PMC:4845325 / 34631-57521 JSONTXT

{"target":"https://pubannotation.org/docs/sourcedb/PMC/sourceid/4845325","sourcedb":"PMC","sourceid":"4845325","source_url":"https://www.ncbi.nlm.nih.gov/pmc/4845325","text":"Methods\n\nNomenclature\nThis is a proteomics study. To highlight this fact, we labeled hits in the first analyses (Figs. 1b and 2a) with protein names. Due to space restrictions in figure design, we labeled hits in all other figures with the HUPO gene names of the corresponding protein or mRNA products.\n\nHuman plasma samples\nAll participants underwent thorough and standardized history and physical exams. For the antibody arrays, we used a total of 99 archived human plasma samples with ethylenediaminetetraacetate (EDTA) as anticoagulant collected at the University of California San Francisco (San Francisco, CA) and the Mayo Clinic (Rochester, MN and Jacksonville, FL; Additional file 1: Table S1). Plasma was produced by standard blood processing, then aliquots were frozen and stored in aliquots at −80 °C, avoiding freeze thaw cycles. Informed consent was obtained from human subjects according to the ethics committee guidelines at the respective clinical centers. All patients were clinically diagnosed with AD based on the 1984 NINCDS-ADRDA Alzheimer’s criteria with additional attention to the 2011 revisions [61] and (if possible) post-mortem tissue analysis (27 of the 47 AD cases). Details on the svPPA plasma samples are provided in Additional file 1: Table S2. A total of 92 svPPA patients from the University of California San Francisco or Mayo Clinic Jacksonville were identified whose clinical features conformed to revised consensus diagnostic criteria for svPPA [62]. Patient consent had been administered at the respective sample collection centers and research has been conducted according to the principles expressed in the Declaration of Helsinki. Analysis of de-identified samples was performed with research approval by the Stanford University institutional review board.\n\nAntibody-microarray production\nPlasma protein levels were measured using antibody-based protein microarrays. We used a custom-expanded, commercially available microarray with modified antibody content (custom L-Series, RayBiotech Inc., Norcross, GA) containing 474 antibodies against chemokines and cytokines printed in triplicates by the company, plus 17 control antibodies. Additionally, we produced a custom-made in-house array that contained a separate set of 119 antibodies against secreted signaling factors printed in quadruplicates, plus 7 control antibodies. A total of 617 antibodies were measured. Subsequent quality control steps removed 11 antibodies with extremely low or no signal and control antibodies (see Antibody-Microarray Data Preparation below) yielding a total of 582 analyzed antibodies (Additional file 2 and Additional file 1: Figure S1 and S2C). Some antibodies target the same protein multiple times (such as precursor/full-length/truncated forms, 14 proteins and 32 antibodies total; see Additional file 2). The microarray production protocol was the following: antibodies of interest were selected based on their biological role as secreted signaling factors and the availability of ELISA-grade quality batches to ensure likely detection of the epitope in liquid solution. The arrays were printed onto SuperEpoxy glass slides (Arrayit, Sunnyvale, CA) using a custom-built robotic microarrayer fitted with sixteen SMP4B pins (Arrayit). After drying the slides overnight they were vacuum-sealed and stored at −20 °C until use.