthesis (Milecki  ##UREF## ) or de novo biosynthesis (Schultheisz et al.  ##REF## ; Schultheisz et al.  ##REF## ). Selective introduction of <sup>13</sup>C isotopes into the ribose ring using chemical synthesis entails difficult regio- and stereo-selective synthesis with low yields (Milecki  ##UREF## ). The de novo biosynthetic method requires a large number of enzymes of which only a few are commercially available (Schultheisz et al.  ##REF## ; Schultheisz et al.  ##REF## ). Additionally, the de novo method requires prohibitively expensive precursors that are limited in the labeling pattern that can be achieved with the biomass methods.</p>",
-        "<p>Unfortunately a biomass method based on <italic>E</italic>. <italic>coli</italic> metabolism, until recently, suffered from the disadvantage of isotopic scrambling in both the ribose and nucleobase atoms. This scrambling is caused by the admixture of the metabolic flux from both the oxidative and non-oxidative pentose phosphate pathways (noPPP) and a very efficient TCA cycle (Johnson et al.  ##REF## ; Johnson and Hoogstraten  ##REF## ). Hoffman and Holland earlier showed that wild-type <italic>E</italic>. <italic>coli</italic> grown using selectively labeled acetate enables the incorporation of <sup>13</sup>C-labels at either the protonated or non-protonated carbon sites in the RNA ribose and nucleobases. LeMaster and coworkers also showed that a modified bacterial strain, in which the TCA cycle enzymes succinate dehydrogenase and malate dehydrogenase were disabled (DL323), could be grown on selectively labeled glycerol to provide alternative site labeling that removed the <sup>13</sup>C-<sup>13</sup>C coupling in protein NMR spectra (LeMaster and Richards  ##REF## ; LeMaster and Kushlan  ##UREF## ). Hoogstraten and coworkers then showed that the DL323 strain grown on selectively labeled glycerol also enabled site selective labeling within the ribose ring (Johnson et al.  ##REF## ; Johnson and Hoogstraten  ##REF## ). Earlier, Pardi and colleagues introduced labeling using <sup>13</sup>C-formate in a background of <sup>12</sup>C-glucose which site specifically labels purine C8 position exclusively; results were not presented for purine C2 position (Latham et al.  ##REF## ). Building on these previous ideas, we showed that an isolated two spin system could be achieved for both base and ribose depending on the strain and complementation of the growth media with formate (Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). Even under these newer conditions, the purine C2 positions were only labeled to ~26%, indicating a general limitation of the formate supplementation methodology (Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ).</p>",
-        "<p>However until now, most biomass production of nucleotides had focused on using symmetric carbon sources such as glycerol, glucose, and acetate. For instance if glucose is the carbon source, then at the onset of the TCA cycle this input carbon source is already diluted to 50% by glycolysis and similarly the purine precursor 3-phosphoglycerate (3PG) is only 50% labeled. Similarly if [1, 3-<sup>13</sup>C]-glycerol is the carbon source, selectivity in the ribose and base regions are again reduced because of the symmetric nature of glycerol such that the purine C5 and all the ribose C1\u2032 have substantial residual coupling that can render relaxation measurements inaccurate (Johnson et al.  ##REF## ; Dayie and Thakur  ##REF## ). This residual coupling arises from the C-1 carbon of three carbon precursors such as pyruvate and glycerol, and thus [1-<sup>13</sup>C]-pyruvate will similarly introduce coupling between C1\u2032-C2\u2032-C3\u2032 and C2\u2032-C3\u2032. We, therefore, reasoned that the asymmetry of <sup>13</sup>C -labeled pyruvate would provide selective labeling in both the ribose and base moieties of nucleotides that would be impossible with symmetric carbon sources.</p>",
-        "<p>Unlike the previous result with the use of [1, 3-<sup>13</sup>C]-glycerol (Johnson et al.  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ), the use of [3-<sup>13</sup>C]-pyruvate affords selective <sup>13</sup>C enrichment of the ribose C1\u2032 and C5\u2032, of the purine C2 and C8, and of the pyrimidine C5 without any of the residual couplings described above. NMR T<sub>1</sub> relaxation experiments and <sup>13</sup>C-methylene TROSY or HSQC on a site selectively <sup>13</sup>C -labeled 27-nt A-site RNA demonstrate that accurate T<sub>1</sub> relaxation parameters and higher resolution spectra can be obtained. Thus, the technology to incorporate site specific labels into both the base and ribose moieties of purine and pyrimidine nucleotides promises to expand the scope of NMR studies to regions hitherto inaccessible.</p>"
-    ],
-    "methods": [
-        "<title>Materials and methods</title>",
-        "<title>Bacterial strains</title>",
-        "<p>The mutant strains DL323 (CGSC # 7538, F-, sdh-1, &amp;lambda-, mdh-2, rph-1) (Hansen and Juni  ##REF## ; LeMaster and Kushlan  ##UREF## ) and K10-15-16 (CGSC # 4858 Hfr fhuA22, zwf-2, relA1, T2R, pfk-10) (Fraenkel  ##REF## ) (referred to here as K10zwf) and the wild-type K12 strain (Clowes and Hayes  ##UREF## ; Soupene et al.  ##REF## ) (CGSC # 4401:F+) used in this work were obtained from the Yale Coli Genetic Stock Center (CGSC). Dr. Paliy kindly provided the wild-type K12 NCM3722 (Soupene et al.  ##REF## ).</p>",
-        "<title>Isotopes</title>",
-        "<p>Isotopic labeled compounds were purchased from Cambridge Isotope Laboratory (Andover, MA) and Isotec-Sigma-Aldrich (Miamisburg, OH): [3-<sup>13</sup>C]-pyruvate (99%), <sup>15</sup>N-(NH4)<sub>2</sub>SO4 (99%).</p>",
-        "<title>Media for <italic>E</italic>. <italic>coli</italic> growth</title>",
-        "<p>Luria\u2013Bertani (LB) media was prepared as described (Sambrook and Russell  ##UREF## ) and LeMaster-Richards (LMR) minimal media was prepared as described (LeMaster and Richards  ##REF## ; Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). The LMR media contains 176\u00a0mM KH<sub>2</sub>PO<sub>4</sub>, 25\u00a0mM NaOH, 10\u00a0\u03bcl H<sub>2</sub>SO<sub>4</sub>, 12.6\u00a0mM (NH<sub>4</sub>)<sub>2</sub>SO<sub>4</sub>, 2\u00a0mM MgSO<sub>4</sub>, 10\u00a0\u03bcM FeSO<sub>4</sub> and 0.2% trace metals, supplemented with the [3-<sup>13</sup>C]-pyruvate and <sup>15</sup>N-(NH<sub>4</sub>)<sub>2</sub>SO<sub>4</sub> enriched nitrogen source.</p>",
-        "<title>Method for <italic>E</italic>. <italic>coli</italic> growth optimization</title>",
-        "<p>The growth evaluation of the <italic>E</italic>. <italic>coli</italic> mutant (DL323) was performed as described previously (Thakur et al.  ##REF## ,  ##REF## ). Briefly, a 5\u00a0ml starter culture in LeMaster media supplemented with unlabeled carbon sources was inoculated from a single colony of DL323 <italic>E</italic>. <italic>coli</italic> grown on LB agar plates at 37\u00b0C overnight (~16\u00a0h). The overnight culture was pelleted, cells were washed twice in 1\u00d7 phosphate-buffered saline (PBS), and re-suspended in fresh 5\u00a0ml of LMR medium; 1\u00a0ml of this re-suspension was added to a 50\u00a0ml culture in LMR medium for another overnight growth at 37\u00b0C. The overnight culture was pelleted prior to complete saturation of these cells, the cells were washed twice in 1\u00d7PBS, resuspended in a 50\u00a0ml of LMR medium with no carbon supplements, and then 5\u00a0ml from this resuspension was added to a 500\u00a0ml LMR medium supplemented with [3-<sup>13</sup>C]-pyruvate and <sup>15</sup>N-(NH<sub>4</sub>)<sub>2</sub>SO<sub>4</sub>, and the solution incubated at 37\u00b0C.</p>",
-        "<title>P1 nuclease digestion, nucleotide separation and purification</title>",
-        "<p>The ribonucleoside monophosphates (rNMPs) were isolated from <italic>E</italic>. <italic>coli</italic> (DL323) mutant culture as described earlier (Batey et al.  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). The nucleic acid mixture was digested with P1 nuclease and separated into individual ribo- or deoxy-monophosphates using a cis-diol boronate affinity column chromatography as described (Batey et al.  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). The site specific labeling pattern of rNMPs was verified by NMR prior to re-phosphorylation to rNTPs.</p>",
