PMC:4331676 / 19078-39076 JSONTXT

{"target":"https://pubannotation.org/docs/sourcedb/PMC/sourceid/4331676","sourcedb":"PMC","sourceid":"4331676","source_url":"https://www.ncbi.nlm.nih.gov/pmc/4331676","text":"Results and discussion\n\nThe selection of LG and features\nTo evaluate the PSSM-DT method, we analyzed the impact of parameter LG on the predictive performance of the proposed model. The predictive results of SVM-PSSM-DT with different values of LG on the benchmark dataset by using five-fold cross validation is shown in Figure 1. As can be seen from the Figure, the value of LG has modest impact on both the ACC and MCC metrics. The ACC firstly increases to a maximum value and then slightly goes down as the LG value increases. So we can conclude that SVM-PSSM-DT achieves the best performance when LG = 5, which mean that the dimension of the feature space applied in this work is 2000. Therefore, the parameter LG was set as 5 for the following experiments.\nFigure 1 The performance of SVM-PSSM-DT with different LG. LG is a parameter in the present method SVM-PSSM-DT. The average ACC and MCC values were used to evaluate the impact of different LG values on the performance of SVM-PSSM-DT. The results were got by testing the model on the benchmark dataset by five-fold-cross-validation. In this study, we proposed three protein representations, including PSSM-DT, PSSM-SDT and PSSM-DDT. Table 1 lists predictive results of the three proposed protein representation according to jackknife validation on benchmark dataset. As a result, the predictor using PSSM-DT yields the highest ACC of 79.96%, MCC of 0.622 and AUC of 86.50%. So in the following experiments, the PSSM-DT based representation was applied to encode the features from PSSM profile.\nTable 1 Results on benchmark dataset of different features through jackknife validation.\nMethods Acc(%) MCC SN(%) SP(%)\nPSSM-DDTa 79.72 0.607 81.33 78.18\nPSSM-SDTb 74.79 0.512 77.147 74.93\nPSSM-DTc 79.96 0.622 81.91 78.00\nPSSM-DT can extract two kinds protein features, called PSSM-DDT and PSSM-SDT respectively. And PSSM-DT represents the combination of PSSM-DDT and PSSM-SDT. The results were got by testing the models on benchmark dataset through jackknife validation.\nathe predictor using PSSM-DDT as protein representation\nbthe predictor using PSSM-SDT as protein representation\ncthe predictor using PSSM-DT as protein representation\n\nFeature analysis\nTo further investigate the importance of the features and reveal the biological meaning of the features in PSSM-DT, we followed the study [50,70,71] to calculate the discriminant weight vector in the feature space. The sequence-specific weight obtained from the SVM training process can be used to calculate the discriminant weight of each feature to measure the importance of the features. Given the weight vectors of the training set with N samples obtained from the kernel-based training A = [a1, a2, a3,...,aN], the feature discriminant weight vector W in the feature space can be calculated by the following equation:\n(12) W = A ⋅ M = a 1 a 2 ⋮ a N T m 11 m 12 ⋯ m 1 j m 21 m 22 ⋯ m 2 j ⋮ ⋮ ⋱ ⋮ m N 1 m N 2 ⋯ m N j\nwhere M is the matrix of sequence representatives in PSSM-DT; A is the weight vectors of the training samples; N is the number of training samples; j is the dimension of the feature vector. The element in W represents the discriminative power of the corresponding feature.\nIn this study, we are only interested in the descriptors frequently occurring in positive samples (DNA-binding proteins). Therefore, the discriminant weight of an amino acid pair is calculated as the quadratic sum of the discriminant weights of the corresponding descriptors with positive discriminant weight for this amino acid pair. The discriminant weights of all the 400 amino acid pairs in PSSM-DT are depicted in Figure 2A. According to this figure, the top four most discriminative amino acid pairs are (R, R), (R, P), (P, R) and (A, R), which indicate that the amino acid R (Arg) and A (Ala) are important for identifying the DNA-protein interaction. This conclusion is consistent with Szilágyi and Skolnick's study [34], in which they found that the percentage of Arg, Ala, Gly, Lys and Asp are useful for identification of DNA-binding proteins. Sieber and Allemann [72] found that R (348) can't directly interact with the nucleobases, but can determine the DNA binding specificity of the basic helix-loop-helix proteins (BHLH) E12 by directly interacting with both the phosphate backbone and the carboxylate of E(345) resulting in locking the side chain conformation of E(345). what's more, by comprehensively analyzing the three dimensional structures of protein-DNA complexes, Rohs and West et al. [73] demonstrated that the binding of R to narrow minor grooves can be applied to mode for protein-DNA recognition, indicating that R is an important component in protein-DNA binding activity. It has been previously reported that the DNA usually enveloped with negative electrostatic potential and the amino acid R shows positive charge [12], which explain the reason why the amino acid R is important for DNA-binding protein identification.