A high-accuracy protein structural class prediction algorithm using predicted secondary structural information

One major problem with the existing algorithm for the prediction of protein structural classes is low accuracies for proteins from α/β and α+β classes. In this study, three novel features were rationally designed to model the differences between proteins from these two classes. In combination with other rational designed features, an 11-dimensional vector prediction method was proposed. By means of this method, the overall prediction accuracy based on 25PDB dataset was 1.5% higher than the previous best-performing method, MODAS. Furthermore, the prediction accuracy for proteins from α+β class based on 25PDB dataset was 5% higher than the previous best-performing method, SCPRED. The prediction accuracies obtained with the D675 and FC699 datasets were also improved.     

An optimization approach to predicting protein structural class from amino acid composition.

Proteins are generally classified into four structural classes: all-alpha proteins, all-beta proteins, alpha + beta proteins, and alpha/beta proteins. In this article, a protein is expressed as a vector of 20-dimensional space, in which its 20 components are defined by the composition of its 20 amino acids. Based on this, a new method, the so-called maximum component coefficient method, is proposed for predicting the structural class of a protein according to its amino acid composition. In comparison with the existing methods, the new method yields a higher general accuracy of prediction. Especially for the all-alpha proteins, the rate of correct prediction obtained by the new method is much higher than that by any of the existing methods. For instance, for the 19 all-alpha proteins investigated previously by P.Y. Chou, the rate of correct prediction by means of his method was 84.2%, but the correct rate when predicted with the new method would be 100%! Furthermore, the new method is characterized by an explicable physical picture. This is reflected by the process in which the vector representing a protein to be predicted is decomposed into four component vectors, each of which corresponds to one of the norms of the four protein structural classes.     

Structure based function prediction of proteins using fragment library frequency vectors

The function of the protein is primarily dictated by its structure. Therefore it is far more logical to find the functional clues of the protein in its overall 3-dimensional fold or its global structure. In this paper, we have developed a novel Support Vector Machines (SVM) based prediction model for functional classification and prediction of proteins using features extracted from its global structure based on fragment libraries. Fragment libraries have been previously used for abintio modelling of proteins and protein structure comparisons. The query protein structure is broken down into a collection of short contiguous backbone fragments and this collection is discretized using a library of fragments. The input feature vector is frequency vector that counts the number of each library fragment in the collection of fragments by all-to-all fragment comparisons. SVM models were trained and optimised for obtaining the best 10-fold Cross validation accuracy for classification. As an example, this method was applied for prediction and classification of Cell Adhesion molecules (CAMs). Thirty-four different fragment libraries with sizes ranging from 4 to 400 and fragment lengths ranging from 4 to 12 were used for obtaining the best prediction model. The best 10-fold CV accuracy of 95.25% was obtained for library of 400 fragments of length 10. An accuracy of 87.5% was obtained on an unseen test dataset consisting of 20 CAMs and 20 NonCAMs. This shows that protein structure can be accurately and uniquely described using 400 representative fragments of length 10.     

Prediction of helix-helix contacts and interacting helices in polytopic membrane proteins using neural networks

Despite rapidly increasing numbers of available 3D structures, membrane proteins still account for less than 1% of all structures in the Protein Data Bank. Recent high-resolution structures indicate a clearly broader structural diversity of membrane proteins than initially anticipated, motivating the development of reliable structure prediction methods specifically tailored for this class of molecules. One important prediction target capturing all major aspects of a protein's 3D structure is its contact map. Our analysis shows that computational methods trained to predict residue contacts in globular proteins perform poorly when applied to membrane proteins. We have recently published a method to identify interacting alpha-helices in membrane proteins based on the analysis of coevolving residues in predicted transmembrane regions. Here, we present a substantially improved algorithm for the same problem, which uses a newly developed neural network approach to predict helix-helix contacts. In addition to the input features commonly used for contact prediction of soluble proteins, such as windowed residue profiles and residue distance in the sequence, our network also incorporates features that apply to membrane proteins only, such as residue position within the transmembrane segment and its orientation toward the lipophilic environment. The obtained neural network can predict contacts between residues in transmembrane segments with nearly 26% accuracy. It is therefore the first published contact predictor developed specifically for membrane proteins performing with equal accuracy to state-of-the-art contact predictors available for soluble proteins. The predicted helix-helix contacts were employed in a second step to identify interacting helices. For our dataset consisting of 62 membrane proteins of solved structure, we gained an accuracy of 78.1%. Because the reliable prediction of helix interaction patterns is an important step in the classification and prediction of membrane protein folds, our method will be a helpful tool in compiling a structural census of membrane proteins.     

