Improved method for predicting protein fold patterns with ensemble classifiers

Protein folding is recognized as a critical problem in the field of biophysics in the 21st century. Predicting protein-folding patterns is challenging due to the complex structure of proteins. In an attempt to solve this problem, we employed ensemble classifiers to improve prediction accuracy. In our experiments, 188-dimensional features were extracted based on the composition and physical-chemical property of proteins and 20-dimensional features were selected using a coupled position-specific scoring matrix. Compared with traditional prediction methods, these methods were superior in terms of prediction accuracy. The 188-dimensional feature-based method achieved 71.2% accuracy in five cross-validations. The accuracy rose to 77% when we used a 20-dimensional feature vector. These methods were used on recent data, with 54.2% accuracy. Source codes and dataset, together with web server and software tools for prediction, are available at: http://datamining.xmu.edu.cn/main/~cwc/ProteinPredict.html.     

Prediction of RNA-binding proteins from primary sequence by a support vector machine approach

Elucidation of the interaction of proteins with different molecules is of significance in the understanding of cellular processes. Computational methods have been developed for the prediction of protein-protein interactions. But insufficient attention has been paid to the prediction of protein-RNA interactions, which play central roles in regulating gene expression and certain RNA-mediated enzymatic processes. This work explored the use of a machine learning method, support vector machines (SVM), for the prediction of RNA-binding proteins directly from their primary sequence. Based on the knowledge of known RNA-binding and non-RNA-binding proteins, an SVM system was trained to recognize RNA-binding proteins. A total of 4011 RNA-binding and 9781 non-RNA-binding proteins was used to train and test the SVM classification system, and an independent set of 447 RNA-binding and 4881 non-RNA-binding proteins was used to evaluate the classification accuracy. Testing results using this independent evaluation set show a prediction accuracy of 94.1%, 79.3%, and 94.1% for rRNA-, mRNA-, and tRNA-binding proteins, and 98.7%, 96.5%, and 99.9% for non-rRNA-, non-mRNA-, and non-tRNA-binding proteins, respectively. The SVM classification system was further tested on a small class of snRNA-binding proteins with only 60 available sequences. The prediction accuracy is 40.0% and 99.9% for snRNA-binding and non-snRNA-binding proteins, indicating a need for a sufficient number of proteins to train SVM. The SVM classification systems trained in this work were added to our Web-based protein functional classification software SVMProt, at http://jing.cz3.nus.edu.sg/cgi-bin/svmprot.cgi. Our study suggests the potential of SVM as a useful tool for facilitating the prediction of protein-RNA interactions.     

A seqlet-based maximum entropy Markov approach for protein secondary structure prediction

A novel method for predicting the secondary structures of proteins from amino acid sequence has been presented. The protein secondary structure seqlets that are analogous to the words in natural language have been extracted. These seqlets will capture the relationship between amino acid sequence and the secondary structures of proteins and further form the protein secondary structure dictionary. To be elaborate, the dictionary is organism-specific. Protein secondary structure prediction is formulated as an integrated word segmentation and part of speech tagging problem. The word-lattice is used to represent the results of the word segmentation and the maximum entropy model is used to calculate the probability of a seqlet tagged as a certain secondary structure type. The method is markovian in the seqlets, permitting efficient exact calculation of the posterior probability distribution over all possible word segmentations and their tags by viterbi algorithm. The optimal segmentations and their tags are computed as the results of protein secondary structure prediction. The method is applied to predict the secondary structures of proteins of four organisms respectively and compared with the PHD method. The results show that the performance of this method is higher than that of PHD by about 3.9% Q3 accuracy and 4.6% SOV accuracy. Combining with the local similarity protein sequences that are obtained by BLAST can give better prediction. The method is also tested on the 50 CASP5 target proteins with Q₃ accuracy 78.9% and SOV accuracy 77.1%. A web server for protein secondary structure prediction has been constructed which is available at http://www.insun.hit.edu.cn:81/demos/biology/index.html.     