\nFigure 2 Feature analysis on protein 1AKHchain A. (A) The discriminant weights of the 400 amino acid pairs. Each element in the figure refers to the quadratic sum of the discriminant weights of descriptors with positive discriminant weight for a certain amino acid pair. A amino acid pair is identified by two amino acids, the x-axis and y-axis represent its second amino acid and first amino acid, respectively. (B) The discriminant weights of the descriptors with different lg values for the top four most discriminant amino acid pairs, including pair(R,R), pair(R,P), pair(P,R) and pair(A,R). (C) The occurrence distributions of the descriptors for the top four most discriminant amino acid pairs on the DNA-binding regions and non DNA-binding regions of protein 1AKH chain A, respectively. The regions in green color are non DNA-binding regions and the region in grey color is DNA-binding protein. (D) The occurrence distributions of the descriptors for the top four most discriminant amino acid pairs on the three dimensional structure of protein 1AKH chain A. The green sections are the three dimensional structure of protein and the brown sections are the three dimensional structure of the DNA. The discriminant weight of the descriptors for pairs (R, R), (R, P), (P, R) and (A, R) with different lg values are shown in Figure 2B. As indicated by the figure, the descriptor with lg of 4 for pair (R, R) has the highest discriminant power. For pair (R, P) and (P, R), the discriminant weight of all descriptors are slightly different. In case of pair (A, R), the descriptor with lg of 5 is the most discriminative feature. In conclusion, for an amino acid pair, the distance between the two amino acids along the sequence can impact its discriminant power in DNA-binding protein identification.\nAdditionally, we take protein 1AKH [PDB:1AKH] chain A as an example to show the availability of PSSM-DT based protein representation on DNA-binding protein identification. 1AKH is known as the MATa1/MATα2 homeodomain heterodimer and its chain A is the yeast mating type transcription factors (MATa1). MATa1 proteins are members of the homeodomain superfamily of DNA-binding proteins and contact the DNA with its homeodomain. It always folds into a compact three-helix domain containing a helix-turn-helix DNA-binding motif. Figure 2C lists the distributions of descriptors for the top four most discriminative pairs on the sequence of MATa1 protein. From this figure we can see that there are 5 occurrences of the proposed descriptors in the DNA-binding region and no occurrence in the non DNA-binding regions. There are totally 5 descriptors occurred in the DNA-binding region, including pair(R, R) with lg of 1, pair(R, R) with lg of 3, pair(P, R) with lg of 2, pair(P, R) with lg of 3 and pair(A, R) with lg of 1. This is further confirmed by the three dimensional structure shown in Figure 2D. As indicated by the figure, there is no descriptor for the four top most discriminative amino acid pairs that occur in the non DNA-binding regions, and all the five occurrences are within the one DNA-binding region. Furthermore, the figure showed that the pair(R, R) with lg of 1and pair(P, R) with lg of 3 are very closed to the three dimensional structure of DNA, indicating that these two descriptors are very discriminative for DNA and protein interaction.\n\nComparison with existing PSSM based encoding schemes\nIn this section, four protein encoding schemes based on PSSM are introduced for a comparison. They are the average score of the residues with respect to the column of certain AA type called AvePscore-20 [21], the average score of the residues of certain AA type with respect to the column of certain AA type called AvePscore-400 [74], the percentile value of the PSSM scores along with the column of certain AA type according to percent thresholds called Pscore-100 [75], and auto-correlation coefficient (ACC) transformation that can transform the PSSMs of different lengths into fixed-length vectors by measuring the correlation between two scores separated by a distance of lg along the sequence [76], respectively. Table 2 lists the predictive results of the proposed protein representation and other four considered protein representations on the benchmark dataset using jackknife validation.