-        "<title>Enzymatic phosphorylation of rNMPs</title>",
-        "<p>The detailed procedure for enzymatic phosphorylation of rNMP was described previously (Nyholm et al.  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). Briefly, the individual rAMP, rCMP, rGMP or rUMP and kinases were freshly prepared and pre-incubated at 37\u00b0C for 10\u00a0min prior to mixing. The enzymatic conversion of the labeled rNMPs to rNTP was complete in 5\u00a0h. The progression of the phosphorylation reaction was monitored on a TARGA C18 column using conditions similar to those in the purification step (Supplementary Figure S1). Collected rNTP fractions were pooled and dialyzed overnight and lyophilized.</p>",
-        "<title>Site specific enrichment of A-Site RNA and in vitro transcription</title>",
-        "<p>A-Site RNA (5\u2032-GGC GUC ACA CCU UCG GGU GAA GUC GCC-3\u2032) was synthesized using in vitro transcription (Milligan et al.  ##REF## ; Milligan and Uhlenbeck  ##REF## ), with a His-tagged mutant (P266L) T7 RNA polymerase (Guillerez et al.  ##REF## ) using a mixture of uniformly labeled guanine and cytosine triphosphates (rGTP and rCTP) (Sigma Aldrich) or site specific labeled rGTP and rCTP. Another set of samples were made using a mixture of uniformly labeled adenosine triphosphates (rATP) (Sigma Aldrich) or site specific labeled rATP. Two terminal 2\u2032-O-methyl modifications in the template strand were introduced to substantially reduce the amount of transcripts with extra nucleotides at the 3\u2032-end (Kao et al.  ##REF## ). The optimal transcription conditions were as follows: 10\u00a0mM total NTP and 15\u00a0mM\u00a0Mg<sup>2+</sup>. The reactions were carried out in a transcription buffer A (40\u00a0mM Tris\u2013HCl, pH 8.1, 1\u00a0mM spermidine, 5\u00a0mM dithiothreitol (DTT), 0.01% Triton X-100, 80\u00a0mg/ml PEG 8000), 300 nM each DNA strand, and 1.5\u00a0\u03bcl of 2.4\u00a0mg/ml T7 RNAP (optimized amount) per 40\u00a0\u03bcl of transcription volume. After three hours of incubation at 310\u00a0K, the RNA from the transcription reactions were purified and dialyzed as described (Dayie  ##REF## ). After dialysis, the RNA was lyophilized, and resuspended into NMR buffer (100\u00a0mM KCl, 10\u00a0mM potassium phosphate pH 6.2, 8% or 100% D<sub>2</sub>O), with or without 3\u00a0mM MgCl<sub>2</sub>, and a trace of sodium azide). The concentration of each NMR sample in Shigemi tubes in 250\u00a0\u03bcl of buffer was ~0.1\u00a0mM.</p>",
-        "<title>NMR experiments</title>",
-        "<p>All NMR experiments were run on a four channel Bruker Avance III 600\u00a0MHz spectrometer equipped with actively shielded z-axis gradient triple resonance probe. The experiments were conducted at temperatures ranging from 15\u00b0 to 45\u00b0C. The NMR data sets were processed and the peak positions and intensities were analyzed with Bruker\u2019s TOPSPIN 2.1 as described previously (Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). One dimensional (1D) <sup>13</sup>C spectra and two-dimensional non-constant-time (<sup>1</sup>H, <sup>13</sup>C) heteronuclear single quantum correlation (HSQC) spectra (Bodenhausen and Ruben  ##UREF## ; Bax et al.  ##UREF## ) were acquired to analyze the rNMP fractions extracted from each bacterial strain. The fractional <sup>13</sup>C enrichment at each carbon site was quantified directly by 1D proton methods or indirectly using 2-bond (<sup>2</sup>J<sub>HN</sub>) HSQC as described previously (Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ).</p>",
-        "<p>To demonstrate the usefulness of site selective labeling 2D <sup>1</sup>H-<sup>13</sup>CH<sub>2</sub> TROSY (Meissner et al.  ##UREF## ; Miclet et al.  ##REF## ) spectra were measured using non-constant time evolution in the carbon dimension. Methylene CH<sub>2</sub>-TROSY experiments (Miclet et al.  ##REF## ) were run with the following modifications that dispense with selective decoupling: the WURST-4 decoupling waveform was omitted during the carbon t<sub>1</sub> evolution period, and a non-selective 180\u00b0 pulse replaced the selective pulse on the C5\u2032 carbon. To compare the usefulness of the modified CH<sub>2</sub>-TROSY experiments, the normal CH<sub>2</sub>-methylene optimized HSQC (Schleucher et al.  ##REF## ; Sattler et al.  ##UREF## ) were also run under identical conditions: for each data set, 16 scans and 1024 complex points were collected in t<sub>2</sub> and 512 complex points were collected in t<sub>1</sub> using the Echo-Anti echo method (Palmer et al.  ##UREF## ; Kay et al.  ##UREF## ) for quadrature detection; proton and carbon carrier was placed at 4.7 and 64\u00a0ppm, respectively, as described (Thakur et al.  ##REF## ,  ##REF## ). The time domain data sets were zero filled in t<sub>1</sub> and t<sub>2</sub> before Fourier transformation to give a final real matrix size of 2,048\u00a0\u00d7\u00a01,024 points.</p>",
-        "<p>Longitudinal (R<sub>1</sub>) relaxation rates were measured for ribose C1\u2032 and cytosine nucleobase C5 carbons using TROSY detected experiments (Hansen and Al-Hashimi  ##REF## ). To compare the usefulness of the selective labels relative to uniform labels, the R<sub>1</sub> relaxation experiments were run under identical conditions: for each data set, 64 scans and 1024 complex points were collected in t<sub>2</sub>, and 128 complex points were collected in t<sub>1</sub> using the Echo-Anti echo method (Palmer et al.  ##UREF## ; Kay et al.  ##UREF## ) for quadrature detection. Proton and carbon carrier was placed at 4.7 and 93\u00a0ppm respectively. The time domain data sets were zero filled in t<sub>1</sub> and t<sub>2</sub> before Fourier transformation to give a final real matrix size of 2,048\u00a0\u00d7\u00a01,024 points. For the 2D longitudinal (R<sub>1</sub>) experiments, relaxation delays of 21.1 (2x), 61.4, 141.1, 301.2, 381.3, 461.3 (2x), 781.6 ms were used. The longitudinal relaxation decay curves were fitted using the Levenberg\u2013Marquardt algorithm by assuming monoexponential decay.</p>"
-    ],
-    "results": [
-        "<title>Results</title>",
-        "<title>Incorporation of <sup>13</sup>C into ribose and base of nucleotides</title>",
-        "<p>To test the hypothesis that an asymmetric carbon source such as [3-<sup>13</sup>C]-pyruvate will provide superior site-selective labeling, we compared the growth of wild type and two mutant <italic>E</italic>. <italic>coli</italic> strains on [3-<sup>13</sup>C]-pyruvate (Fig.\u00a0 ##FIG## , Table\u00a0 ##TAB## ). To place our results in context, we outline nucleotide metabolism in <italic>E</italic>. <italic>coli</italic> via glycolysis, gluconeogenesis, and the Kreb cycle using pyruvate as the sole carbon source. If glucose is used as the sole carbon source, the conversion of PEP (phosphoenolpyruvate) to pyruvate by pyruvate kinase is an irreversible reaction under physiological conditions (Fig.\u00a0 ##FIG## ). However, under minimal growth conditions in <italic>E. coli</italic> using pyruvate as the sole carbon source, PEP synthetase (PPS) can convert pyruvate directly to PEP (Fig.\u00a0 ##FIG## ), and PEP carboxylase (PPC) can convert PEP to an oxaloacetate intermediate (Chao et al.  ##REF## ; Schilling et al.  ##UREF## ). In eukaryotes, pyruvate is converted to PEP via a two-step process with an oxaloacetate intermediate catalyzed by pyruvate carboxylase and PEP carboxykinase. However, pyruvate carboxylase is a mitochondrial protein not operative in <italic>E</italic>. <italic>coli</italic>, and so we expect all PEP will be derived from the externally supplied pyruvate without dilution by the TCA cycle for the DL323 <italic>E</italic>. <italic>coli</italic> strain. Consequently purine and pyrimidine moieties derived from oxaloacetate and serine should have enrichment levels quite near that of the initial pyruvate. Similarly the labeling pattern in the ribose ring derived from PEP, which is not diluted by the TCA cycle, should label only C1\u2032 and C5\u2032 in 1:2 ratio and all other sites should have minimal label. ##FIG## \n ##TAB## \n</p>",