COMSAT: Residue contact prediction of transmembrane proteins based on support vector machines and mixed integer linear programming

In this article, we present COMSAT, a hybrid framework for residue contact prediction of transmembrane (TM) proteins, integrating a support vector machine (SVM) method and a mixed integer linear programming (MILP) method. COMSAT consists of two modules: COMSAT_SVM which is trained mainly on position–specific scoring matrix features, and COMSAT_MILP which is an ab initio method based on optimization models. Contacts predicted by the SVM model are ranked by SVM confidence scores, and a threshold is trained to improve the reliability of the predicted contacts. For TM proteins with no contacts above the threshold, COMSAT_MILP is used. The proposed hybrid contact prediction scheme was tested on two independent TM protein sets based on the contact definition of 14 Å between Cα‐Cα atoms. First, using a rigorous leave‐one‐protein‐out cross validation on the training set of 90 TM proteins, an accuracy of 66.8%, a coverage of 12.3%, a specificity of 99.3% and a Matthews' correlation coefficient (MCC) of 0.184 were obtained for residue pairs that are at least six amino acids apart. Second, when tested on a test set of 87 TM proteins, the proposed method showed a prediction accuracy of 64.5%, a coverage of 5.3%, a specificity of 99.4% and a MCC of 0.106. COMSAT shows satisfactory results when compared with 12 other state‐of‐the‐art predictors, and is more robust in terms of prediction accuracy as the length and complexity of TM protein increase. COMSAT is freely accessible at http://hpcc.siat.ac.cn/COMSAT/ . Proteins 2016; 84:332–348. © 2016 Wiley Periodicals, Inc.     

Cysteine separations profiles on protein sequences infer disulfide connectivity

MOTIVATION: Disulfide bonds play an important role in protein folding. A precise prediction of disulfide connectivity can strongly reduce the conformational search space and increase the accuracy in protein structure prediction. Conventional disulfide connectivity predictions use sequence information, and prediction accuracy is limited. Here, by using an alternative scheme with global information for disulfide connectivity prediction, higher performance is obtained with respect to other approaches. RESULT: Cysteine separation profiles have been used to predict the disulfide connectivity of proteins. The separations among oxidized cysteine residues on a protein sequence have been encoded into vectors named cysteine separation profiles (CSPs). Through comparisons of their CSPs, the disulfide connectivity of a test protein is inferred from a non-redundant template set. For non-redundant proteins in SwissProt 39 (SP39) sharing less than 30% sequence identity, the prediction accuracy of a fourfold cross-validation is 49%. The prediction accuracy of disulfide connectivity for proteins in SwissProt 43 (SP43) is even higher (53%). The relationship between the similarity of CSPs and the prediction accuracy is also discussed. The method proposed in this work is relatively simple and can generate higher accuracies compared to conventional methods. It may be also combined with other algorithms for further improvements in protein structure prediction. AVAILABILITY: The program and datasets are available from the authors upon request. CONTACT: cykao@csie.ntu.edu.tw.     

Discrimination of outer membrane proteins using support vector machines

MOTIVATION: Discriminating outer membrane proteins from other folding types of globular and membrane proteins is an important task both for dissecting outer membrane proteins (OMPs) from genomic sequences and for the successful prediction of their secondary and tertiary structures. RESULTS: We have developed a method based on support vector machines using amino acid composition and residue pair information. Our approach with amino acid composition has correctly predicted the OMPs with a cross-validated accuracy of 94% in a set of 208 proteins. Further, this method has successfully excluded 633 of 673 globular proteins and 191 of 206 alpha-helical membrane proteins. We obtained an overall accuracy of 92% for correctly picking up the OMPs from a dataset of 1087 proteins belonging to all different types of globular and membrane proteins. Furthermore, residue pair information improved the accuracy from 92 to 94%. This accuracy of discriminating OMPs is higher than that of other methods in the literature, which could be used for dissecting OMPs from genomic sequences. AVAILABILITY: Discrimination results are available at http://tmbeta-svm.cbrc.jp.     