\n\nPlasma sample preparation and antibody-microarray incubation\nThe human plasma samples were thawed at room temperature and diluted 5-times in PBS without Ca2+/Mg2+ (pH 6.5) followed by 10,000 g centrifugation in a swing bucket centrifuge for 10 min at Room temperature. The lipid layer on top was carefully removed with the house vacuum. Without disturbing the platelet pellet 300 μl was carefully removed for dialysis (96 well Dispodialyzer/5 kDa, Harvard Apparatus, Holliston, MA) into PBS (pH 6.5) at 4 °C in multiple steps including a last over-night step to yield a maximally pure plasma protein fraction in an appropriate buffer for the biotinylation reaction. The dialyzed plasma was diluted again 6-times in PBS and recombinant Green fluorescent protein (GFP) was spiked into the samples as positive control at a final concentration of 1 μg/ml. The plasma proteins were N-terminally biotinylated (NHS-SulfoBiotin, Thermo Scientific, Rockford, IL), reaction was stopped with 0.1 M glycin final concentration and unbound biotin removed by multiple dialysis against PBS (pH 6.5 and last at pH 8). Then samples were diluted in 3 % casein in PBS (pH 7.4) and the individual samples were incubated on blocked antibody arrays over-night at 4 °C. Blocking was performed by incubating dried arrays in 4 °C precooled 3 % casein in PBS (pH 7.4) overnight on a shaker (30 rpm) at 4 °C. After multiple washing steps antibody-bound protein was detected using 0.5 μg/ml Alexa Fluor 555 conjugated streptavidin (Invitrogen) on a GenePix Pro 4000B scanner (Molecular Devices, Sunnyvale, CA, Additional file 1: Figure S1D). Samples for both AD and svPPA studies were processed in parallel in randomized order in one batch.\n\nData processing and figure generation\nRaw data from the array scanner were provided as images (.tif files) and spot intensities (Excel.xls files; Microsoft, Seattle, WA). Excel files were condensed into one file (tab-delimited.txt file) and non-analyzed data rows/columns were removed using RDBmerge (Ron de Bruin, www.rondebruin.nl). Unless otherwise stated, data processing and statistical testing were performed in Matlab R2012a (MathWorks, Natick, MT). Figures were generated directly in Matlab or data were transferred and plotted in Prism 5.0f (GraphPad Software, La Jolla, CA). Figures were then arranged for publishing using Illustrator CS5 and Photoshop CS5 (both Adobe, San Diego, CA).\n\nAntibody-microarray data preparation\nTo determine spot intensities, we calculated the mean pixel intensity per spot. To determine background intensities we calculated the median pixel intensity per background “doughnut” (Additional file 1: Figure S1D). Individual array spots were background subtracted locally (by subtracting the median background across spot replicates in each sample). Spots with a residual intensity less than 10 % above background were set to ‘ND’ (non-detectable). Antibodies with more than 55 % ‘ND’ values were excluded from the analysis (N = 11), yielding a total of 582 quantifiable antibodies (Additional file 2, for ‘ND’-count distribution see Additional file 1: Figure S2C). ‘ND’ values were then replaced with the greater of the half the minimum non-‘ND’ value per sample replicates, the half the minimum non-’ND’ value of that antibody across all samples (if the sample replicates were all ‘ND’), or 1. The spot data were Log2 transformed, replicate averaged, and iteratively (i = 50) row- and column-wise median centered (subtract the column-wise median from the values in each column/row of data, so that the mean or median value of each column/row is 0) and normalized (multiply all values in each column/row of data by a scale factor S so that the sum of the squares of the values in each column/row is constant across columns/rows) following a procedure described in the Cluster 3.0 manual [63]. Finally the data were Z-scored, leaving approximately normally distributed data for analysis with a mean of 0 and a standard deviation of 1 (Additional file 1: Figure S2D to F, 86 % of all antibodies have normal distributions based on one-sample Kolmogorov-Smirnov test).