\nTable 2 Results on benchmark dataset of different PSSM based encoding schemes through jackknife validation.\nMethods Acc(%) MCC SN(%) SP(%) AUC(%)\nAvePscore-20a 73.95 0.480 68.57 79.09 81.40\nAvePscore-400d 73.58 0.470 66.47 80.36 81.50\nPscore-100c 73.12 0.463 72.76 73.45 80.50\nACCd 73.77 0.475 73.14 74.36 81.90\nPSSM-DTf 79.96 0.622 81.91 78.00 86.50\nThe four protein representation methods in the front of the table are four protein encoding methods for identification of DNA-binding proteins proposed in the past. The four methods and the current method PSSM-DT are based on PSSMs property of protein sequences, but the encoding method applied by them are different. The results were got by testing on benchmark dataset through jackknife validation.\naresults obtained by in-house implementation of AvePscore-20 [21]\nbresults obtained by in-house implementation of AvePscore-400 [21]\ncresults obtained by in-house implementation of Pscore-100 [75]\ndresults obtained by in-house implementation of ACC [76]\nfresults obtained by using PSSM-DT as protein representation Furthermore, to provide a graphic illustration to show the performance of the five protein representations, the corresponding ROC (receiver operating characteristic) curves were drawn in Figure 3, where the horizontal coordinate X is for the false positive rate or 1-SP and the vertical coordinate Y is for the true positive rate or SN. The best method would yield a point with the coordinate (0,1) meaning 0 false positive rate and 100% true positive rate. Therefore a perfect classification method would give a point with the coordinate (0,1) and a completely random guess would give a point along the diagonal from point (0,0) to (1,1). The area under the ROC curve called AUC is often used to indicate the performance quality of binary classification methods, where the larger the area, the better the predictive quality is.\nFigure 3 The ROC curves of several PSSM based protein encoding methods on benchmark dataset. The receiver operating characteristic (ROC) curves of PSSM-DT and several other existing protein encoding methods were got by testing the models on benchmark dataset through jackknife validation, where the horizontal coordinate X is for the false positive rate or 1-SP and the vertical coordinate Y is for the true positive rate or SN and a good method would yield a curve close to the coordinate (0,1) meaning low false positive rate and high true positive rate. As shown in Table 2 and Figure 3, the PSSM-DT based protein representation generated the highest performance and outperformed the other four protein representations based on PSSM, indicating that PSSM-DT based protein representation is effective for DNA-binding protein identification.\n\nComparison with existing prediction methods\nTable 3 shows the predictive results of SVM-PSSM-DT and four other state-of-the-art methods on the benchmark dataset through jackknife validation, including DNAbinder(dimension 21) [21], DNAbinder(dimension 400) [21], DNA-Port [74] and iDNA-Prot [16]. DNAbinder(dimension 21) and DNAbinder(dimension 400) encode features from their PSSM based evolutionary information and utilize SVM to build prediction model. iDNA-Prot applies grey model to integrate the features from protein sequence into the general form of pseudo amino acid composition and then inputs into a Random Forest classifier. DNA-Prot is a Random Forest classifier based on the amino acid composition, predicted second structure and some physicochemical properties. The ROC curves of the proposed method and the four predictive methods are shown in Figure 4.\nTable 3 Results on benchmark dataset of different predictors through jackknife validation.\nmetric ACC (%) MCC SN (%) SP (%) AUC(%)\nDNAbinder(dimension 21)a 73.95 0.480 68.57 79.09 81.40\nDNAbinder(dimension 400)b 73.58 0.470 66.47 80.36 81.50\nDNA-Protc 72.55 0.440 82.67 59.76 78.90\niDNA-Protd 75.40 0.500 83.81 64.73 76.10\nPSSM-DTf 79.96 0.622 81.91 78.00 86.50\nThe four methods in the front of the table are four state-of-the-art predicting methods for identification of DNA-binding proteins proposed in the past and were demonstrated to have good performance. The results of the four existing methods and SVM-PSSM-DT were got by testing on benchmark dataset through jackknife validation.