-        "<p>In agreement with this metabolic analysis (Fig.\u00a0 ##FIG## ) and for DL323 <italic>E</italic>. <italic>coli</italic> grown on [3-<sup>13</sup>C]-pyruvate, the ribose ring is labeled only in the C1\u2032 and C5\u2032 carbon positions at ~42% and 98% respectively (Fig.\u00a0 ##FIG## , Table\u00a0 ##TAB## ). Thus these two positions remain singlet. This labeling pattern is very different from what was observed for DL323 <italic>E</italic>. <italic>coli</italic> grown on [1, 3-<sup>13</sup>C]-glycerol (Johnson and Hoogstraten  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). In this case of growth on glycerol, the ribose ring is labeled in all but the C4\u2032 carbon position, so that the C2\u2032 and C3\u2032 positions suffer from multiplet splitting by carbon\u2013carbon coupling and the C1\u2032 position has some admixture of residual C1\u2032-C2\u2032 coupling (Thakur et al.  ##REF## ,  ##REF## ). ##FIG## \n</p>",
-        "<p>Similar to DL323, K10zwf <italic>E</italic>. <italic>coli</italic> grown on [3-<sup>13</sup>C]-pyruvate leads to exclusive labeling of the C1\u2032 and C5\u2032 carbon position without <sup>13</sup>C-<sup>13</sup>C coupling evident (Fig.\u00a0 ##FIG## ). Importantly, in this case the abundance of <sup>13</sup>C label at C1\u2032 is higher than that obtained with DL323 (Table\u00a0 ##TAB## ). Again similar to DL323 and K10zwf <italic>E</italic>. <italic>coli</italic>, wildtype K12 grown on [3-<sup>13</sup>C]-pyruvate also leads to labeling of the C1\u2032 and C5\u2032 carbon positions, but unlike the other two strains, sufficient residual labeling of the C2\u2032 and C3\u2032 carbon atoms leads to <sup>13</sup>C-<sup>13</sup>C coupling (Fig.\u00a0 ##FIG## ,  ##FIG## ). This residual labeling presumably arises from PEP derived from the mixture of [1, 2, 3-<sup>13</sup>C]-oxaloacetate and [2, 3, 4-<sup>13</sup>C]-oxaloacetate intermediates (Fig.\u00a0 ##FIG## ).</p>",
-        "<p>The dramatic improvement of the labeling afforded by using [3-<sup>13</sup>C]-pyruvate is most visible in the base region. For nucleotides derived from K10zwf and K12 cultures, both the protonated C5 and C6 carbon positions of pyrimidines are substantially labeled (\u226560%) because the flux through the TCA cycle is efficient in these two <italic>E</italic>. <italic>coli</italic> strains. This efficient flux also ensures that the purine C4 and C5 positions are labeled. A single pass through the TCA cycle leads to an equal probability of carrying the <sup>13</sup>C label into either the pyrimidine C5 or C6 position because the oxaloacetate is generated by [1, 3-<sup>13</sup>C]-malate or [2, 4-<sup>13</sup>C]-malate (Fig.\u00a0 ##FIG## ). If the precursor is [1, 3-<sup>13</sup>C]-oxaloacetate, then the desired label at pyrimidine C5 or C6 is diluted after the first and subsequent rounds of the TCA cycle (Fig.\u00a0 ##FIG## ). This dilution of labeling is observed for the nucleotides derived from the cultures of <italic>E</italic>. <italic>coli</italic> K10zwf and K12 strains (Table\u00a0 ##TAB## ). Again, if the precursor is [2, 4-<sup>13</sup>C]-oxaloacetate, then the second pass through the TCA cycle leads to a [2, 3-<sup>13</sup>C]-oxaloacetate and subsequent cycles leads to a mixture of [1, 2, 3-<sup>13</sup>C]-oxaloacetate and [2, 3, 4-<sup>13</sup>C]-oxaloacetate. The net effect of the TCA cycle is to thoroughly scramble the labels arising from oxaloacetate such that, in addition to the desired selective label at either the pyrimidine C5 or C6, undesired pyrimidine C5-C6 or C4-C5-C6 labeled spin pairs would be obtained. Thus the significant coupling observed between pyrimidine C4 and C5 or C5 and C6 (Fig.\u00a0 ##FIG## ; Supplementary Figure S2) is a consequence of this scrambling by the TCA cycle. These labeling patterns, similar to those observed with [1, 3-<sup>13</sup>C]-glycerol grown on DL323, have the undesirable consequence of substantive <sup>13</sup>C-<sup>13</sup>C couplings that degrade the resolution and the accuracy of measured relaxation parameters (as shown below). In contrast to K10zwf and K12, <italic>E</italic>. <italic>coli</italic> strain DL323 provides a much cleaner labeling pattern that is devoid of unwanted labeling at the pyrimidine C6 and purine C4 and C5 positions (Fig.\u00a0 ##FIG## ; Supplementary Figure S2). For the DL323 strain, the flux through the TCA cycle is reduced essentially to zero, thereby preventing the dilution of the labels arising from oxaloacetate described above. ##FIG## \n</p>",
-        "<p>As expected, all of the C2 and C8 atoms of purine are substantially labeled to &gt;90% for nucleotides derived from DL323 and K12 cultures, but those from K10zwf have these labels reduced to 85 and 57% respectively partly because the scrambling from the TCA cycle apparently feeds more efficiently into the purine biosynthetic pathway for K10zwf than K12 (Figs.\u00a0 ##FIG## ,  ##FIG## ; Supplementary Figure S2). The origin of these differences is not clear. Nonetheless, some of the scrambled malate or oxaloacetate from the TCA cycle is likely fed back into PEP such that purine C4, purine C5, and pyrimidine C6 positions are labeled in K10zwf and K12 but not DL323 (Fig.\u00a0 ##FIG## ). As a result, K10zwf and K12 provide <sup>13</sup>C -<sup>13</sup>C spin pairs in the purine and pyrimidine base moieties that can potentially hinder accurate extraction of relaxation parameters.</p>",
-        "<title>Applications of site selective labels for <sup>13</sup>C NMR study of nucleic acids</title>",
-        "<p>Spectra obtained using our selective labeling methodology is of high quality (Figs.\u00a0 ##FIG## ,  ##FIG## ). A 27\u2013nt A-site ribosomal RNA fragment derived from the 30 S ribosomal subunit (Purohit and Stern  ##REF## ; Fourmy et al.  ##REF## ; Kaul et al.  ##REF## ; Wirmer and Westhof  ##REF## ; Schmeing and Ramakrishnan  ##REF## ) was transcribed using either the uniformly <sup>13</sup>C/<sup>15</sup>N-labeled GTP and CTP or the site selectively-labeled GTP and CTP derived from DL323 <italic>E</italic>. <italic>coli</italic> cells grown on [3-<sup>13</sup>C]-pyruvate as described above. For the C,G- uniformly <sup>13</sup>C\u2014labeled sample, the peaks are not only broader, but they also overlap extensively (Fig.\u00a0 ##FIG## ). These unwanted splittings can also be removed using either constant time experiments, adiabatic band selective decoupling schemes, or maximum entropy reconstruction-deconstruction; as discussed previously, these have their own limitations (Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). In particular, the length of the constant time period (T) limits the acquisition times to multiples of the homonuclear coupling constant. To obtain reasonable digital resolution, large values of T are needed and this leads to significant signal attenuation for RNA molecules larger than 30 nucleotides (Dayie  ##REF## ). But with the selective labels, one is not forced to compromise resolution for improved sensitivity. ##FIG## \n ##FIG## \n</p>",
-        "<p>Use of the site selectively labeled NTPs removes the crowding due to the splitting of signals from J-coupling and eliminates the C2\u2032, C3\u2032, and C4\u2032 resonances while preserving only the C1\u2032 and C5\u2032 signals (Fig.\u00a0 ##FIG## ). In addition to improving resolution, sensitivity is also greatly enhanced. For instance, cytosine C5 has a large chemical shielding anisotropy (CSA) and so decays rapidly, and as expected most of the C5 resonances are absent in the spectra of the uniformly labeled sample. These signals are preserved in the spectra of the site selectively labeled sample (Fig.\u00a0 ##FIG## ).</p>",
-        "<p>In the C8 purine and C6 pyrimidine regions, the selective labels once again improve the resolution and sensitivity (Fig.\u00a0 ##FIG## ). By eliminating the C6 pyrimidine signals, the guanosine C8 signals can be readily identified without concerns about potential overlap with cytosine C6, especially for a sample that is labeled with all four nucleotides. These observations underscore the difference selective labeling makes on sensitivity and resolution.</p>",