\n\nPrinciple component analysis\nTo assess the influence of potentially confounding factors such as plasma source, patient age, or patient gender, we performed a Principle Component Analysis in Matlab using the Z-scored data and the built-in princomp function.\n\nDifferential protein level analysis\nTo identify proteins with significant changes in plasma levels (based on Z-score values) we calculated permutation-corrected p-values (pcorr) for Control vs. AD for every protein (unpaired two-tailed t-test, 10,000 class label permutations) using the mattest Matlab function. To compute false-discovery rates, we adopted a direct approach to estimate q-values [64] using the mafdr Matlab function. Proteins with a significant difference between AD and Control samples and a q \u003c 0.05 were considered having different plasma levels (a total of 50 proteins). An identical approach was used for the svPPA data.\n\nNetwork representation\nTo link the proteins with changed plasma levels to biological pathways, we mapped these proteins onto known protein networks using the Genemania-app [65] in Cytoscape 3.0.1 [www.cytoscape.org, [66]]. Pathway data came from NCI-Nature [67], Reactome [68–70], and [71]. Physical interaction data came from Biogrid Small Scale [72], IREF Interact, and IREF Small Scale [73]. We allowed for some nodes above the significance threshold to be added by the algorithm to connect cliques (dashed nodes, p-value indicted in figure). To test for enrichment in biological function, we queried Gene Ontology [74, 75], KEGG [76], and Panther [77] databases using DAVID [13] with the 564 unique genes representing the 582 proteins tested as background.\n\nMini-mental State Exam (MMSE) correlation\nMMSE scores were recorded at the time of plasma acquisition at the respective clinical centers. Scores were available for 44/47 AD patients and 26/52 Control patients (Additional file 2 and Additional file 1: Table S1). MMSE scores were correlated to Z-scored protein levels using Spearman’s rank correlation. Correlation significance was assessed using Spearman’s p-value (pRho: significance that slope is not 0) and by computing an empirical p-value by permuting protein scores over MMSE scores 1,000-times (pPerm: Number of times that random MMSE-Protein data yields correlation greater than observed/1,000). Network representation was performed as described above.\n\nDifferential co-expression analysis\nCo-expression analysis was performed to identify proteins involved in AD pathogenesis that would inform us on more specific pathways than the broad ones implicated in the differential analysis above. Besides the mere difference in expression levels, proteins may differ in how they correlate with each other between disease and controls. We sought to discover protein networks that are changed between AD and Control by evaluating differential correlation matrices in an approach analogous to methods developed for analyzing genetic interaction profiles [78]. We thus created separate correlation matrices (Spearman’s rank correlation) in Matlab of all of the proteins measured for AD and Control signaling proteomes, respectively. Since cohort sizes were almost equal for AD and Controls we then calculated differential correlation profiles from these correlation matrices and used unsupervised clustering to identify 8 distinct clusters of proteins with highly similar differential correlation profiles (Fig. 4c, boxes). To demonstrate that differential co-expression data contains valuable biological information, we created semantic similarity scores for each protein pair [79]. Protein pairs with high differential co-expression profile correlation exhibited high semantic-similarity profile correlations as well (Additional file 1: Figure S5).