\naresults obtained by in-house implementation of DNAbinder [21]\nbresults obtained by in-house implementation of DNAbinder [21]\ncresults obtained by in-house implementation of DNA-Prot [74]\ndresults obtained by in-house implementation of iDNA-Prot [16]\nfresults obtained by using PSSM-DT as protein representation\nFigure 4 The ROC curves of several predictive methods on benchmark dataset. The receiver operating characteristic (ROC) curves of SVM-PSSM-DT and several other existing DNA-binding protein predictors were got by testing the models on benchmark dataset through jackknife validation, where the horizontal coordinate X is for the false positive rate or 1-SP and the vertical coordinate Y is for the true positive rate or SN and a good method would yield a curve close to the coordinate (0,1) meaning low false positive rate and high true positive rate. From Table 3 and Figure 4 we can see that SVM-PSSM-DT achieved the best performance with ACC of 79.96%, MCC of 0.62 and AUC of 86.50%, outperforming other four methods by 4.56-7.41% in terms of ACC, 0.12-0.18 in terms of MCC and 5-10.4% in terms of AUC. It indicates that PSSM-DT can advance the prhedictive performance of DNA-binding proteins identification from PSSM based sequence information.\n\nIndependent test\nIn order to further compare the predictive performance of SVM-PSSM-DT with other existing methods, we evaluated the proposed method on the independent dataset PDB186. It was recently constructed by Lou et al [75] to validate the quality of predictions, which consists 93 DNA-binding proteins and equal number of non DNA-binding proteins selected from PDB. Since there are some sequences from the benchmark dataset that shared high sequence identity with the independent dataset PDB186, the tool CD-HIT [77] was applied to remove the sequences from the benchmark dataset having more than 25% sequence identity to any one in a same subset in the independent dataset PDB186 to avoid homology bias. Table 4 lists the predictive results of the proposed method and several relevant existing methods, including iDNA-Prot [16], DNA-Prot [74], DNAbinder [21], DNABIND [34], and DNA-Threader [78], to our best knowledge.\nTable 4 Results on Independent dataset PDB186 of different predictorsa\nMethods Acc(%) MCC Sn(%) Sp(%) AUC(%)\niDNA-Prot 67.20 0.344 67.70 66.70 83.30\nDNA-Prot 61.80 0.240 69.90 53.80 79.60\nDNAbinder 60.80 0.216 57.00 64.50 60.70\nDNABIND 67.70 0.355 66.70 68.80 69.40\nDNA-Threader 59.70 0.279 23.70 95.70 N/A\nDBPPred 76.90 0.538 79.60 74.20 79.10\nPSSM-DT 80.00 0.647 87.09 72.83 87.40\nThe six methods in the front of the table are six useful predicting methods for identification of DNA-binding proteins proposed in the past and were demonstrated to have good performance. The results of the six existing predicting methods and the SVM-PSSM-DT were achieved on the dataset PDB186 by their model trained on benchmark dataset.\naThe results of iDNA-Prot [16], DNA-Prot [74], DNAbinder[21], DNABIND [34], DNA-Threader [78] and DDPPred [75] were obtained from [75]. Moreover, to provide a graphic illustration to show the performance comparisons of the SVM-PSSM-DT with other existing state-of-the-art predictors, the corresponding ROC curves were drawn in Figure 5. The experimental real value results of three predictors are provided by [75], including DBPPred [75], DNAbinder [21] and DNABIND [23]. And the real value outputs of the proposed method, iDNA-Prot and DNA-Prot are obtained by testing their predictors trained on benchmark dataset on independent dataset PDB186.\nFigure 5 The ROC curves of several predictive methods on Independent dataset. The receiver operating characteristic (ROC) curves of SVM-PSSM-DT and several other existing DNA-binding protein predictors were got by testing the models trained by benchmark dataset on independent dataset PDB186, where the horizontal coordinate X is for the false positive rate or 1-SP and the vertical coordinate Y is for the true positive rate or SN and a good method would yield a curve close to the coordinate (0,1) meaning low false positive rate and high true positive rate. From Table 4 and Figure 5 we can see that among the seven predictive methods, the proposed method has the highest performance with ACC of 80.00%, MCC of 0.674 and AUC of 87.40% and DBPPred is the known reported predictive method with the best predictive performance (ACC = 76.90%, MCC = 0.538 and AUC = 79.10%). So the independent prediction of SVM-PSSM-DT is improved by ACC of 3.105%, MCC of 0.136 and AUC of 8.30% when compared with the DBPPred method, indicating that SVM-PSSM-DT is an effective prediction model for DNA-binding protein identification.