-        "<p>A non-constant time version of the <sup>13</sup>CH<sub>2</sub> TROSY experiment (Miclet et al.  ##REF## ) that enables the rescue of the slowest relaxing multiplet component was next performed to demonstrate the usefulness of these site selective labels. Compared to uniform labeling, the site-selective labels shows two clear advantages of improved resolution and sensitivity (Fig.\u00a0 ##FIG## A, B) of a spectrum of a 27-nt A-Site RNA fragment labeled only in G and C. Compared to the normal CH<sub>2</sub>-methylene optimized experiment (Schleucher et al.  ##REF## ; Sattler et al.  ##UREF## ), 14 out of the expected 18 GC C5\u2032 resonances are visible in the TROSY experiment but most of these are either overlapped, or weak, or absent in the uniformly labeled sample. Thus, it is anticipated that these and other new experiments that incorporate the <sup>13</sup>CH<sub>2</sub> TROSY module can be designed to probe RNA-ligand interactions at very high resolution using the site specific labels described here. ##FIG## \n</p>",
-        "<title>Site selective labeling affords more accurate relaxation rate measurements using non-constant time non-selective pulse experiments</title>",
-        "<p>To place the results of the relaxation in context, it is instructive to examine how spins are dipolar coupled to each magnetization of interest in a relaxation measurement. For instance, the longitudinal relaxation of the <sup>13</sup>C5 spin during a relaxation period T for a uniformly labeled RNA sample has the following time dependency (Yamazaki et al.  ##UREF## ; Dayie and Wagner  ##UREF## ; Boisbouvier et al.  ##REF## ; Hansen and Al-Hashimi  ##REF## ): ##FORMU## \u0394C<sup>i</sup> are the deviations of each carbon\u2019s longitudinal magnetization from its equilibrium magnetization, \u0394H5 is the deviation of the H5 proton magnetization from its equilibrium magnetization, and the autorelaxation and cross relaxation rates are given by R<sub>C5-X</sub>(z) and R<sub>X</sub>(x\u00a0\u2192\u00a0C5) respectively. Similar equations can be written for the ribose C1\u2032 spin and all its dipolar coupled neighbors. These rates are given as follows (Eldho and Dayie  ##REF## ): ##FORMU## \n ##FORMU## \n ##FORMU## \n ##FORMU## \n ##FORMU##  and \u03941\u00a0=\u00a0\u03c3xx\u00a0\u2212\u00a0\u03c3zz and \u03942\u00a0=\u00a0\u03c3yy\u00a0\u2212\u00a0\u03c3zz.</p>",
-        "<p>Here X\u00a0=\u00a0C4 or C6. The physical constants \u03b3<sub>H</sub> and \u03b3<sub>C</sub> are the gyromagnetic ratios of <sup>1</sup>H and <sup>13</sup>C respectively (\u03b3<sub>C</sub>\u00a0=\u00a06.7283\u00a0\u00d7\u00a010<sup>7</sup> rad (Ts)<sup>\u22121</sup>, \u03b3<sub>H</sub>\u00a0=\u00a026.75\u00a0\u00d7\u00a010<sup>7</sup> rad (Ts)<sup>\u22121</sup>), r<sub>HC</sub> is the length of the internuclear <sup>1</sup>H-<sup>13</sup>C bond vector, r<sub>CC</sub> is the length of the internuclear <sup>13</sup>C-<sup>13</sup>C bond vector, and  ##FORMU##  is Planck\u2019s constant divided by 2\u03c0 (1.054592\u00a0\u00d7\u00a010<sup>\u221227</sup>). An asymmetric chemical shielding tensor can be decomposed into an isotropic and two axially symmetric anisotropic terms with the symmetry axes arbitrarily chosen as the two principal axes along X and Y, as discussed previously for carbonyl backbone motional analysis (Dayie and Wagner  ##UREF## ). More general expressions for anisotropic motion can be found in the literature (Werbelow  ##UREF## ). Because each term J(\u03c9<sub>X</sub>\u00a0\u2212\u00a0\u03c9<sub>C</sub>) in ( ##FORMU## ) and ( ##FORMU## ) is proportional to J(0) which scales linearly with molecular weight, it is clear that the dipole\u2013dipole contribution cannot be neglected for uniformly labeled RNA samples.</p>",
-        "<p>We expected therefore for the case of a uniformly labeled cytosine or uridine, the H5, C4, C6 spins will contribute to the longitudinal relaxation of the <sup>13</sup>C5 nucleus, and H1\u2032 and C2\u2032 spins will contribute to the relaxation of the <sup>13</sup>C1\u2032 nucleus. For the site selectively labeled nucleotides, only H5 will contribute to the longitudinal relaxation of the <sup>13</sup>C5 nucleus, and only H1\u2032 will contribute to the relaxation of the <sup>13</sup>C1\u2032 nucleus. Thus, the labeling pattern of rNMPs derived from DL323 grown on [3-<sup>13</sup>C]-pyruvate is particularly attractive for relaxation studies of the ribose C1\u2032 and the base C5 (as well as C2 and C8 positions that are not discussed further here). Each of these positions is essentially singlet. The isolation of the C1\u2032 ribose and the C5 base positions from labeled adjacent neighbors means that the interference described above that arises from strong <sup>13</sup>C-<sup>13</sup>C magnetic interactions in the base and ribose rings is no longer a hindrance for extracting accurate relaxation parameters. Therefore as a third demonstration of the usefulness of the site selectively labeled versus the uniformly labeled nucleotides for quantifying dynamics in RNA, R<sub>1</sub> experiments were carried on the 27\u2013nt A-site ribosomal RNA fragment transcribed using uniformly <sup>13</sup>C/<sup>15</sup>N-labeled GTP and CTP and site selectively-labeled GTP and CTP derived from DL323 <italic>E</italic>. <italic>coli</italic> cells grown on [3-<sup>13</sup>C]-pyruvate. Examples of <sup>13</sup>C R<sub>1</sub> decay curves for both ribose C1\u2032 and base carbon C5 sites are depicted in Fig.\u00a0 ##FIG##  and Supplementary Figure S4. The decay curves fit well to a mono-exponential function for the C1\u2032 position of a site selective ATP- labeled sample (Fig.\u00a0 ##FIG## A), but not as well for a uniform ATP-labeled sample (Fig.\u00a0 ##FIG## B). The discrepancy between R<sub>1</sub> measured for uniform (2.0\u00a0\u00b1\u00a00.2\u00a0s<sup>\u22121</sup>) and site-selective (1.8\u00a0\u00b1\u00a00.08\u00a0s<sup>\u22121</sup>) labeled sample is 0.2\u00a0s<sup>\u22121</sup> (Fig.\u00a0 ##FIG## A). Similarly, the fit to a monoexponential decay function is only applicable to the site selectively CTP-labeled RNA sample (Fig.\u00a0 ##FIG## C). The fit to the uniformly CTP-labeled sample is poor. Again fit to a monoexponential decay function is only applicable to the C5 position of the site selectively CTP-labeled RNA sample. This is discrepancy, also worse for the C5 carbons with relaxation rates of 3.4 \u00b1\u00a01.0\u00a0s<sup>\u22121</sup> for the uniformly labeled sample and 2.9\u00a0\u00b1\u00a00.4\u00a0s<sup>\u22121</sup> for the selectively labeled sample, is partly because of the expected dipolar coupled interactions, but also because of substantial decrease in signal-to noise for the uniformly labeled sample (Supplementary Figure S4). These observations suggest that the contributions from the cross-relaxation terms in ( ##FORMU## ), ( ##FORMU## ), and ( ##FORMU## ) are not negligible for the uniformly labeled sample. ##FIG## \n</p>"
-    ],
-    "discussion": [
-        "<title>Discussion</title>",
-        "<p>Previous labeling technologies using wild type <italic>E</italic>. <italic>coli</italic> strains and symmetric carbon sources spurred the development of new solution NMR methods to characterize the structure and dynamics of small to medium sized uniformly labeled RNA molecules (Batey et al.  ##REF## ; Hall  ##REF## ; F\u00fcrtig et al.  ##REF## ; Latham et al.  ##REF## ; Dayie  ##REF## ; Lu et al.  ##REF## ). A number of alternative labeling approaches have been proposed since then, and it is important therefore to discuss the advantages and limitations of these different methods. First, uniform labeling introduces direct one-bond and residual dipolar couplings that prevents accurate measurement of <sup>13</sup>C relaxation rates (Johnson et al.  ##REF## ; Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ) and decreases the resolution and sensitivity of NMR experiments. For instance, growth of wildtype <italic>E</italic>. <italic>coli</italic> strain K12 on [1-<sup>13</sup>C]-acetate yields <sup>13</sup>C label at the ribose C1\u2032, C2\u2032, and C3\u2032 atomic sites but very little label at C4\u2032 and C5\u2032 (Hoffman and Holland  ##REF## ). As a result, C2\u2032 and C3\u2032 have substantial multiplet structure and even C1\u2032 has residual multiplet structure (Supplementary Figure S3). In addition, the protonated aromatic carbons (purine C2 and C8 and pyrimidine C5 and C6) are not labeled, making this particular labeling pattern less useful for routine NMR applications. In contrast, growth of wildtype <italic>E</italic>. <italic>coli</italic> strain K12 on [2-<sup>13</sup>C]-acetate yields <sup>13</sup>C label at all five ribose carbon positions of C1\u2032, C2\u2032, C3\u2032, C4\u2032 and C5\u2032 (Hoffman and Holland  ##REF## ). Since each site is highly enriched, all five positions have substantial multiplet structure (Supplementary Figure S3). Also for the aromatic carbons, the protonated carbon C5 and C6 of pyrimidines and C2 and C8 of purines are labeled at very high level (Hoffman and Holland  ##REF## ; Supplementary Figure S3). Similarly, growth of wildtype <italic>E</italic>. <italic>coli</italic> strain K12 on [2-<sup>13</sup>C]-glucose yields <sup>13</sup>C label only at C1\u2032, C2\u2032, and C4\u2032. No label is observed at C3\u2032 and C5\u2032. For the aromatic carbons, the protonated carbon C5 and C6 of pyrimidine and C2 and C8 of purine are also labeled at very high level (Hoffman and Holland  ##REF## ). These second set of labeling patterns are therefore very useful for NMR applications that require substantial coupling for transfer of coherences, but problematic for applications that require suppression of these coupling networks (Supplementary Figure S3 E).</p>",