\n\nBraak staging and atrophy correlation\nBraak staging and atrophy data were downloaded from the supplemental data file of [22]. We then filtered the data for mRNAs that had been reported to exhibit significant correlation to either Braak staging or atrophy data in the pre-frontal cortex. Expected values were calculated by determining the total number of proteins tested that were reported to have significant correlations. Observed values were calculated by determining the total number of proteins within our MMSE correlation hit list that were reported to have significant correlations. Significance testing was performed using the chi2-test.\n\nSingle Nucleotide Polymorphism (SNP) analysis\nDatasets and SNP association testing. Summarized information from tests of genetic association of AD with SNPs located in the candidate gene regions was culled from a recent large genome-wide association study (GWAS) conducted by the Alzheimer Disease Genetics Consortium (ADGC) [23]. Naj et al. computed results for SNPs throughout the genome in their discovery sample composed of 8,309 AD cases and 7,366 cognitively normal elders from ten independent Caucasian data sets. Details of the procedures for quality control, genotype imputation, and population substructure adjustment are published elsewhere [23]. Genotyped and imputed SNPs were tested for association with AD in each dataset separately using a logistic generalized linear model (GLM) in case–control datasets and a logistic generalized estimating equation (GEE) in family-based datasets, controlling for intra-study population substructure. Genotyped SNPs were coded as 0, 1, or 2 according to the number of minor alleles under the additive genetic model. For imputed SNPs, a quantitative estimate between 0 and 2 for the dose of the minor allele were used to incorporate the uncertainty of the imputation estimates. All analyses were performed using the GEE [80] and GWAF [81] programs in the R statistical software package. SNP association results obtained from individual datasets were combined by meta-analysis using the inverse variance method implemented in the software package METAL [82] (http://www.sph.umich.edu/csg/abecasis/Metal/index.html).\n\nGene-based multiple testing corrections\nWe corrected for testing multiple SNPs in a gene after accounting for correlation between SNP genotypes due to linkage disequilibrium. Each gene tested was treated as an independent hypothesis and the effective number of tests per gene was obtained by a previously described method [83]. The Versatile Gene-based Association Study (VEGAS) approach [84] was used to summarize the strength of association of a gene with AD based on the number of SNPs tested in the gene and size of the gene. This method computes a gene-based test statistic based on the SNP p-values within the gene, and then uses simulation to calculate an empirical gene-based p-value. The distribution of empirical p-values was then plotted and tested against an expected distribution of p-values using the Kolmogorov-Smirnov test.\n\nmRNA expression analysis\nThe dataset comprises gene expression data from brain tissues that were posthumously collected from more than 600 individuals with AD diagnosis, HD diagnosis, or with normal non-demented brains. We used a subset of dorsolateral prefrontal cortex (PFC, Brodmann area 9) samples from 181 AD case and 125 controls. Only neuropathologically confirmed AD subjects with Braak stage \u003e III were included in this profiling experiment; Braak stage and atrophy were assessed by pathologists at McLean Hospital (Belmont, MA). The samples were flash frozen in liquid nitrogen vapor with an average postmortem interval (PMI) of about 18 h.\nA total of 1 μg of mRNA extracted from each tissue sample was amplified to fluorescently labeled cRNA, and profiled by the Rosetta Gene Expression Laboratory in two phases using the Rosetta/Merck 44 k 1.1 microarray (GPL4372) (Agilent Technologies, Santa Clara, CA). The average RNA integrity number of 6.81 was sufficiently high for the microarray experiment monitoring 40,638 transcripts representing more than 31,000 unique genes. The expression levels were processed and normalized to the average of all samples in the batch from the same region using Rosetta Resolver (Rosetta Biosoftware, Seattle, WA).