\n\nWeb-server guide\nWe have constructed a user-friendly web-server of SVM-PSSM-DT freely accessible to the public. Moreover, for the convenience of the vast majority of experimental scientists, a step-by-step guide is provided below on how to use the web-server to get the desired results.\nStep 1. Open the web-server by clicking the link [79] and you will see the home page as shown in Figure 6. Click on the Read Me button you can obtain the brief introduction about this web-server.\nFigure 6 The top page of the web-server. In the top page, you can type or copy and paste the query protein sequences into the input box at the center, obtain the brief introduction about this web-server by clicking on Read Me button and see information about FASTA format by clicking on the Example button right above the input box. Step 2. Either type or copy and paste the query protein sequences into the input box at the center of Figure 6. AS this server need calculate the PSSM profile for every protein sequence through PSI-BLAST, which is a time-consuming operation, thus it receive only a query protein sequence at a time. The input sequence should be in the FASTA format and example sequences in FASTA format can be seen by clicking on the Example button right above the input box.\nStep 3. Click on the Submit button to submit the query sequence to the server, then you will see the predicted results on your screen. For example, use the protein 1IGN chain B as a query sequence, you will see on your screen that the predictive result is \"DNA-binding protein\".","divisions":[{"label":"title","span":{"begin":0,"end":22}},{"label":"sec","span":{"begin":24,"end":2210}},{"label":"title","span":{"begin":24,"end":56}},{"label":"p","span":{"begin":57,"end":760}},{"label":"figure","span":{"begin":761,"end":1093}},{"label":"label","span":{"begin":761,"end":769}},{"label":"caption","span":{"begin":771,"end":1093}},{"label":"p","span":{"begin":771,"end":1093}},{"label":"p","span":{"begin":1094,"end":1554}},{"label":"table-wrap","span":{"begin":1555,"end":2210}},{"label":"label","span":{"begin":1555,"end":1562}},{"label":"caption","span":{"begin":1564,"end":1644}},{"label":"p","span":{"begin":1564,"end":1644}},{"label":"table","span":{"begin":1645,"end":1793}},{"label":"tr","span":{"begin":1645,"end":1679}},{"label":"th","span":{"begin":1645,"end":1652}},{"label":"th","span":{"begin":1654,"end":1660}},{"label":"th","span":{"begin":1662,"end":1665}},{"label":"th","span":{"begin":1667,"end":1672}},{"label":"th","span":{"begin":1674,"end":1679}},{"label":"tr","span":{"begin":1680,"end":1717}},{"label":"td","span":{"begin":1680,"end":1689}},{"label":"td","span":{"begin":1691,"end":1696}},{"label":"td","span":{"begin":1698,"end":1703}},{"label":"td","span":{"begin":1705,"end":1710}},{"label":"td","span":{"begin":1712,"end":1717}},{"label":"tr","span":{"begin":1718,"end":1756}},{"label":"td","span":{"begin":1718,"end":1727}},{"label":"td","span":{"begin":1729,"end":1734}},{"label":"td","span":{"begin":1736,"end":1741}},{"label":"td","span":{"begin":1743,"end":1749}},{"label":"td","span":{"begin":1751,"end":1756}},{"label":"tr","span":{"begin":1757,"end":1793}},{"label":"td","span":{"begin":1757,"end":1765}},{"label":"td","span":{"begin":1767,"end":1772}},{"label":"td","span":{"begin":1774,"end":1779}},{"label":"td","span":{"begin":1781,"end":1786}},{"label":"td","span":{"begin":1788,"end":1793}},{"label":"table-wrap-foot","span":{"begin":1794,"end":2210}},{"label":"p","span":{"begin":1794,"end":2043}},{"label":"p","span":{"begin":2044,"end":2099}},{"label":"p","span":{"begin":2100,"end":2155}},{"label":"p","span":{"begin":2156,"end":2210}},{"label":"sec","span":{"begin":2212,"end":8409}},{"label":"title","span":{"begin":2212,"end":2228}},{"label":"p","span":{"begin":2229,"end":2851}},{"label":"p","span":{"begin":2852,"end":3022}},{"label":"label","span":{"begin":2852,"end":2856}},{"label":"p","span":{"begin":3023,"end":3295}},{"label":"p","span":{"begin":3296,"end":5047}},{"label":"figure","span":{"begin":5048,"end":6251}},{"label":"label","span":{"begin":5048,"end":5056}},{"label":"caption","span":{"begin":5058,"end":6251}},{"label":"p","span":{"begin":5058,"end":6251}},{"label":"p","span":{"begin":6252,"end":6850}},{"label":"p","span":{"begin":6851,"end":8409}},{"label":"sec","span":{"begin":8411,"end":12132}},{"label":"title","span":{"begin":8411,"end":8463}},{"label":"p","span":{"begin":8464,"end":9361}},{"label":"table-wrap","span":{"begin":9362,"end":10459}},{"label":"label","span":{"begin":9362,"end":9369}},{"label":"caption","span":{"begin":9371,"end":9470}},{"label":"p","span":{"begin":9371,"end":9470}},{"label":"table","span":{"begin":9471,"end":9743}},{"label":"tr","span":{"begin":9471,"end":9513}},{"label