Accurate prediction of protein structural classes using functional domains and predicted secondary structure sequences

Protein structural class prediction is one of the challenging problems in bioinformatics. Previous methods directly based on the similarity of amino acid (AA) sequences have been shown to be insufficient for low-similarity protein data-sets. To improve the prediction accuracy for such low-similarity proteins, different methods have been recently proposed that explore the novel feature sets based on predicted secondary structure propensities. In this paper, we focus on protein structural class prediction using combinations of the novel features including secondary structure propensities as well as functional domain (FD) features extracted from the InterPro signature database. Our comprehensive experimental results based on several benchmark data-sets have shown that the integration of new FD features substantially improves the accuracy of structural class prediction for low-similarity proteins as they capture meaningful relationships among AA residues that are far away in protein sequence. The proposed prediction method has also been tested to predict structural classes for partially disordered proteins with the reasonable prediction accuracy, which is a more difficult problem comparing to structural class prediction for commonly used benchmark data-sets and has never been done before to the best of our knowledge. In addition, to avoid overfitting with a large number of features, feature selection is applied to select discriminating features that contribute to achieve high prediction accuracy. The selected features have been shown to achieve stable prediction performance across different benchmark data-sets.     

Accurate prediction of protein structural classes using functional domains and predicted secondary structure sequences

Protein structural class prediction is one of the challenging problems in bioinformatics. Previous methods directly based on the similarity of amino acid (AA) sequences have been shown to be insufficient for low-similarity protein data-sets. To improve the prediction accuracy for such low-similarity proteins, different methods have been recently proposed that explore the novel feature sets based on predicted secondary structure propensities. In this paper, we focus on protein structural class prediction using combinations of the novel features including secondary structure propensities as well as functional domain (FD) features extracted from the InterPro signature database. Our comprehensive experimental results based on several benchmark data-sets have shown that the integration of new FD features substantially improves the accuracy of structural class prediction for low-similarity proteins as they capture meaningful relationships among AA residues that are far away in protein sequence. The proposed prediction method has also been tested to predict structural classes for partially disordered proteins with the reasonable prediction accuracy, which is a more difficult problem comparing to structural class prediction for commonly used benchmark data-sets and has never been done before to the best of our knowledge. In addition, to avoid overfitting with a large number of features, feature selection is applied to select discriminating features that contribute to achieve high prediction accuracy. The selected features have been shown to achieve stable prediction performance across different benchmark data-sets.     

Enhanced recognition of protein transmembrane domains with prediction-based structural profiles

MOTIVATION: Membrane domain prediction has recently been re-evaluated by several groups, suggesting that the accuracy of existing methods is still rather limited. In this work, we revisit this problem and propose novel methods for prediction of alpha-helical as well as beta-sheet transmembrane (TM) domains. The new approach is based on a compact representation of an amino acid residue and its environment, which consists of predicted solvent accessibility and secondary structure of each amino acid. A recently introduced method for solvent accessibility prediction trained on a set of soluble proteins is used here to indicate segments of residues that are predicted not to be accessible to water and, therefore, may be 'buried' in the membrane. While evolutionary profiles in the form of a multiple alignment are used to derive these simple 'structural profiles', they are not used explicitly for the membrane domain prediction and the overall number of parameters in the model is significantly reduced. This offers the possibility of a more reliable estimation of the free parameters in the model with a limited number of experimentally resolved membrane protein structures. RESULTS: Using cross-validated training on available sets of structurally resolved and non-redundant alpha and beta membrane proteins, we demonstrate that membrane domain prediction methods based on such a compact representation outperform approaches that utilize explicitly evolutionary profiles and multiple alignments. Moreover, using an external evaluation by the TMH Benchmark server we show that our final prediction protocol for the TM helix prediction is competitive with the state-of-the-art methods, achieving per-residue accuracy of approximately 89% and per-segment accuracy of approximately 80% on the set of high resolution structures used by the TMH Benchmark server. At the same time the observed rates of confusion with signal peptides and globular proteins are the lowest among the tested methods. The new method is available online at http://minnou.cchmc.org.     