Improving transmembrane protein consensus topology prediction using inter-helical interaction

Alpha helix transmembrane proteins (αTMPs) represent roughly 30% of all open reading frames (ORFs) in a typical genome and are involved in many critical biological processes. Due to the special physicochemical properties, it is hard to crystallize and obtain high resolution structures experimentally, thus, sequence-based topology prediction is highly desirable for the study of transmembrane proteins (TMPs), both in structure prediction and function prediction. Various model-based topology prediction methods have been developed, but the accuracy of those individual predictors remain poor due to the limitation of the methods or the features they used. Thus, the consensus topology prediction method becomes practical for high accuracy applications by combining the advances of the individual predictors. Here, based on the observation that inter-helical interactions are commonly found within the transmembrane helixes (TMHs) and strongly indicate the existence of them, we present a novel consensus topology prediction method for αTMPs, CNTOP, which incorporates four top leading individual topology predictors, and further improves the prediction accuracy by using the predicted inter-helical interactions. The method achieved 87% prediction accuracy based on a benchmark dataset and 78% accuracy based on a non-redundant dataset which is composed of polytopic αTMPs. Our method derives the highest topology accuracy than any other individual predictors and consensus predictors, at the same time, the TMHs are more accurately predicted in their length and locations, where both the false positives (FPs) and the false negatives (FNs) decreased dramatically. The CNTOP is available at: http://ccst.jlu.edu.cn/JCSB/cntop/CNTOP.html.     

A seqlet-based maximum entropy Markov approach for protein secondary structure prediction

1 Introduction The prediction of protein structure and function from amino acid sequences is one of the most impor-tant problems in molecular biology. This problem is becoming more pressing as the number of known pro-tein sequences is explored as a result of genome and other sequencing projects, and the protein sequence- structure gap is widening rapidly[1]. Therefore, com-putational tools to predict protein structures are needed to narrow the widening gap. Although the prediction of three dim…     

Prediction whether a human cDNA sequence contains initiation codon by combining statistical information and similarity with protein sequences

MOTIVATION: In the previous works, we developed ATGpr, a computer program for predicting the fullness of a cDNA, i.e. whether it contains an initiation codon or not. Statistical information of short nucleotide fragments was fully exploited in the prediction algorithm. However, sequence similarities to known proteins, which are becoming increasingly available due to recent rapid growth of protein database, were not used in the prediction. In this work, we present a new prediction algorithm based on both statistical and similarity information, which provides better performance in sensitivity and specificity. RESULTS: We evaluated the accuracy of ATGpr for predicting fullness of cDNA sequences from human clustered ESTs of UniGene, and we obtained specificity, sensitivity, and correlation coefficient of this prediction. Specificity and sensitivity crossed at 46% over the ATGpr score threshold of 0.33 and the maximum correlation coefficient of 0.34 was obtained at this threshold. Without ATGpr we found it effective to use alignments with known proteins for predicting the fullness of cDNA sequences. That is, specificity increased monotonously as similarity (identity of the alignments) increased. Specificity was achieved greater than 80% if identity was greater than 40%. For more effective prediction of fullness of cDNA sequences we combined the similarity (identity of query sequence) with known proteins and ATGpr score. As a result, specificity became greater than 80% if identity was greater than 20%. AVAILABILITY: The prediction program, called ATGpr_ sim, is available at http://www.hri.co.jp/atgpr/ATGpr_sim.html CONTACT: nisikawa@crl.hitachi.co.jp     

Identify submitochondria and subchloroplast locations with pseudo amino acid composition: Approach from the strategy of discrete wavelet transform feature extraction

It is very challenging and complicated to predict protein locations at the sub-subcellular level. The key to enhancing the prediction quality for protein sub-subcellular locations is to grasp the core features of a protein that can discriminate among proteins with different subcompartment locations. In this study, a different formulation of pseudoamino acid composition by the approach of discrete wavelet transform feature extraction was developed to predict submitochondria and subchloroplast locations. As a result of jackknife cross-validation, with our method, it can efficiently distinguish mitochondrial proteins from chloroplast proteins with total accuracy of 98.8% and obtained a promising total accuracy of 93.38% for predicting submitochondria locations. Especially the predictive accuracy for mitochondrial outer membrane and chloroplast thylakoid lumen were 82.93% and 82.22%, respectively, showing an improvement of 4.88% and 27.22% when other existing methods were compared. The results indicated that the proposed method might be employed as a useful assistant technique for identifying sub-subcellular locations. We have implemented our algorithm as an online service called SubIdent (http://bioinfo.ncu.edu.cn/services.aspx).     