":"th","span":{"begin":9471,"end":9478}},{"label":"th","span":{"begin":9480,"end":9486}},{"label":"th","span":{"begin":9488,"end":9491}},{"label":"th","span":{"begin":9493,"end":9498}},{"label":"th","span":{"begin":9500,"end":9505}},{"label":"th","span":{"begin":9507,"end":9513}},{"label":"tr","span":{"begin":9514,"end":9562}},{"label":"td","span":{"begin":9514,"end":9527}},{"label":"td","span":{"begin":9529,"end":9534}},{"label":"td","span":{"begin":9536,"end":9541}},{"label":"td","span":{"begin":9543,"end":9548}},{"label":"td","span":{"begin":9550,"end":9555}},{"label":"td","span":{"begin":9557,"end":9562}},{"label":"tr","span":{"begin":9563,"end":9612}},{"label":"td","span":{"begin":9563,"end":9577}},{"label":"td","span":{"begin":9579,"end":9584}},{"label":"td","span":{"begin":9586,"end":9591}},{"label":"td","span":{"begin":9593,"end":9598}},{"label":"td","span":{"begin":9600,"end":9605}},{"label":"td","span":{"begin":9607,"end":9612}},{"label":"tr","span":{"begin":9613,"end":9659}},{"label":"td","span":{"begin":9613,"end":9624}},{"label":"td","span":{"begin":9626,"end":9631}},{"label":"td","span":{"begin":9633,"end":9638}},{"label":"td","span":{"begin":9640,"end":9645}},{"label":"td","span":{"begin":9647,"end":9652}},{"label":"td","span":{"begin":9654,"end":9659}},{"label":"tr","span":{"begin":9660,"end":9699}},{"label":"td","span":{"begin":9660,"end":9664}},{"label":"td","span":{"begin":9666,"end":9671}},{"label":"td","span":{"begin":9673,"end":9678}},{"label":"td","span":{"begin":9680,"end":9685}},{"label":"td","span":{"begin":9687,"end":9692}},{"label":"td","span":{"begin":9694,"end":9699}},{"label":"tr","span":{"begin":9700,"end":9743}},{"label":"td","span":{"begin":9700,"end":9708}},{"label":"td","span":{"begin":9710,"end":9715}},{"label":"td","span":{"begin":9717,"end":9722}},{"label":"td","span":{"begin":9724,"end":9729}},{"label":"td","span":{"begin":9731,"end":9736}},{"label":"td","span":{"begin":9738,"end":9743}},{"label":"table-wrap-foot","span":{"begin":9744,"end":10459}},{"label":"p","span":{"begin":9744,"end":10144}},{"label":"p","span":{"begin":10145,"end":10210}},{"label":"p","span":{"begin":10211,"end":10277}},{"label":"p","span":{"begin":10278,"end":10341}},{"label":"p","span":{"begin":10342,"end":10398}},{"label":"p","span":{"begin":10399,"end":10459}},{"label":"p","span":{"begin":10460,"end":11288}},{"label":"figure","span":{"begin":11289,"end":11846}},{"label":"label","span":{"begin":11289,"end":11297}},{"label":"caption","span":{"begin":11299,"end":11846}},{"label":"p","span":{"begin":11299,"end":11846}},{"label":"p","span":{"begin":11847,"end":12132}},{"label":"sec","span":{"begin":12134,"end":14983}},{"label":"title","span":{"begin":12134,"end":12177}},{"label":"p","span":{"begin":12178,"end":13002}},{"label":"table-wrap","span":{"begin":13003,"end":14035}},{"label":"label","span":{"begin":13003,"end":13010}},{"label":"caption","span":{"begin":13012,"end":13094}},{"label":"p","span":{"begin":13012,"end":13094}},{"label":"table","span":{"begin":13095,"end":13395}},{"label":"tr","span":{"begin":13095,"end":13139}},{"label":"th","span":{"begin":13095,"end":13101}},{"label":"th","span":{"begin":13103,"end":13110}},{"label":"th","span":{"begin":13112,"end":13115}},{"label":"th","span":{"begin":13117,"end":13123}},{"label":"th","span":{"begin":13125,"end":13131}},{"label":"th","span":{"begin":13133,"end":13139}},{"label":"tr","span":{"begin":13140,"end":13199}},{"label":"td","span":{"begin":13140,"end":13164}},{"label":"td","span":{"begin":13166,"end":13171}},{"label":"td","span":{"begin":13173,"end":13178}},{"label":"td","span":{"begin":13180,"end":13185}},{"label":"td","span":{"begin":13187,"end":13192}},{"label":"td","span":{"begin":13194,"end":13199}},{"label":"tr","span":{"begin":13200,"end":13260}},{"label":"td","span":{"begin":13200,"end":13225}},{"label":"td","span":{"begin":13227,"end":13232}},{"label":"td","span":{"begin":13234,"end":13239}},{"label":"td","span":{"begin":13241,"end":13246}},{"label":"td","span":{"begin":13248,"end":13253}},{"label":"td","span":{"begin":13255,"end":13260}},{"label":"tr","span":{"begin":13261,"end":13305}},{"label":"td","span":{"begin":13261,"end":13270}},{"label":"td","span":{"begin":13272,"end":13277}},{"label":"td","span":{"begin":13279,"end":13284}},{"label":"td","span":{"begin":13286,"end":13291}},{"label":"td","span":{"begin":13293,"end":13298}},{"label":"td","span":{"begin":13300,"end":13305}},{"label":"tr","span":{"begin":13306,"end":13351}},{"label":"td","span":{"begin":13306,"end":13316}},{"label":"td","span":{"begin":13318,"end":13323}},{"label":"td","span":{"begin":13325,"end":13330}},{"label