RSARF: prediction of residue solvent accessibility from protein sequence using random forest method

Prediction of protein structure from its amino acid sequence is still a challenging problem. The complete physicochemical understanding of protein folding is essential for the accurate structure prediction. Knowledge of residue solvent accessibility gives useful insights into protein structure prediction and function prediction. In this work, we propose a random forest method, RSARF, to predict residue accessible surface area from protein sequence information. The training and testing was performed using 120 proteins containing 22006 residues. For each residue, buried and exposed state was computed using five thresholds (0%, 5%, 10%, 25%, and 50%). The prediction accuracy for 0%, 5%, 10%, 25%, and 50% thresholds are 72.9%, 78.25%, 78.12%, 77.57% and 72.07% respectively. Further, comparison of RSARF with other methods using a benchmark dataset containing 20 proteins shows that our approach is useful for prediction of residue solvent accessibility from protein sequence without using structural information. The RSARF program, datasets and supplementary data are available at http://caps.ncbs.res.in/download/pugal/RSARF/.     

Analysis of protein features and machine learning algorithms for prediction of druggable proteins

Background: Computational tools have been widely used in drug discovery process since they reduce the time and cost. Prediction of whether a protein is druggable is fundamental and crucial for drug research pipeline. Sequence based protein function prediction plays vital roles in many research areas. Training data, protein features selection and machine learning algorithms are three indispensable elements that drive the successfulness of the models. Methods: In this study, we tested the performance of different combinations of protein features and machine learning algorithms, based on FDA-approved small molecules’ targets, in druggable proteins prediction. We also enlarged the dataset to include the targets of small molecules that were in experiment or clinical investigation. Results: We found that although the 146-d vector used by Li et al. with neuron network achieved the best training accuracy of 91.10%, overlapped 3-gram word2vec with logistic regression achieved best prediction accuracy on independent test set (89.55%) and on newly approved-targets. Enlarged dataset with targets of small molecules in experiment and clinical investigation were trained. Unfortunately, the best training accuracy was only 75.48%. In addition, we applied our models to predict potential targets for references in future study. Conclusions: Our study indicates the potential ability of word2vec in the prediction of druggable protein. And the training dataset of druggable protein should not be extended to targets that are lack of verification. The target prediction package could be found on https://github.com/pkumdl/target_prediction.     

Multi-class protein fold recognition using support vector machines and neural networks

MOTIVATION: Protein fold recognition is an important approach to structure discovery without relying on sequence similarity. We study this approach with new multi-class classification methods and examined many issues important for a practical recognition system. RESULTS: Most current discriminative methods for protein fold prediction use the one-against-others method, which has the well-known 'False Positives' problem. We investigated two new methods: the unique one-against-others and the all-against-all methods. Both improve prediction accuracy by 14-110% on a dataset containing 27 SCOP folds. We used the Support Vector Machine (SVM) and the Neural Network (NN) learning methods as base classifiers. SVMs converges fast and leads to high accuracy. When scores of multiple parameter datasets are combined, majority voting reduces noise and increases recognition accuracy. We examined many issues involved with large number of classes, including dependencies of prediction accuracy on the number of folds and on the number of representatives in a fold. Overall, recognition systems achieve 56% fold prediction accuracy on a protein test dataset, where most of the proteins have below 25% sequence identity with the proteins used in training.     

Residue-level prediction of DNA-binding sites and its application on DNA-binding protein predictions

Protein-DNA interactions are crucial to many cellular activities such as expression-control and DNA-repair. These interactions between amino acids and nucleotides are highly specific and any aberrance at the binding site can render the interaction completely incompetent. In this study, we have three aims focusing on DNA-binding residues on the protein surface: to develop an automated approach for fast and reliable recognition of DNA-binding sites; to improve the prediction by distance-dependent refinement; use these predictions to identify DNA-binding proteins. We use a support vector machines (SVM)-based approach to harness the features of the DNA-binding residues to distinguish them from non-binding residues. Features used for distinction include the residue's identity, charge, solvent accessibility, average potential, the secondary structure it is embedded in, neighboring residues, and location in a cationic patch. These features collected from 50 proteins are used to train SVM. Testing is then performed on another set of 37 proteins, much larger than any testing set used in previous studies. The testing set has no more than 20% sequence identity not only among its pairs, but also with the proteins in the training set, thus removing any undesired redundancy due to homology. This set also has proteins with an unseen DNA-binding structural class not present in the training set. With the above features, an accuracy of 66% with balanced sensitivity and specificity is achieved without relying on homology or evolutionary information. We then develop a post-processing scheme to improve the prediction using the relative location of the predicted residues. Balanced success is then achieved with average sensitivity, specificity and accuracy pegged at 71.3%, 69.3% and 70.5%, respectively. Average net prediction is also around 70%. Finally, we show that the number of predicted DNA-binding residues can be used to differentiate DNA-binding proteins from non-DNA-binding proteins with an accuracy of 78%. Results presented here demonstrate that machine-learning can be applied to automated identification of DNA-binding residues and that the success rate can be ameliorated as more features are added. Such functional site prediction protocols can be useful in guiding consequent works such as site-directed mutagenesis and macromolecular docking.     