\nAll microarray data generated in this study are available through the National Brain Databank at the Harvard Brain Tissue Resource Center (http://www.brainbank.mclean.org/). This microarray dataset is MIAME compliant. The raw and final processed data for each hybridization are available upon request. The essential sample annotation including experimental factors and their values (e.g., gender, age, PMI, pH) is available and summarized in [24].\nThe differential gene expression was assessed using the standard t-test. The distribution of p-values was then plotted and tested against an expected distribution of p-values using the Kolmogorov-Smirnov test.\n\nWestern blot\nActive GDF3 levels were determined in fresh tissue samples not part of the plasma screen (Additional file 1: Table S3). Hippocampal samples were a random subset picked blindly from the same donors as the cortical samples. All tissues or cells were lysed in RIPA buffer and total protein concentrations were determined with a BCA Protein Assay Kit (Thermo Scientific, Waltham, MA). 10–20 μg of total protein was loaded for each sample into pre-cast 4–12 % bis-tris gels and run with MOPS buffer (Invitrogen, Carlsbad, CA). Gels were transferred onto PVDF membranes (Millipore, Billerica, MA). Antigen specific primary antibodies were incubated overnight at 4 °C and detected with species-specific horseradish-peroxidase labeled secondary antibodies. An ECL Western Blotting Detection kit (GE Healthcare, Cleveland, OH) was used to obtain a chemiluminescence signal, which was detected using Amersham Hyperfilm ECL (GE Healthcare). Band quantification was performed using ImageJ software (version 1.46; NIH, Bethesda, MD). Bands of interest were normalized to actin or neuron specific enolase for a loading control. For active GDF3 we used anti-GDF3 antibodies from Novus Biologicals (Littleton, CO; NBP1-96508).\n\nCell culture assays\nHuman NTERA cells expressing eGFP under the DCX promoter were maintained in DMEM media containing 10 % FBS. To induce differentiation, cells were plated in 96 well plates. One day after seeding, 10 μM of retinoic acid and designated concentrations of recombinant carrier-free human GDF3 (R\u0026D Systems, Minneapolis, MN; at 0, 10, 50, or 150 ng/mL) were added to the culture media of each corresponding NTERA treatment well. Cells were maintained under these conditions for 2 weeks, during which media was replaced every 3 days. Cells were then cultured for an additional 2 weeks with continued GDF3 treatment, in the absence of retinoic acid.\n\nCellavista\nAdult neurosphere number, eGFP expression (relative to cell confluence), and number of proliferating NPCs were quantified after GDF3 treatment using an Innovatis Cellavista Imager (Dynamic Devices, Wilmington, DE). To quantify NPC proliferation, 10x images were collected by Cellavista and BrdU+ nuclei were detected and quantified by Cellavista software using the cell nuclei count function.\n\nAdult hippocampal NPC isolation\nHippocampal NPCs were isolated from 6 week old male and female mice [85]. NPCs were maintained on poly-D-lysine (Sigma, St. Lous, MO) and laminin (Invitrogen) coated 10 cm plastic plates in neurobasal A media (Invitrogen) with 1× B27 supplement without vitamin A (Invitrogen) and 1× GlutaMAX-I supplement (Invitrogen) and 20 ng/ml each of recombinant human FGF-basic (Peprotech, Rocky Hill, NJ) and recombinant human EGF (Peprotech) at 37 °C and 5 % CO2. All experiments used NPCs below passage 20 and were repeated at least once with male NPCs and once with female NPCs.