Sann: solvent accessibility prediction of proteins by nearest neighbor method

We present a method to predict the solvent accessibility of proteins which is based on a nearest neighbor method applied to the sequence profiles. Using the method, continuous real-value prediction as well as two-state and three-state discrete predictions can be obtained. The method utilizes the z-score value of the distance measure in the feature vector space to estimate the relative contribution among the k-nearest neighbors for prediction of the discrete and continuous solvent accessibility. The Solvent accessibility database is constructed from 5717 proteins extracted from PISCES culling server with the cutoff of 25% sequence identities. Using optimal parameters, the prediction accuracies (for discrete predictions) of 78.38% (two-state prediction with the threshold of 25%), 65.1% (three-state prediction with the thresholds of 9 and 36%), and the Pearson correlation coefficient (between the predicted and true RSA's for continuous prediction) of 0.676 are achieved An independent benchmark test was performed with the CASP8 targets where we find that the proposed method outperforms existing methods. The prediction accuracies are 80.89% (for two state prediction with the threshold of 25%), 67.58% (three-state prediction), and the Pearson correlation coefficient of 0.727 (for continuous prediction) with mean absolute error of 0.148. We have also investigated the effect of increasing database sizes on the prediction accuracy, where additional improvement in the accuracy is observed as the database size increases. The SANN web server is available at http://lee.kias.re.kr/~newton/sann/.     

Topology prediction for helical transmembrane proteins at 86% accuracy.

Previously, we introduced a neural network system predicting locations of transmembrane helices (HTMs) based on evolutionary profiles (PHDhtm, Rost B, Casadio R, Fariselli P, Sander C, 1995, Protein Sci 4:521-533). Here, we describe an improvement and an extension of that system. The improvement is achieved by a dynamic programming-like algorithm that optimizes helices compatible with the neural network output. The extension is the prediction of topology (orientation of first loop region with respect to membrane) by applying to the refined prediction the observation that positively charged residues are more abundant in extra-cytoplasmic regions. Furthermore, we introduce a method to reduce the number of false positives, i.e., proteins falsely predicted with membrane helices. The evaluation of prediction accuracy is based on a cross-validation and a double-blind test set (in total 131 proteins). The final method appears to be more accurate than other methods published: (1) For almost 89% (+/-3%) of the test proteins, all HTMs are predicted correctly. (2) For more than 86% (+/-3%) of the proteins, topology is predicted correctly. (3) We define reliability indices that correlate with prediction accuracy: for one half of the proteins, segment accuracy raises to 98%; and for two-thirds, accuracy of topology prediction is 95%. (4) The rate of proteins for which HTMs are predicted falsely is below 2% (+/-1%). Finally, the method is applied to 1,616 sequences of Haemophilus influenzae. We predict 19% of the genome sequences to contain one or more HTMs. This appears to be lower than what we predicted previously for the yeast VIII chromosome (about 25%).     

RSLpred: an integrative system for predicting subcellular localization of rice proteins combining compositional and evolutionary information

The attainment of complete map‐based sequence for rice (Oryza sativa) is clearly a major milestone for the research community. Identifying the localization of encoded proteins is the key to understanding their functional characteristics and facilitating their purification. Our proposed method, RSLpred, is an effort in this direction for genome‐scale subcellular prediction of encoded rice proteins. First, the support vector machine (SVM)‐based modules have been developed using traditional amino acid‐, dipeptide‐ (i+1) and four parts‐amino acid composition and achieved an overall accuracy of 81.43, 80.88 and 81.10%, respectively. Secondly, a similarity search‐based module has been developed using position‐specific iterated‐basic local alignment search tool and achieved 68.35% accuracy. Another module developed using evolutionary information of a protein sequence extracted from position‐specific scoring matrix achieved an accuracy of 87.10%. In this study, a large number of modules have been developed using various encoding schemes like higher‐order dipeptide composition, N‐ and C‐terminal, splitted amino acid composition and the hybrid information. In order to benchmark RSLpred, it was tested on an independent set of rice proteins where it outperformed widely used prediction methods such as TargetP, Wolf‐PSORT, PA‐SUB, Plant‐Ploc and ESLpred. To assist the plant research community, an online web tool ‘RSLpred’ has been developed for subcellular prediction of query rice proteins, which is freely accessible at http://www.imtech.res.in/raghava/rslpred.     