":"td","span":{"begin":13332,"end":13337}},{"label":"td","span":{"begin":13339,"end":13344}},{"label":"td","span":{"begin":13346,"end":13351}},{"label":"tr","span":{"begin":13352,"end":13395}},{"label":"td","span":{"begin":13352,"end":13360}},{"label":"td","span":{"begin":13362,"end":13367}},{"label":"td","span":{"begin":13369,"end":13374}},{"label":"td","span":{"begin":13376,"end":13381}},{"label":"td","span":{"begin":13383,"end":13388}},{"label":"td","span":{"begin":13390,"end":13395}},{"label":"table-wrap-foot","span":{"begin":13396,"end":14035}},{"label":"p","span":{"begin":13396,"end":13723}},{"label":"p","span":{"begin":13724,"end":13786}},{"label":"p","span":{"begin":13787,"end":13849}},{"label":"p","span":{"begin":13850,"end":13911}},{"label":"p","span":{"begin":13912,"end":13974}},{"label":"p","span":{"begin":13975,"end":14035}},{"label":"figure","span":{"begin":14036,"end":14586}},{"label":"label","span":{"begin":14036,"end":14044}},{"label":"caption","span":{"begin":14046,"end":14586}},{"label":"p","span":{"begin":14046,"end":14586}},{"label":"p","span":{"begin":14587,"end":14983}},{"label":"sec","span":{"begin":14985,"end":18442}},{"label":"title","span":{"begin":14985,"end":15001}},{"label":"p","span":{"begin":15002,"end":15912}},{"label":"table-wrap","span":{"begin":15913,"end":16812}},{"label":"label","span":{"begin":15913,"end":15920}},{"label":"caption","span":{"begin":15922,"end":15984}},{"label":"p","span":{"begin":15922,"end":15984}},{"label":"table","span":{"begin":15985,"end":16336}},{"label":"tr","span":{"begin":15985,"end":16027}},{"label":"th","span":{"begin":15985,"end":15992}},{"label":"th","span":{"begin":15994,"end":16000}},{"label":"th","span":{"begin":16002,"end":16005}},{"label":"th","span":{"begin":16007,"end":16012}},{"label":"th","span":{"begin":16014,"end":16019}},{"label":"th","span":{"begin":16021,"end":16027}},{"label":"tr","span":{"begin":16028,"end":16072}},{"label":"td","span":{"begin":16028,"end":16037}},{"label":"td","span":{"begin":16039,"end":16044}},{"label":"td","span":{"begin":16046,"end":16051}},{"label":"td","span":{"begin":16053,"end":16058}},{"label":"td","span":{"begin":16060,"end":16065}},{"label":"td","span":{"begin":16067,"end":16072}},{"label":"tr","span":{"begin":16073,"end":16116}},{"label":"td","span":{"begin":16073,"end":16081}},{"label":"td","span":{"begin":16083,"end":16088}},{"label":"td","span":{"begin":16090,"end":16095}},{"label":"td","span":{"begin":16097,"end":16102}},{"label":"td","span":{"begin":16104,"end":16109}},{"label":"td","span":{"begin":16111,"end":16116}},{"label":"tr","span":{"begin":16117,"end":16161}},{"label":"td","span":{"begin":16117,"end":16126}},{"label":"td","span":{"begin":16128,"end":16133}},{"label":"td","span":{"begin":16135,"end":16140}},{"label":"td","span":{"begin":16142,"end":16147}},{"label":"td","span":{"begin":16149,"end":16154}},{"label":"td","span":{"begin":16156,"end":16161}},{"label":"tr","span":{"begin":16162,"end":16204}},{"label":"td","span":{"begin":16162,"end":16169}},{"label":"td","span":{"begin":16171,"end":16176}},{"label":"td","span":{"begin":16178,"end":16183}},{"label":"td","span":{"begin":16185,"end":16190}},{"label":"td","span":{"begin":16192,"end":16197}},{"label":"td","span":{"begin":16199,"end":16204}},{"label":"tr","span":{"begin":16205,"end":16250}},{"label":"td","span":{"begin":16205,"end":16217}},{"label":"td","span":{"begin":16219,"end":16224}},{"label":"td","span":{"begin":16226,"end":16231}},{"label":"td","span":{"begin":16233,"end":16238}},{"label":"td","span":{"begin":16240,"end":16245}},{"label":"td","span":{"begin":16247,"end":16250}},{"label":"tr","span":{"begin":16251,"end":16293}},{"label":"td","span":{"begin":16251,"end":16258}},{"label":"td","span":{"begin":16260,"end":16265}},{"label":"td","span":{"begin":16267,"end":16272}},{"label":"td","span":{"begin":16274,"end":16279}},{"label":"td","span":{"begin":16281,"end":16286}},{"label":"td","span":{"begin":16288,"end":16293}},{"label":"tr","span":{"begin":16294,"end":16336}},{"label":"td","span":{"begin":16294,"end":16301}},{"label":"td","span":{"begin":16303,"end":16308}},{"label":"td","span":{"begin":16310,"end":16315}},{"label":"td","span":{"begin":16317,"end":16322}},{"label":"td","span":{"begin":16324,"end":16329}},{"label":"td","span":{"begin":16331,"end":16336}},{"label":"table-wrap-foot","span":{"begin":16337,"end":16812}},{"label":"p","span":{"begin":16337,"end":16676}},{"label":"p","span":{"begin":16677,"end":16812}},{"label":"p","span":{"begin":16813,"end":17323}},{"label":"figure","span":{"begin":17324,"end":17885}},{"label":"label","span":{"begin":17324,"end":17332}},{"label":