\n\nGDF3 treatment of proliferating NPCs\n5000 cells were plated per well in a 96 well poly-D-lysine/laminin-coated plate with 0, 0.01, 0.1, 1, 10 or 100 ng/ml recombinant mouse GDF3 (R\u0026D Systems) in normal growth media. Cells were allowed to grow for 4 days, with a ½ media change on day 2 (which replaced full growth factors [85] and ½ of GDF3 treatment). After 4 days in treatment, 20 μM bromodeoxyuridine (BrdU, Sigma) in sterile PBS was added to all wells and cells were fixed with 4 % paraformaldehyde 2 h later for 10 min.\n\nImmunocytochemistry\nFixed cells were rinsed with 0.1 M phosphate buffered saline (PBS) 3 times then blocked with 10 % normal donkey serum (NDS, Jackson ImmunoResearch, West Grove, PA) and 0.3 % Triton-X 100 (Sigma) in PBS for 30 min. Cells were incubated overnight in primary antibody, rat anti-BrdU (1:500, AbD Serotech, Raleigh NC) in 10 % NDS in PBS at 4 °C. Cells were then rinsed and incubated in secondary antibody, Alexa488 anti-rat (1:200, Invitrogen) in 10 % NDS in PBS. After rinsing, total BrdU+ cells were imaged and quantified using an automated Cellavista microscope system (Hoffman-La Roche, Basel, Switzerland).\n\nNPC differentiation\nMurine NPCs were differentiated for 8 days [85]. Briefly, cells were plated at 200,000 cells/well in a poly-D-lysine/laminin-coated plate in either full growth factor media (20 ng/ml of EGF and FGF2; proliferative conditions) or in media with only 5 ng/ml FGF2. After 2 days, the proliferative wells and the 2d differentiation wells were harvested while the 4d and 8d wells received a complete media change to media with no growth factors added. At day 4, the 4d wells were harvested and the 8d wells received a ½ media change with no growth factors added. A ½ media change was repeated on day 6 and the remaining wells were harvested on day 8.\n\nRNA harvesting and conversion to cDNA\nCells were removed from the plate using Accutase Cell Dissociation Reagent (Invitrogen) then centrifuged at 400 g for 5 min. The cell pellet was stored at −80 °C until later RNA extraction. RNA extraction was performed using the RNeasy Mini Kit (Qiagen, Venlo, Netherlands) as per manufacturer instructions. The resulting RNA was quantified using a nanodrop spectrophotometer and RNA purity was confirmed using A260/A280 ratios. 500 ng of RNA was treated with DNase I as per manufacturer instructions (Invitrogen) to eliminate any genomic DNA contamination and then converted to cDNA using SuperScript III first-strand synthesis system (Invitrogen) as per manufacturer instructions. cDNA was diluted 1:5 in water.\n\nReal-time quantitative PCR\n2 μl of cDNA was quantified in duplicate for each sample using LightCycler 480 SYBR Green I (Roche) on a LightCycler 480 II as per manufacturer instructions. Cycling conditions were: 15 min at 95 °C, 45 cycles of [15 s at 94 °C, 25 s at 58 °C, 20 s at 72 °C]. Melt curve cycles immediately followed and were: 5 s at 95 °C, 1 min at 65 °C and then gradual temperature rise to 97 °C at a rate of 0.11 °C/s followed by 30s at 40 °C. GDF3 levels were normalized to MAPK3 [86] as a reference gene because MAPK3 has been shown not to change with differentiation in contrast to many other standard housekeeping genes such as actin, which change dramatically during the differentiation process [87]. Melt curve analysis was performed to verify primer specificity and all primers were tested in a dilution series before use. Data is displayed as fold change above proliferative condition mRNA levels using 2^(ΔΔCt) values.\nPrimer sequences were obtained from the MIT/Harvard PrimerBank.