PocketDepth: a new depth based algorithm for identification of ligand binding sites in proteins

Predicting functional sites in proteins is important in structural biology for understanding the function and also for structure-based drug design. Here we report a new binding site prediction method PocketDepth, which is geometry based and uses a depth based clustering. Depth is an important parameter considered during protein structure visualisation and analysis but has been used more often intuitively than systematically. Our current implementation of depth reflects how central a given subspace is to a putative pocket. We have tested the algorithm against PDBbind, a large curated set of 1091 proteins. A prediction was considered a true-positive if the predicted pocket had at least 10% overlap with the actual ligand. Two different parameter sets, 'deeper' and 'surface' were used, for wider coverage of different types of binding sites in proteins. With deeper parameters, true-positives were observed for 841 proteins, resulting in a prediction accuracy of 77%, for any ranked prediction. Of these, 55.2% were first ranked predictions, whereas 91.2% and 97.4% were covered in the first 5 and 10 ranks, respectively. With the 'surface' parameters, a prediction rate of 95.8% was observed, albeit with much poorer ranks. The deeper set identified pocket boundaries more precisely and yielded better ranks, while the latter missed fewer predictions and hence had better coverage. The two parameter sets were therefore algorithmically combined, resulting in prediction accuracies of 96.5% for any ranked prediction. About 41.8% of these were in the first rank, 82% and 94% were in top 5 and 10 ranks, respectively. The algorithm is available at http://proline.physics.iisc.ernet.in/pocketdepth.     

Monte Carlo simulation studies on the prediction of protein folding types from amino acid composition.

In the methodology development for statistical prediction of protein structures, the founders of different methods usually selected different sets of proteins to test their predicted results. Therefore, it is hard to make a fair comparison according to the results they reported. Even if the predictions by different methods are performed for the same set of proteins, there is still such a problem: a method better that the other for one set of proteins would not necessarily remain so when applied to another set of proteins. To tackle this problem, a Monte Carlo simulation method is proposed to establish an objective criterion to measure the accuracy of prediction for the protein folding type. Such an objective accuracy is actually corresponding to the asymptotical limit genereated during the Monte Carlo simulation process. Based on that, it has been found that the average objective accuracy for predicting the all-alpha, all-beta, alpha + beta, and alpha/beta proteins by the least Euclid's distance method (Nakashima, H., K. Nishikawa, and T. Ooi. 1986. J. Biochem. 99:152-162) is 73.0% and that by the least Minkowski's distance method (Chou, P.Y. 1989. Prediction in Protein Structure and the Principles of Protein Conformation. Plenum Press. New York. 549-586) is 70.9%, indicating that the former is better than the latter. However, according to the original reports, the latter claimed a rate of correct prediction with 79.7% but the former with only 70.2%, leading to a completely opposite conclusion. This indicates the necessity of establishing an objective criterion, and a comparison is meaningful only when it is based on the objective criterion. The simulation method and the idea developed here also can be applied to examine any other statistical prediction methods.     

SPSSM8: An accurate approach for predicting eight-state secondary structures of proteins

Protein eight-state secondary structure prediction is challenging, but is necessary to determine protein structure and function. Here, we report the development of a novel approach, SPSSM8, to predict eight-state secondary structures of proteins accurately from sequences based on the structural position-specific scoring matrix (SPSSM). The SPSSM has been successfully utilized to predict three-state secondary structures. Now we employ an eight-state SPSSM as a feature that is obtained from sequence structure alignment against a large database of 9 million sequences with putative structural information. The SPSSM8 uses a low sequence identity dataset (9062 entries) as a training set and conditional random field for the classification algorithm. The SPSSM8 achieved an average eight-state secondary structure accuracy (Q8) of 71.7% (Q3, 81.6%) for an independent testing set (463 entries), which had an improved accuracy of 10.1% and 4.6% compared with SSPro8 and CNF, respectively, and significantly improved the accuracy of eight-state secondary structure prediction. For CASP 9 dataset (92 entries) the SPSSM8 achieved a Q8 accuracy of 80.1% (Q3, 83.0%). The SPSSM8 was confirmed as an outstanding predictor for eight-state secondary structures of proteins. SPSSM8 is freely available at http://cal.tongji.edu.cn/SPSSM8.     