"caption","span":{"begin":17334,"end":17885}},{"label":"p","span":{"begin":17334,"end":17885}},{"label":"p","span":{"begin":17886,"end":18442}},{"label":"title","span":{"begin":18444,"end":18460}},{"label":"p","span":{"begin":18461,"end":18730}},{"label":"p","span":{"begin":18731,"end":18926}},{"label":"figure","span":{"begin":18927,"end":19260}},{"label":"label","span":{"begin":18927,"end":18935}},{"label":"caption","span":{"begin":18937,"end":19260}},{"label":"p","span":{"begin":18937,"end":19260}},{"label":"p","span":{"begin":19261,"end":19719}}],"tracks":[{"project":"2_test","denotations":[{"id":"25708928-23029559-14842653","span":{"begin":2368,"end":2370},"obj":"23029559"},{"id":"25708928-12967962-14842654","span":{"begin":2371,"end":2373},"obj":"12967962"},{"id":"25708928-16551468-14842655","span":{"begin":4021,"end":4023},"obj":"16551468"},{"id":"25708928-9687024-14842656","span":{"begin":4172,"end":4174},"obj":"9687024"},{"id":"25708928-19865164-14842657","span":{"begin":4607,"end":4609},"obj":"19865164"},{"id":"25708928-15356290-14842658","span":{"begin":4944,"end":4946},"obj":"15356290"},{"id":"25708928-19385697-14842659","span":{"begin":8794,"end":8796},"obj":"19385697"},{"id":"25708928-24475169-14842660","span":{"begin":8931,"end":8933},"obj":"24475169"},{"id":"25708928-19706744-14842661","span":{"begin":9164,"end":9166},"obj":"19706744"},{"id":"25708928-24475169-14842662","span":{"begin":10338,"end":10340},"obj":"24475169"},{"id":"25708928-19706744-14842663","span":{"begin":10395,"end":10397},"obj":"19706744"},{"id":"25708928-19385697-14842664","span":{"begin":12406,"end":12408},"obj":"19385697"},{"id":"25708928-19385697-14842665","span":{"begin":13908,"end":13910},"obj":"19385697"},{"id":"25708928-24475169-14842666","span":{"begin":15211,"end":15213},"obj":"24475169"},{"id":"25708928-19385697-14842667","span":{"begin":15832,"end":15834},"obj":"19385697"},{"id":"25708928-16551468-14842668","span":{"begin":15862,"end":15864},"obj":"16551468"},{"id":"25708928-19911048-14842669","span":{"begin":15885,"end":15887},"obj":"19911048"},{"id":"25708928-19385697-14842670","span":{"begin":16719,"end":16721},"obj":"19385697"},{"id":"25708928-16551468-14842671","span":{"begin":16748,"end":16750},"obj":"16551468"},{"id":"25708928-19911048-14842672","span":{"begin":16767,"end":16769},"obj":"19911048"},{"id":"25708928-24475169-14842673","span":{"begin":16784,"end":16786},"obj":"24475169"},{"id":"25708928-24475169-14842674","span":{"begin":16808,"end":16810},"obj":"24475169"},{"id":"25708928-24475169-14842675","span":{"begin":17087,"end":17089},"obj":"24475169"},{"id":"25708928-24475169-14842676","span":{"begin":17111,"end":17113},"obj":"24475169"},{"id":"25708928-20089514-14842677","span":{"begin":17144,"end":17146},"obj":"20089514"}],"attributes":[{"subj":"25708928-23029559-14842653","pred":"source","obj":"2_test"},{"subj":"25708928-12967962-14842654","pred":"source","obj":"2_test"},{"subj":"25708928-16551468-14842655","pred":"source","obj":"2_test"},{"subj":"25708928-9687024-14842656","pred":"source","obj":"2_test"},{"subj":"25708928-19865164-14842657","pred":"source","obj":"2_test"},{"subj":"25708928-15356290-14842658","pred":"source","obj":"2_test"},{"subj":"25708928-19385697-14842659","pred":"source","obj":"2_test"},{"subj":"25708928-24475169-14842660","pred":"source","obj":"2_test"},{"subj":"25708928-19706744-14842661","pred":"source","obj":"2_test"},{"subj":"25708928-24475169-14842662","pred":"source","obj":"2_test"},{"subj":"25708928-19706744-14842663","pred":"source","obj":"2_test"},{"subj":"25708928-19385697-14842664","pred":"source","obj":"2_test"},{"subj":"25708928-19385697-14842665","pred":"source","obj":"2_test"},{"subj":"25708928-24475169-14842666","pred":"source","obj":"2_test"},{"subj":"25708928-19385697-14842667","pred":"source","obj":"2_test"},{"subj":"25708928-16551468-14842668","pred":"source","obj":"2_test"},{"subj":"25708928-19911048-14842669","pred":"source","obj":"2_test"},{"subj":"25708928-19385697-14842670","pred":"source","obj":"2_test"},{"subj":"25708928-16551468-14842671","pred":"source","obj":"2_test"},{"subj":"25708928-19911048-14842672","pred":"source","obj":"2_test"},{"subj":"25708928-24475169-14842673","pred":"source","obj":"2_test"},{"subj":"25708928-24475169-14842674","pred":"source","obj":"2_test"},{"subj":"25708928-24475169-14842675","pred":"source","obj":"2_test"},{"subj":"25708928-24475169-14842676","pred":"source","obj":"2_test"},{"subj":"25708928-20089514-14842677","pred":"source","obj":"2_test"}]}],"config":{"attribute types":[{"pred":"source","value type":"selection","values":[{"id":"2_test","color":"#e5ec93","default":true}]}]}}