-        "<p>To remove these coupling networks, alternate <sup>13</sup>C-<sup>12</sup>C labeling schemes have been shown to be important for tackling these difficulties in RNA molecules (Johnson et al.  ##REF## ; Johnson and Hoogstraten  ##REF## ; Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). However, some of the earlier schemes had their unique problems. For instance, the labeling pattern of rNTPs derived from DL323 grown on [1,3-<sup>13</sup>C]-glycerol is not attractive for relaxation studies because the ribose C2\u2032 position is doublet, the C1\u2032 retains some residual doublet arising from <sup>13</sup>C2\u2032-<sup>13</sup>C1\u2032 isotopomers, and the C4\u2032 has no label (Thakur et al.  ##REF## ,  ##REF## ). In the base region the pyrimidine C5 site has multiplet structure, again precluding its use for accurate relaxation measurements (Thakur et al.  ##REF## ,  ##REF## ). The use of G6PDH mutant strains rather than DL323 using 2-<sup>13</sup>C-glycerol was also recommended in previous work (Johnson et al.  ##REF## ).</p>",
-        "<p>Recently de novo biosynthetic methods have been proposed that potentially resolve these problems (Schultheisz et al.  ##REF## ,  ##REF## ). However, the new method requires 29 enzymes for making ATP and GTP and 18 enzymes for making CTP and UTP. Of these enzymes only a few are commercially available, which makes this method excellent for adaptation by companies but not by individual laboratories. Additionally, the de novo method requires selectively labeled glucose and serine which are not readily available commercially, or prohibitively expensive, or both. For instance the de novo method of labeling ribose C1\u2032 and C5\u2032 in the context of purine C2, C5, and C8, indicated above by our method, requires <sup>13</sup>C-2,6-glucose that is not commercially available and <sup>13</sup>C-3-serine that costs ~$4000/g; the biomass method is able to make these labels readily.</p>",
-        "<p>Of greater interest the main advantage of using the <italic>E</italic>. <italic>coli</italic> strain DL323 grown on a non-symmetric carbon source such as [3-<sup>13</sup>C]-pyruvate is that it provides isolated two-spin systems that are ideal for relaxation measurements as well as improving resolution and sensitivity. For example in the ribose region, the C1\u2032 and C5\u2032 labels are completely isolated from C2\u2032 and C4\u2032 spins respectively. Similarly in the base region, the purine C2 and C8 and pyrimidine C5 spins are isolated from the network of carbon spins within the aromatic ring. For instance in a uniformly labeled sample, adenine C2 has substantial <sup>2</sup>J<sub>C2C5</sub> coupling of 11\u00a0Hz to the C5 carbon, and the purine C8 carbons have <sup>2</sup>J<sub>C4C8</sub> coupling of\u00a0~9\u00a0Hz to the C5 carbon (Wijmenga and van Buuren  ##UREF## ). In addition, the carbon atoms C4 (149\u2013154\u00a0ppm)/C5 (116\u2013120\u00a0ppm)/C6 (156\u2013161\u00a0ppm)/C8 (131\u2013142\u00a0ppm) resonate within a chemical shift range that precludes effectively decoupling C8 from the C4, C5, or C6 carbon atoms. These complications again prevent accurate relaxation measurements using CPMG experiments. Formate supplementation has been proposed as a potential cost-effective solution (Latham et al.  ##REF## ; Dayie and Thakur  ##REF## ; Thakur et al.  ##REF## ,  ##REF## ). However, the C8 is labeled to ~90% and C2 to only ~25%, limiting the general usefulness of formate labeling. Thus the confounding effects of scalar coupling from adjacent labeled sites are eliminated using the current method proposed in this work, unlike the case for previous labels.</p>"
-    ],
-    "conclusion": [
-        "<title>Conclusions</title>",
-        "<p>Any RNA sequence can be site specifically labeled with site alternate <sup>13</sup>C-enriched isotopes efficiently prepared from site-specifically labeled rNTPs using in vitro transcription by T7 RNA polymerase. We have presented an efficient method for preparing these ribonucleotides containing site specific distributions of <sup>13</sup>C enrichment that has significant advantages in terms of cost, flexibility, resolution and sensitivity over previously reported procedures. These labels will be particularly beneficial to investigators using heteronuclear NMR spectroscopy to study the structure and dynamics of RNA and RNA complexes of increasingly large sizes, currently an important and active area of research.</p>"
-    ],
-    "front": [
-        "<p>Selective isotopic labeling provides an unparalleled window within which to study the structure and dynamics of RNAs by high resolution NMR spectroscopy. Unlike commonly used carbon sources, the asymmetry of <sup>13</sup>C-labeled pyruvate provides selective labeling in both the ribose and base moieties of nucleotides using <italic>E</italic>. <italic>coli</italic> variants, that until now were not feasible. Here we show that an <italic>E</italic>. <italic>coli</italic> mutant strain that lacks succinate and malate dehydrogenases (DL323) and grown on [3-<sup>13</sup>C]-pyruvate affords ribonucleotides with site specific labeling at C5\u2032 (~95%) and C1\u2032 (~42%) and minimal enrichment elsewhere in the ribose ring. Enrichment is also achieved at purine C2 and C8 (~95%) and pyrimidine C5 (~100%) positions with minimal labeling at pyrimidine C6 and purine C5 positions. These labeling patterns contrast with those obtained with DL323 <italic>E</italic>. <italic>coli</italic> grown on [1, 3-<sup>13</sup>C]-glycerol for which the ribose ring is labeled in all but the C4\u2032 carbon position, leading to multiplet splitting of the C1\u2032, C2\u2032 and C3\u2032 carbon atoms. The usefulness of these labeling patterns is demonstrated with a 27-nt RNA fragment derived from the 30S ribosomal subunit. Removal of the strong magnetic coupling within the ribose and base leads to increased sensitivity, substantial simplification of NMR spectra, and more precise and accurate dynamic parameters derived from NMR relaxation measurements. Thus these new labels offer valuable probes for characterizing the structure and dynamics of RNA that were previously limited by the constraint of uniformly labeled nucleotides.</p>",
-        "<title>Electronic supplementary material</title>",
-        "<p>The online version of this article (doi:10.1007/s10858-011-9581-6) contains supplementary material, which is available to authorized users.</p>",
-        "<title>Keywords</title>"
-    ],
-    "body": [
-        "<title>Electronic supplementary material</title>",
-        "<p>Below is the link to the electronic supplementary material.\n ##SUPPL## \n</p>"
-    ],
-    "back": [
-        "<p>This work was supported in part by the University of Maryland Nano-Biotechnology Award and the National Institutes of Health grant GM077326 to T.K.D. We are grateful to Professors Jon Dinman and Steve Rokita for helpful discussions.</p>",
-        "<title>Open Access</title>",
-        "<p>This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.</p>"
-    ],
-    "figure": [