Prediction of the functional class of lipid binding proteins from sequence-derived properties irrespective of sequence similarity

Lipid binding proteins play important roles in signaling, regulation, membrane trafficking, immune response, lipid metabolism, and transport. Because of their functional and sequence diversity, it is desirable to explore additional methods for predicting lipid binding proteins irrespective of sequence similarity. This work explores the use of support vector machines (SVMs) as such a method. SVM prediction systems are developed using 14,776 lipid binding and 133,441 nonlipid binding proteins and are evaluated by an independent set of 6,768 lipid binding and 64,761 nonlipid binding proteins. The computed prediction accuracy is 78.9, 79.5, 82.2, 79.5, 84.4, 76.6, 90.6, 79.0, and 89.9% for lipid degradation, lipid metabolism, lipid synthesis, lipid transport, lipid binding, lipopolysaccharide biosynthesis, lipoprotein, lipoyl, and all lipid binding proteins, respectively. The accuracy for the nonmember proteins of each class is 99.9, 99.2, 99.6, 99.8, 99.9, 99.8, 98.5, 99.9, and 97.0%, respectively. Comparable accuracies are obtained when homologous proteins are considered as one, or by using a different SVM kernel function. Our method predicts 86.8% of the 76 lipid binding proteins nonhomologous to any protein in the Swiss-Prot database and 89.0% of the 73 known lipid binding domains as lipid binding. These findings suggest the usefulness of SVMs for facilitating the prediction of lipid binding proteins. Our software can be accessed at the SVMProt server (http://jing.cz3.nus.edu.sg/cgi-bin/svmprot.cgi).     

Prediction of subcellular localization using sequence-biased recurrent networks

MOTIVATION: Targeting peptides direct nascent proteins to their specific subcellular compartment. Knowledge of targeting signals enables informed drug design and reliable annotation of gene products. However, due to the low similarity of such sequences and the dynamical nature of the sorting process, the computational prediction of subcellular localization of proteins is challenging. RESULTS: We contrast the use of feed forward models as employed by the popular TargetP/SignalP predictors with a sequence-biased recurrent network model. The models are evaluated in terms of performance at the residue level and at the sequence level, and demonstrate that recurrent networks improve the overall prediction performance. Compared to the original results reported for TargetP, an ensemble of the tested models increases the accuracy by 6 and 5% on non-plant and plant data, respectively. AVAILABILITY: The Protein Prowler incorporating the recurrent network predictor described in this paper is available online at http://pprowler.imb.uq.edu.au/     

MITOPRED: a genome-scale method for prediction of nucleus-encoded mitochondrial proteins

MOTIVATION: Currently available methods for the prediction of subcellular location of mitochondrial proteins rely largely on the presence of mitochondrial targeting signals in the protein sequences. However, a large fraction of mitochondrial proteins lack such signals, making those tools ineffective for genome-scale prediction of mitochondria-targeted proteins. Here, we propose a method for genome-scale prediction of nucleus-encoded mitochondrial proteins. The new method, MITOPRED, is based on the Pfam domain occurrence patterns and the amino acid compositional differences between mitochondrial and non-mitochondrial proteins. RESULTS: MITOPRED could predict mitochondrial proteins with 100% specificity at a 44% sensitivity rate and with 67% specificity at 99% sensitivity. Additionally, it was sufficiently robust to predict mitochondrial proteins across different eukaryotic species with similar accuracy. Based on Matthews correlation coefficient measure, the prediction performance of MITOPRED is clearly superior (0.73) to those of the two popular methods TargetP (0.51) and PSORT (0.53). Using this method, we predicted the nucleus-encoded mitochondrial proteins from six complete genomes (three invertebrate, two vertebrate and one plant species) and estimated the total number in each genome. In human, our method estimated the existence of 1362 mitochondrial proteins corresponding to 4.8% of the total proteome. AVAILABILITY: MITOPRED program is freely accessible at http://mitopred.sdsc.edu. Source code is available on request from the authors. SUPPLEMENTARY INFORMATION: Training data sets are also available at http://mitopred.sdsc.edu