\nGDF3 fwd: 5’ ATGCAGCCTTATCAACGGCTT\nGDF3 rev: 5’ AGGCGCTTTCTCTAATCCCAG\nGDF3 PrimerBankID: 6679979a1\nMAPK3 fwd: 5’ TCCGCCATGAGAATGTTATAGGC\nMAPK3 rev: 5’ GGTGGTGTTGATAAGCAGATTGG\nMAPK3 PrimerBankID: 21489933a1","divisions":[{"label":"title","span":{"begin":0,"end":7}},{"label":"sec","span":{"begin":9,"end":302}},{"label":"title","span":{"begin":9,"end":21}},{"label":"p","span":{"begin":22,"end":302}},{"label":"sec","span":{"begin":304,"end":1798}},{"label":"title","span":{"begin":304,"end":324}},{"label":"p","span":{"begin":325,"end":1798}},{"label":"sec","span":{"begin":1800,"end":3355}},{"label":"title","span":{"begin":1800,"end":1830}},{"label":"p","span":{"begin":1831,"end":3355}},{"label":"sec","span":{"begin":3357,"end":5068}},{"label":"title","span":{"begin":3357,"end":3417}},{"label":"p","span":{"begin":3418,"end":5068}},{"label":"sec","span":{"begin":5070,"end":5765}},{"label":"title","span":{"begin":5070,"end":5107}},{"label":"p","span":{"begin":5108,"end":5765}},{"label":"sec","span":{"begin":5767,"end":7471}},{"label":"title","span":{"begin":5767,"end":5803}},{"label":"p","span":{"begin":5804,"end":7471}},{"label":"sec","span":{"begin":7473,"end":7729}},{"label":"title","span":{"begin":7473,"end":7501}},{"label":"p","span":{"begin":7502,"end":7729}},{"label":"sec","span":{"begin":7731,"end":8373}},{"label":"title","span":{"begin":7731,"end":7766}},{"label":"p","span":{"begin":7767,"end":8373}},{"label":"sec","span":{"begin":8375,"end":9135}},{"label":"title","span":{"begin":8375,"end":8397}},{"label":"p","span":{"begin":8398,"end":9135}},{"label":"sec","span":{"begin":9137,"end":9847}},{"label":"title","span":{"begin":9137,"end":9178}},{"label":"p","span":{"begin":9179,"end":9847}},{"label":"sec","span":{"begin":9849,"end":11234}},{"label":"title","span":{"begin":9849,"end":9884}},{"label":"p","span":{"begin":9885,"end":11234}},{"label":"sec","span":{"begin":11236,"end":11880}},{"label":"title","span":{"begin":11236,"end":11273}},{"label":"p","span":{"begin":11274,"end":11880}},{"label":"sec","span":{"begin":11882,"end":13447}},{"label":"title","span":{"begin":11882,"end":11927}},{"label":"p","span":{"begin":11928,"end":13447}},{"label":"sec","span":{"begin":13449,"end":14288}},{"label":"title","span":{"begin":13449,"end":13488}},{"label":"p","span":{"begin":13489,"end":14288}},{"label":"sec","span":{"begin":14290,"end":16207}},{"label":"title","span":{"begin":14290,"end":14314}},{"label":"p","span":{"begin":14315,"end":14940}},{"label":"p","span":{"begin":14941,"end":15549}},{"label":"p","span":{"begin":15550,"end":15997}},{"label":"p","span":{"begin":15998,"end":16207}},{"label":"sec","span":{"begin":16209,"end":17432}},{"label":"title","span":{"begin":16209,"end":16221}},{"label":"p","span":{"begin":16222,"end":17432}},{"label":"sec","span":{"begin":17434,"end":18094}},{"label":"title","span":{"begin":17434,"end":17453}},{"label":"p","span":{"begin":17454,"end":18094}},{"label":"sec","span":{"begin":18096,"end":18499}},{"label":"title","span":{"begin":18096,"end":18106}},{"label":"p","span":{"begin":18107,"end":18499}},{"label":"sec","span":{"begin":18501,"end":19104}},{"label":"title","span":{"begin":18501,"end":18532}},{"label":"p","span":{"begin":18533,"end":19104}},{"label":"sec","span":{"begin":19106,"end":19630}},{"label":"title","span":{"begin":19106,"end":19142}},{"label":"p","span":{"begin":19143,"end":19630}},{"label":"sec","span":{"begin":19632,"end":20259}},{"label":"title","span":{"begin":19632,"end":19651}},{"label":"p","span":{"begin":19652,"end":20259}},{"label":"sec","span":{"begin":20261,"end":20925}},{"label":"title","span":{"begin":20261,"end":20280}},{"label":"p","span":{"begin":20281,"end":20925}},{"label":"sec","span":{"begin":20927,"end":21678}},{"label":"title","span":{"begin":20927,"end":20964}},{"label":"p","span":{"begin":20965,"end":21678}},{"label":"title","span":{"begin":21680,"end":21706}},{"label":"p","span":{"begin":21707,"end":22620}},{"label":"p","span":{"begin":22621,"end":22684}},{"label":"p","span":{"begin":22685,"end":22719}},{"label":"p","span":{"begin":22720,"end":22754}},{"label":"p","span":{"begin":22755,"end":22783}},{"label":"p","span":{"begin":22784,"end":22821}},{"label":"p","span":{"begin":22822,"end":22859}}],"tracks":[{"project":"2_test","denotations