-        "<fig id=\"Fig1\"><label>Fig.\u00a01</label><caption><p>Major metabolic pathways involved in the production of nucleic acid nucleotides from pyruvate, including key steps in glycolysis, gluconeogenesis and several passes through the tricarboxylic (TCA) cycle as derived from Covert and Palsson ( ##REF## ). With the <italic>E</italic>. <italic>coli</italic> strain lacking succinate and malate dehydrogenases (DL323), the oxidative branch of the pentose phosphate pathway remains intact but the TCA cycle is severed in two places such that the oxaloacetate is derived exclusively from carboxylation of phosphoenolpyruvate (PEP) and the resulting label is not diluted by the TCA cycle. Atom labels for the terminal (3) carbon (magenta and <italic>thin</italic>\n<italic>circle</italic>), the central (2) carbon (<italic>square</italic>), and the terminal (1) carbon (<italic>triangle</italic>) of pyruvate are highlighted. Positions that are enriched due to the presence of <sup>13</sup>CO<sub>2</sub> in the growth medium are shown with a <italic>circled</italic> x. Pyrimidine base derived from the oxaloacetate (OAA) produced by carboxylation of PEP is shown via the aspartate intermediate. This OAA cannot be used as a substrate in the first and subsequent rounds of the TCA cycle because of the two mutations. Consequently OAA derived aspartate amino acid can be produced with <sup>13</sup>C labeling at only the C<sup>\u03b1</sup> (C6) position if [2-<sup>13</sup>C]-pyruvate is used. If [3-<sup>13</sup>C]-pyruvate is used only C<sup>\u03b2</sup> (C5) position is labeled. In either case carboxylation of PEP leads to labeling of the C<sup>\u03b3</sup> (C4) position. Similarly the labeling pattern of purines from glycine derived from 3PG are shown such that if [2-<sup>13</sup>C]-pyruvate is used only the C<sup>\u03b1</sup> position of Gly and therefore C5 position of the purine ring is labeled. Otherwise if [3-<sup>13</sup>C]-pyruvate is used the CO of Gly and therefore C4 of purine ring is labeled, and the labeling of the C<sup>\u03b2</sup> position of Ser also leads to labeling of the purine C2 and C8 positions</p></caption> ##GRAPH## </fig>",
-        "<fig id=\"Fig2\"><label>Fig.\u00a02</label><caption><p>2D non-constant time HSQC spectra of all four labeled nucleotides extracted from K10zwf (<italic>blue</italic> contours, left shifted) or K12 (<italic>red</italic> contours) or DL323 (purple contours, right shifted) <italic>E</italic>. <italic>coli</italic> strains grown on [3-<sup>13</sup>C]-pyruvate. Growth on [3-<sup>13</sup>C]-pyruvate results in label at only C1\u2032 and C5\u2032 for DL323 and K10zwf strains, whereas growth of K12 leads to some residual label at C2\u2032, C3\u2032, and C4\u2032 in addition to the major labels at the C1\u2032 and C5\u2032 carbons. The cytosine (Cyt) and Uracil (Ura) C5 resonances at 96.67 and 102.69\u00a0ppm respectively are folded into the spectrum</p></caption> ##GRAPH## </fig>",
-        "<fig id=\"Fig3\"><label>Fig.\u00a03</label><caption><p>Labeling pattern of a mixture of four rNMPs isolated from K10zwf, K12, and DL323 <italic>E</italic>. <italic>coli</italic> grown on a [3-<sup>13</sup>C]-Pyruvate background, and DL323 <italic>E</italic>. <italic>coli</italic> grown on a [1, 3-<sup>13</sup>C]-glycerol background. The direct carbon detection 1D spectrum shows all the labeled carbon positions. A long recycle delay of 5\u00a0s was used to allow for sufficient magnetization recovery and proton decoupling was limited to the acquisition period only. The residual <sup>13</sup>C-<sup>13</sup>C coupling observed in a [1, 3-<sup>13</sup>C]-glycerol is absent with the [3-<sup>13</sup>C]-Pyruvate only in DL323; K10zwf and K12 grown on [3-<sup>13</sup>C]-Pyruvate still retain the residual <sup>13</sup>C-<sup>13</sup>C coupling</p></caption> ##GRAPH## </fig>",
-        "<fig id=\"Fig4\"><label>Fig.\u00a04</label><caption><p>NMR spectra showing enhanced resolution afforded by site selective labeling of A-Site RNA bound to paromomycin. The experiments were performed on the A-Site RNA <bold>A</bold> site-selectively <sup>13</sup>C-GTP and CTP labeled and <bold>B</bold> uniformly <sup>13</sup>C-GTP and CTP labeled. 2D non-constant time HSQC spectra of the ribose region. The cytosine (Cyt) and Uracil (Ura) C5 resonances at 96.67 and 102.69\u00a0ppm respectively are folded into the spectrum and are <italic>boxed</italic> to highlight the reduced signal from the uniformly labeled sample. The C2\u2032, C3\u2032, and C4\u2032 regions are <italic>boxed</italic> to highlight the absence of labeling in selectively labeled sample. Blown-up views of the C2\u2032, C3\u2032, and C4\u2032 regions boxed in (<bold>A</bold>) and (<bold>B</bold>) are depicted in (<bold>C</bold>) and (<bold>D</bold>) respectively</p></caption> ##GRAPH## </fig>",
-        "<fig id=\"Fig5\"><label>Fig.\u00a05</label><caption><p>NMR spectra showing enhanced resolution afforded by site selective labeling of A-Site RNA bound to paromomycin. The experiments were performed on the A-Site RNA <bold>A</bold> site-selectively <sup>13</sup>C-GTP and CTP labeled and <bold>B</bold> uniformly <sup>13</sup>C-GTP and CTP labeled. 2D non-constant time HSQC spectra of the base region. The cytosine (Cyt) and Uracil (Ura) C6 resonances are boxed to highlight the absence of labeling in selectively labeled sample</p></caption> ##GRAPH## </fig>",
-        "<fig id=\"Fig6\"><label>Fig.\u00a06</label><caption><p>CH<sub>2</sub>-TROSY spectra depicting the C5\u2032 region of A-Site RNA bound to paromomycin. Experiments were performed on the A-Site RNA <bold>A</bold> site-selectively <sup>13</sup>C-GTP and CTP labeled and <bold>B</bold> uniformly <sup>13</sup>C-GTP and CTP labeled. Spectra are plotted at identical contour levels. Notice the dramatic gain in sensitivity and resolution with the site selectively labeled sample</p></caption> ##GRAPH## </fig>",
-        "<fig id=\"Fig7\"><label>Fig.\u00a07</label><caption><p>Representative longitudinal R<sub>1</sub> relaxation decay curves showing marked deviation from monoexponential decay for uniformly labeled samples. Ribose C1\u2032 R<sub>1</sub> relaxation measurements at 25\u00b0C for the A-site RNA labeled with <bold>A</bold> site selectively-labeled ATP and <bold>B</bold> uniformly <sup>13</sup>C/<sup>15</sup>N-labeled ATP. Ribose C1\u2032 R<sub>1</sub> relaxation measurements at 25\u00b0C for A-site RNA labeled with <bold>C</bold> site selectively-labeled GTP and CTP and <bold>D</bold> uniformly <sup>13</sup>C/<sup>15</sup>N-labeled GTP and CTP</p></caption> ##GRAPH## </fig>"
-    ],
-    "table": [
-        "<table-wrap id=\"Tab1\"><label>Table\u00a01</label><caption><p>\n<sup>13</sup>C enrichment levels at various carbon positions within ribonucleotides harvested from K12, DL323, and K10zwf <italic>E</italic>. <italic>coli</italic> strains grown on [3-<sup>13</sup>C]-pyruvate</p></caption><table frame=\"hsides\" rules=\"groups\"><thead><tr><th align=\"left\">Carbon position labeled</th><th align=\"left\">DL323 <italic>E</italic>. <italic>coli</italic> strain</th><th align=\"left\">K10zwf <italic>E</italic>. <italic>coli</italic> strain</th><th align=\"left\">K12 <italic>E</italic>. <italic>coli</italic> strain</th></tr></thead><tbody><tr><td align=\"left\" colspan=\"4\">Purine<sup>a</sup>\n</td></tr><tr><td align=\"left\">\u00a0Ade C2</td><td align=\"left\">99\u00a0\u00b1\u00a01</td><td align=\"left\">85\u00a0\u00b1\u00a04</td><td align=\"left\">95\u00a0\u00b1\u00a02</td></tr><tr><td align=\"left\">\u00a0C8</td><td align=\"left\">93\u00a0\u00b1\u00a01</td><td align=\"left\">57\u00a0\u00b1\u00a01</td><td align=\"left\">98\u00a0\u00b1\u00a02</td></tr><tr><td align=\"left\" colspan=\"4\">Pyrimidine<sup>a</sup>\n</td></tr><tr><td align=\"left\">\u00a0C5</td><td align=\"left\">98\u00a0\u00b1\u00a04</td><td align=\"left\">81\u00a0\u00b1\u00a04</td><td align=\"left\">91\u00a0\u00b1\u00a01</td></tr><tr><td align=\"left\">\u00a0C6</td><td align=\"left\">15\u00a0\u00b1\u00a01</td><td align=\"left\">60\u00a0\u00b1\u00a04</td><td align=\"left\">64\u00a0\u00b1\u00a01</td></tr><tr><td align=\"left\" colspan=\"4\">Ribose</td></tr><tr><td align=\"left\">\u00a0C1\u2032<sup>a</sup>\n</td><td align=\"left\">42\u00a0\u00b1\u00a04</td><td align=\"left\">82\u00a0\u00b1\u00a02</td><td align=\"left\">83\u00a0\u00b1\u00a02</td></tr><tr><td align=\"left\">\u00a0C2\u2032<sup>b</sup>\n</td><td align=\"left\">4</td><td align=\"left\">9\u00a0\u00b1\u00a01</td><td align=\"left\">21\u00a0\u00b1\u00a01</td></tr><tr><td align=\"left\">\u00a0C3\u2032<sup>b</sup>\n</td><td align=\"left\">&lt;2</td><td align=\"left\">&lt;1</td><td align=\"left\">8</td></tr><tr><td align=\"left\">\u00a0C4\u2032<sup>b</sup>\n</td><td align=\"left\">3</td><td align=\"left\">9\u00a0\u00b1\u00a01</td><td align=\"left\">30\u00a0\u00b1\u00a01</td></tr><tr><td align=\"left\">\u00a0C5\u2032<sup>c</sup>\n</td><td align=\"left\">95\u00a0\u00b1\u00a05</td><td align=\"left\">95</td><td align=\"left\">95\u00a0\u00b1\u00a01</td></tr></tbody></table> ##FOOTN## </table-wrap>"
-    ],
-    "formula": [
-        "<disp-formula id=\"Equ1\"><label>1</label><alternatives><tex-math id=\"M1\">\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$ \\begin{gathered} \\frac{{d(\\Updelta C_{Z}^{5} )}}{dT} = - \\{ R_{C5 - H5} (z) + R_{C5 - C4} (z) + R_{C5 - C6} (z) + \\rho_{others} \\} \\Updelta C_{Z}^{5} - \\hfill \\\\ R_{C6} (C6 \\to C5)\\Updelta C_{Z}^{6} - R_{C4} (C4 \\to C5)\\Updelta C_{Z}^{4} - R_{C5} (H5 \\to C5)\\Updelta H_{Z}^{5} \\hfill \\\\ \\end{gathered} $$\\end{document}</tex-math> ##GRAPH## </alternatives></disp-formula>",
-        "<disp-formula id=\"Equ2\"><label>2</label><alternatives><tex-math id=\"M2\">\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$ {\\text{R}}_{{{\\text{C5}} - {\\text{H}}}} \\left( {\\text{z}} \\right) = A[{\\text{J}}(\\omega_{\\text{H}} - \\omega_{\\text{C}} ) + 6 {\\text{J}}(\\omega_{\\text{H}} + \\omega_{\\text{C}} )\\left] { + \\left( { 3 {\\text{A}} + B} \\right)} \\right[{\\text{J}}(\\omega_{\\text{C}} )] $$\\end{document}</tex-math> ##GRAPH## </alternatives></disp-formula>",
-        "<disp-formula id=\"Equ3\"><label>3</label><alternatives><tex-math id=\"M3\">\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$ {\\text{R}}_{{{\\text{C5}} - {\\text{X}}}} ({\\text{z}}) = A^{\\prime } [{\\text{J}}(\\omega_{\\text{X}} - \\omega_{\\text{C}} ) + 6 {\\text{J}}(\\omega_{\\text{X}} + \\omega_{\\text{C}} )] + ( 3 {\\text{A}}^{\\prime } + B)[{\\text{J}}(\\omega_{\\text{C}} )] $$\\end{document}</tex-math> ##GRAPH## </alternatives></disp-formula>",
-        "<disp-formula id=\"Equ4\"><label>4</label><alternatives><tex-math id=\"M4\">\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$ {\\text{R}}_{\\text{C5}} ({\\text{C5}} \\to {\\text{H5}}) = {\\text{A}}[ 6 {\\text{J}}(\\omega {\\text{H}} + \\omega {\\text{C}}) - {\\text{J}}(\\omega {\\text{H}} - \\omega {\\text{C}})] $$\\end{document}</tex-math> ##GRAPH## </alternatives></disp-formula>",
-        "<disp-formula id=\"Equ5\"><label>5</label><alternatives><tex-math id=\"M5\">\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$ {\\text{R}}_{\\text{X}} \\left( {{\\text{X}} \\to {\\text{C5}}} \\right) = {\\text{A}}^{\\prime } [ 6 {\\text{J}}(\\omega {\\text{C}} + \\omega {\\text{X}}) - {\\text{J}}(\\omega {\\text{X}} - \\omega {\\text{C}})] $$\\end{document}</tex-math> ##GRAPH## </alternatives></disp-formula>",
-        "<inline-formula id=\"IEq1\"><alternatives><tex-math id=\"M6\">\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$ {\\text{A}} = \\frac{{\\hbar^{2} \\gamma_{C}^{2} \\gamma_{H}^{2} }}{{4r_{HC}^{6} }},\\;{\\text{A}}^{\\prime } = \\frac{{\\hbar^{2} \\gamma_{C}^{4} }}{{4r_{CC}^{6} }},\\;{\\text{B}} = \\frac{{\\omega_{C}^{2} \\Updelta_{C}^{2} }}{3},\\quad {\\text{and}}\\quad \\Updelta_{C}^{2} = \\Updelta_{1}^{2} + \\Updelta_{2}^{2} - \\Updelta_{1} \\Updelta_{2} $$\\end{document}</tex-math> ##GRAPH## </alternatives></inline-formula>",
-        "<inline-formula id=\"IEq2\"><alternatives><tex-math id=\"M7\">\\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$ \\hbar $$\\end{document}</tex-math> ##GRAPH## </alternatives></inline-formula>"
-    ],
-    "box": [],
-    "code": [],
-    "quote": [],
-    "chem": [],
-    "supplementary": [
-        "<supplementary-material content-type=\"local-data\" id=\"MOESM1\"> ##MEDIA## </supplementary-material>"
-    ],
-    "footnote": [
-        "<table-wrap-foot><p>\n<sup>a</sup>The percentage label is calculated as an average of two methods: (i) the ratio of the sum of the intensities of satellite peaks to the sum of the intensities of the satellite and center peaks using the 2-bond <sup>15</sup>N HSQC without <sup>13</sup>C decoupling during acquisition (Dayie and Thakur  ##REF## ) and (ii) the ratio of the sum of the intensities of satellite peaks to the sum of the intensities of the satellite and center peaks using the 1D <sup>1</sup>H experiment without <sup>13</sup>C decoupling during acquisition as described in the text</p><p>\n<sup>b,c</sup>The percentage label (Plabel) is calculated as in (a) but this time with only method (ii)</p></table-wrap-foot>"
-    ],
-    "graphic": [
-        "<graphic xlink:href=\"10858_2011_9581_Fig1_HTML\" id=\"MO1\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Fig2_HTML\" id=\"MO2\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Fig3_HTML\" id=\"MO3\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Fig4_HTML\" id=\"MO4\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Fig5_HTML\" id=\"MO5\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Fig6_HTML\" id=\"MO6\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Article_Equ1.gif\" position=\"anchor\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Article_Equ2.gif\" position=\"anchor\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Article_Equ3.gif\" position=\"anchor\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Article_Equ4.gif\" position=\"anchor\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Article_Equ5.gif\" position=\"anchor\"/>",
-        "<inline-graphic xlink:href=\"10858_2011_9581_Article_IEq1.gif\"/>",
-        "<inline-graphic xlink:href=\"10858_2011_9581_Article_IEq2.gif\"/>",
-        "<graphic xlink:href=\"10858_2011_9581_Fig7_HTML\" id=\"MO12\"/>"
-    ],
-    "media": [
-        "<media xlink:href=\"10858_2011_9581_MOESM1_ESM.pdf\"><caption><p>Supplementary material 1 (PDF 858 kb)</p></caption></media>"
-    ],
-    "unknown_pub": "[{\"surname\": [\"Bax\", \"Ikura\", \"Kay\", \"Torchia\", \"Tschudin\"], \"given-names\": [\"A\", \"M\", \"LE\", \"DA\", \"R\"], \"article-title\": [\"Comparison of different modes of two-dimensional reverse correlation NMR for the study of proteins\"], \"source\": [\"J Magn Reson\"], \"year\": [\"1990\"], \"volume\": [\"86\"], \"fpage\": [\"304\"], \"lpage\": [\"318\"], \"pub-id\": [\"10.1016/0022-2364(90)90262-8\"]}, {\"surname\": [\"Bodenhausen\", \"Ruben\"], \"given-names\": [\"G\", \"DJ\"], \"article-title\": [\"Natural abundance nitrogen-\"], \"sup\": [\"15\"], \"source\": [\"Chem Phys Lett\"], \"year\": [\"1980\"], \"volume\": [\"69\"], \"fpage\": [\"185\"], \"lpage\": [\"189\"], \"pub-id\": [\"10.1016/0009-2614(80)80041-8\"]}, {\"mixed-citation\": [\"Clowes R, Hayes W (1968) Experiments in microbial genetics. Wiley, New York\"]}, {\"surname\": [\"Dayie\", \"Wagner\"], \"given-names\": [\"KT\", \"G\"], \"article-title\": [\"Carbonyl carbon probe of local mobility in \"], \"sup\": [\"13\", \"15\"], \"source\": [\"J Am Chem Soc\"], \"year\": [\"1997\"], \"volume\": [\"119\"], \"fpage\": [\"7797\"], \"lpage\": [\"7806\"], \"pub-id\": [\"10.1021/ja9633880\"]}, {\"surname\": [\"Kay\", \"Keifer\", \"Saarinen\"], \"given-names\": [\"LE\", \"P\", \"T\"], \"article-title\": [\"Pure absorption gradient enhanced heteronuclear single quantum correlation spectroscopy with improved sensitivity\"], \"source\": [\"J Am Chem Soc\"], \"year\": [\"1992\"], \"volume\": [\"114\"], \"fpage\": [\"10663\"], \"lpage\": [\"10665\"], \"pub-id\": [\"10.1021/ja00052a088\"]}, {\"surname\": [\"LeMaster\", \"Kushlan\"], \"given-names\": [\"DM\", \"DM\"], \"article-title\": [\"Dynamical mapping of \"], \"italic\": [\"E\", \"coli\"], \"sup\": [\"13\"], \"source\": [\"J Am Chem Soc\"], \"year\": [\"1996\"], \"volume\": [\"118\"], \"fpage\": [\"9255\"], \"lpage\": [\"9264\"], \"pub-id\": [\"10.1021/ja960877r\"]}, {\"surname\": [\"Meissner\", \"Schulte-Herbrueggen\", \"Briand\", \"Sorensen\"], \"given-names\": [\"A\", \"T\", \"J\", \"OW\"], \"article-title\": [\"Double spin-state-selective coherence transfer. Application for two-dimensional selection of multiplet components with long transverse relaxation times\"], \"source\": [\"Mol