":[{"id":"27112350-21325651-72613518","span":{"begin":1484,"end":1486},"obj":"21325651"},{"id":"27112350-20926419-72613519","span":{"begin":8548,"end":8550},"obj":"20926419"},{"id":"27112350-14597658-72613520","span":{"begin":8592,"end":8594},"obj":"14597658"},{"id":"27112350-18832364-72613521","span":{"begin":8633,"end":8635},"obj":"18832364"},{"id":"27112350-21067998-72613522","span":{"begin":8648,"end":8650},"obj":"21067998"},{"id":"27112350-15608231-72613522","span":{"begin":8648,"end":8650},"obj":"15608231"},{"id":"27112350-24243840-72613522","span":{"begin":8648,"end":8650},"obj":"24243840"},{"id":"27112350-20482850-72613523","span":{"begin":8661,"end":8663},"obj":"20482850"},{"id":"27112350-25428363-72613524","span":{"begin":8723,"end":8725},"obj":"25428363"},{"id":"27112350-20940177-72613525","span":{"begin":8765,"end":8767},"obj":"20940177"},{"id":"27112350-10802651-72613526","span":{"begin":8994,"end":8996},"obj":"10802651"},{"id":"27112350-25428369-72613527","span":{"begin":8998,"end":9000},"obj":"25428369"},{"id":"27112350-10592173-72613528","span":{"begin":9009,"end":9011},"obj":"10592173"},{"id":"27112350-23868073-72613529","span":{"begin":9027,"end":9029},"obj":"23868073"},{"id":"27112350-19131956-72613530","span":{"begin":9054,"end":9056},"obj":"19131956"},{"id":"27112350-22252388-72613531","span":{"begin":10440,"end":10442},"obj":"22252388"},{"id":"27112350-20179076-72613532","span":{"begin":11063,"end":11065},"obj":"20179076"},{"id":"27112350-23622250-72613533","span":{"begin":11357,"end":11359},"obj":"23622250"},{"id":"27112350-20040588-72613534","span":{"begin":13168,"end":13170},"obj":"20040588"},{"id":"27112350-20616382-72613535","span":{"begin":13386,"end":13388},"obj":"20616382"},{"id":"27112350-16077740-72613536","span":{"begin":13772,"end":13774},"obj":"16077740"},{"id":"27112350-20598278-72613537","span":{"begin":13838,"end":13840},"obj":"20598278"},{"id":"27112350-22216330-72613538","span":{"begin":15993,"end":15995},"obj":"22216330"},{"id":"27112350-15229246-72613539","span":{"begin":22175,"end":22177},"obj":"15229246"}],"attributes":[{"subj":"27112350-21325651-72613518","pred":"source","obj":"2_test"},{"subj":"27112350-20926419-72613519","pred":"source","obj":"2_test"},{"subj":"27112350-14597658-72613520","pred":"source","obj":"2_test"},{"subj":"27112350-18832364-72613521","pred":"source","obj":"2_test"},{"subj":"27112350-21067998-72613522","pred":"source","obj":"2_test"},{"subj":"27112350-15608231-72613522","pred":"source","obj":"2_test"},{"subj":"27112350-24243840-72613522","pred":"source","obj":"2_test"},{"subj":"27112350-20482850-72613523","pred":"source","obj":"2_test"},{"subj":"27112350-25428363-72613524","pred":"source","obj":"2_test"},{"subj":"27112350-20940177-72613525","pred":"source","obj":"2_test"},{"subj":"27112350-10802651-72613526","pred":"source","obj":"2_test"},{"subj":"27112350-25428369-72613527","pred":"source","obj":"2_test"},{"subj":"27112350-10592173-72613528","pred":"source","obj":"2_test"},{"subj":"27112350-23868073-72613529","pred":"source","obj":"2_test"},{"subj":"27112350-19131956-72613530","pred":"source","obj":"2_test"},{"subj":"27112350-22252388-72613531","pred":"source","obj":"2_test"},{"subj":"27112350-20179076-72613532","pred":"source","obj":"2_test"},{"subj":"27112350-23622250-72613533","pred":"source","obj":"2_test"},{"subj":"27112350-20040588-72613534","pred":"source","obj":"2_test"},{"subj":"27112350-20616382-72613535","pred":"source","obj":"2_test"},{"subj":"27112350-16077740-72613536","pred":"source","obj":"2_test"},{"subj":"27112350-20598278-72613537","pred":"source","obj":"2_test"},{"subj":"27112350-22216330-72613538","pred":"source","obj":"2_test"},{"subj":"27112350-15229246-72613539","pred":"source","obj":"2_test"}]}],"config":{"attribute types":[{"pred":"source","value type":"selection","values":[{"id":"2_test","color":"#ec93a7","default":true}]}]}}