Phys\"], \"year\": [\"1998\"], \"volume\": [\"96\"], \"fpage\": [\"1137\"], \"lpage\": [\"1142\"]}, {\"surname\": [\"Milecki\"], \"given-names\": [\"J\"], \"article-title\": [\"Specific labelling of nucleosides and nucleotides with \"], \"sup\": [\"13\", \"15\"], \"source\": [\"J Label Compd Radiopharm\"], \"year\": [\"2002\"], \"volume\": [\"45\"], \"fpage\": [\"307\"], \"lpage\": [\"337\"], \"pub-id\": [\"10.1002/jlcr.553\"]}, {\"surname\": [\"Palmer\", \"Cavanagh\", \"Wright\", \"Rance\"], \"given-names\": [\"AG\", \"J\", \"PE\", \"M\"], \"article-title\": [\"Sensitivity improvement in proton -detected two-dimensional heteronuclear correlation NMR spectroscopy\"], \"source\": [\"J Magn Reson\"], \"year\": [\"1991\"], \"volume\": [\"93\"], \"fpage\": [\"151\"], \"lpage\": [\"170\"], \"pub-id\": [\"10.1016/0022-2364(91)90036-S\"]}, {\"surname\": [\"Sambrook\", \"Russell\"], \"given-names\": [\"J\", \"D\"], \"source\": [\"Molecular cloning: a laboratory manual\"], \"year\": [\"2001\"], \"publisher-loc\": [\"Cold Spring Harbor\"], \"publisher-name\": [\"Cold Spring Harbor Laboratory\"]}, {\"surname\": [\"Sattler\", \"Schwendinger\", \"Schleucher\", \"Griesinger\"], \"given-names\": [\"M\", \"MG\", \"J\", \"C\"], \"article-title\": [\"Novel strategies for sensitivity enhancement in heteronuclear multi-dimensional NMR experiments employing pulsed field gradients\"], \"source\": [\"J Biomol NMR\"], \"year\": [\"1995\"], \"volume\": [\"5\"], \"fpage\": [\"11\"], \"lpage\": [\"22\"], \"pub-id\": [\"10.1007/BF00417487\"]}, {\"surname\": [\"Schilling\", \"Edwards\", \"Letscher\", \"Palsson\"], \"given-names\": [\"CH\", \"JS\", \"D\", \"B\\u00d8\"], \"article-title\": [\"Combining pathway analysis with flux balance analysis for the comprehensive study of metabolic systems\"], \"source\": [\"Biotechnol Bioeng\"], \"year\": [\"2001\"], \"volume\": [\"71\"], \"fpage\": [\"286\"], \"lpage\": [\"306\"], \"pub-id\": [\"10.1002/1097-0290(2000)71:4<286::AID-BIT1018>3.0.CO;2-R\"]}, {\"mixed-citation\": [\"Werbelow LG (1994) Relaxation-induced transfer of nuclear spin polarization as a probe of molecular structure and dynamics in mobile phases. In: Tycko R (ed) Understanding chemical reactivity: nuclear magnetic resonance probes of molecular dynamics. Kluwer Academic Publishers, Boston, pp 223\\u2013263\"]}, {\"surname\": [\"Wijmenga\", \"van Buuren\"], \"given-names\": [\"S\", \"B\"], \"article-title\": [\"The use of NMR methods for conformational studies of nucleic acids\"], \"source\": [\"Prog Nucl Magn Reson Spectrosc\"], \"year\": [\"1998\"], \"volume\": [\"32\"], \"fpage\": [\"287\"], \"lpage\": [\"387\"], \"pub-id\": [\"10.1016/S0079-6565(97)00023-X\"]}, {\"surname\": [\"Yamazaki\", \"Muhandiram\", \"Kay\"], \"given-names\": [\"T\", \"R\", \"LE\"], \"article-title\": [\"NMR experiments for the measurement of carbon relaxation properties in highly enriched, uniformly \"], \"sup\": [\"13\", \"15\", \"13\", \"\\u03b1\"], \"source\": [\"J Am Chem Soc\"], \"year\": [\"1994\"], \"volume\": [\"114\"], \"fpage\": [\"8266\"], \"lpage\": [\"8278\"], \"pub-id\": [\"10.1021/ja00097a037\"]}]",
-    "glossary": {
-        "acronym": [
-            "AMP",
-            "CMP",
-            "UMP",
-            "GMP",
-            "R5P",
-            "FBP",
-            "F6P",
-            "GA3P",
-            "G6PDH",
-            "K10zwf",
-            "Gly",
-            "Ser",
-            "noPPP",
-            "OAA",
-            "DHAP",
-            "oPPP",
-            "rNTPs",
-            "TIM",
-            "PEP",
-            "G6P",
-            "Ru5P",
-            "X5P",
-            "S7P",
-            "E4P",
-            "3PG",
-            "6PG",
-            "6PGA"
-        ],
-        "definition": [
-            "Adenosine 5\u2032-monophosphate",
-            "Cytidine 5\u2032-monophosphate",
-            "Uridine 5\u2032-monophosphate",
-            "Guanosine 5\u2032-monophosphate",
-            "Ribose-5-phosphate",
-            "Fructose-6-bisphosphate",
-            "Fructose-6-phosphate",
-            "Glyceraldehyde-3-phosphate",
-            "Glucose-6-phosphate dehydrogenase",
-            "Glucose-6-phosphate dehydrogenase mutant",
-            "Glycine",
-            "Serine",
-            "Non-oxidative pentose phosphate pathway",
-            "Oxaloacetate",
-            "Dihydroxyacetone phosphate",
-            "Oxidative pentose phosphate pathway",
-            "Ribonucleoside triphosphates",
-            "Triosephosphate isomerase",
-            "Phosphoenolpyruvate",
-            "Glucose-6-phosphate",
-            "Ribulose-5-phosphate",
-            "Xylulose-5-phosphate",
-            "Sedoheptulose-7-phosphate",
-            "Erythrose-4-phosphate",
-            "3-Phosphoglycerate",
-            "6-Phosphogluconate",
-            "6-Phosphogluconate-\u03b4-lactone"
-        ]
-    },
-    "references": {
-        "introduction": {...},
-        "methods": {...},
-        "results": {
-            "pmid_ref": ["8226637", "19049467", "20730533", "21057854", "20730533", "21057854", "8065453", "8910275", "16433544", "17116475", "19838167", "20309608", "20730533", "21057854", "16034664", "15327312", "8019138", "12913409", "18047338", "17098254"],
-            "unknown_pub_ref": [11, 10, 14, 3, 3, 12],
-            "figure_ref": [0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 0, 0, 0, 2, 2, 2, 1, 2, 2, 3, 4, 3, 3, 4, 3, 3, 4, 5, 5, 6, 6, 6, 6, 6, 6],
-            "table_ref": [0, 0, 0, 0, 0],
-            "formula_ref": [0, 1, 2, 3, 4, 5, 6, 2, 4, 0, 2, 4],
-            "box_ref": [],
-            "code_ref": [],
-            "quote_ref": [],
-            "chem_ref": [],
-            "supplementary_ref": [],
-            "footnote_ref": [],
-            "graphic_ref": [],
-            "media_ref": []
-        },
-        "discussion": {...},
-        "conclusion": {...},
-        "front": {...},
-        "body": {...},
-        "back": {...},
-        "figure": {...},
-        "table": {...},
-        "formula": {...},
-        "box": {...},
-        "code": {...},
-        "quote": {...},
-        "chem": {...},
-        "supplementary": {...},
-        "footnote": {...}
-    },
-    "references_text": {
-        "introduction": {...},
-        "methods": {...},
-        "results": {
-            "pmid_ref": ["1993", "2008", "2010a", "b", "2010a", "b", "1994", "1996", "2006", "2006", "2009", "2010", "2010a", "b", "2005", "2004", "1994", "2003", "2007", "2007"],
-            "unknown_pub_ref": ["2001", "1995", "1994", "1997", "1997", "1994"],
-            "figure_ref": [ "1", "1", "1", "", "1", "2", "", "2", "1", "2", "1", "1", "1", "3", "3", "", "2", "3", "3", "4", "5", "4", "", "", "4", "4", "5", "6", "", "7", "7", "7", "7", "7", ""],
-            "table_ref": ["1", "", "1", "1", "1"],
-            "formula_ref": ["", "", "", "", "", "", "", "3", "5", "1", "3", "5"],
-            "box_ref": [],
-            "code_ref": [],
-            "quote_ref": [],
-            "chem_ref": [],
-            "supplementary_ref": [],
-            "footnote_ref": [],
-            "graphic_ref": [],
-            "media_ref": []
-        },
-        "discussion": {...},
-        "conclusion": {...},
-        "front": {...},
-        "body": {...},
-        "back": {...},
-        "figure": {...},
-        "table": {...},
-        "formula": {...},
-        "box": {...},
-        "code": {...},
-        "quote": {...},
-        "chem": {...},
-        "supplementary": {...},
-    "n_references": 58,
-    "license": "CC BY-NC",
-    "retracted": "no",
-    "last_updated": "2022-01-14 22:48:41",
-    "citation": "J Biomol NMR. 2011 Oct 30; 51(4):505-517",
-    "package_file": "oa_package/a8/33/PMC3222825.tar.gz"
-}
-```
-
 ### Data Fields
 
 - "accession_id": The PMC ID of the article
@@ -394,7 +132,7 @@ a corpus of pre-annotated text for other tasks (e.g. figure caption to graphic,
 - "retracted": If the article was retracted or not
 - "last_updated": Last update of the article
 - "citation": Citation of the article
-- "package_file": path to the folder containing the graphics and media files of the article
+- "package_file": path to the folder containing the graphics and media files of the article (to append to the base URL: ftp.ncbi.nlm.nih.gov/pub/pmc/)
 
 ### Data Splits