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.     

SVM-based method for protein structural class prediction using secondary structural content and structural information of amino acids

The knowledge collated from the known protein structures has revealed that the proteins are usually folded into the four structural classes: all-α, all-β, α/β and α + β. A number of methods have been proposed to predict the protein's structural class from its primary structure; however, it has been observed that these methods fail or perform poorly in the cases of distantly related sequences. In this paper, we propose a new method for protein structural class prediction using low homology (twilight-zone) protein sequences dataset. Since protein structural class prediction is a typical classification problem, we have developed a Support Vector Machine (SVM)-based method for protein structural class prediction that uses features derived from the predicted secondary structure and predicted burial information of amino acid residues. The examination of different individual as well as feature combinations revealed that the combination of secondary structural content, secondary structural and solvent accessibility state frequencies of amino acids gave rise to the best leave-one-out cross-validation accuracy of ~81% which is comparable to the best accuracy reported in the literature so far.     

Prediction of protein disorder on amino acid substitutions

Intrinsically disordered regions of proteins are known to have many functional roles in cell signaling and regulatory pathways. The altered expression of these proteins due to mutations is associated with various diseases. Currently, most of the available methods focus on predicting the disordered proteins or the disordered regions in a protein. On the other hand, methods developed for predicting protein disorder on mutation showed a poor performance with a maximum accuracy of 70%. Hence, in this work, we have developed a novel method to classify the disorder-related amino acid substitutions using amino acid properties, substitution matrices, and the effect of neighboring residues that showed an accuracy of 90.0% with a sensitivity and specificity of 94.9 and 80.6%, respectively, in 10-fold cross-validation. The method was evaluated with a test set of 20% data using 10 iterations, which showed an average accuracy of 88.9%. Furthermore, we systematically analyzed the features responsible for the better performance of our method and observed that neighboring residues play an important role in defining the disorder of a given residue in a protein sequence. We have developed a prediction server to identify disorder-related mutations, and it is available at http://www.iitm.ac.in/bioinfo/DIM_Pred/.     

Graph sharpening plus graph integration: a synergy that improves protein functional classification

MOTIVATION: Predicting protein function is a central problem in bioinformatics, and many approaches use partially or fully automated methods based on various combination of sequence, structure and other information on proteins or genes. Such information establishes relationships between proteins that can be modelled most naturally as edges in graphs. A priori, however, it is often unclear which edges from which graph may contribute most to accurate predictions. For that reason, one established strategy is to integrate all available sources, or graphs as in graph integration, in the hope that the positive signals will add to each other. However, in the problem of functional prediction, noise, i.e. the presence of inaccurate or false edges, can still be large enough that integration alone has little effect on prediction accuracy. In order to reduce noise levels and to improve integration efficiency, we present here a recent method in graph-based learning, graph sharpening, which provides a theoretically firm yet intuitive and practical approach for disconnecting undesirable edges from protein similarity graphs. This approach has several attractive features: it is quick, scalable in the number of proteins, robust with respect to errors and tolerant of very diverse types of protein similarity measures. RESULTS: We tested the classification accuracy in a test set of 599 proteins with remote sequence homology spread over 20 Gene Ontology (GO) functional classes. When compared to integration alone, graph sharpening plus integration of four vastly different molecular similarity measures improved the overall classification by nearly 30% [0.17 average increase in the area under the ROC curve (AUC)]. Moreover, and partially through the increased sparsity of the graphs induced by sharpening, this gain in accuracy came at negligible computational cost: sharpening and integration took on average 4.66 (+/-4.44) CPU seconds. AVAILABILITY: Software and Supplementary data will be available on http://mammoth.bcm.tmc.edu/     

Identification of Multi-Functional Enzyme with Multi-Label Classifier

Enzymes are important and effective biological catalyst proteins participating in almost all active cell processes. Identification of multi-functional enzymes is essential in understanding the function of enzymes. Machine learning methods perform better in protein structure and function prediction than traditional biological wet experiments. Thus, in this study, we explore an efficient and effective machine learning method to categorize enzymes according to their function. Multi-functional enzymes are predicted with a special machine learning strategy, namely, multi-label classifier. Sequence features are extracted from a position-specific scoring matrix with autocross-covariance transformation. Experiment results show that the proposed method obtains an accuracy rate of 94.1% in classifying six main functional classes through five cross-validation tests and outperforms state-of-the-art methods. In addition, 91.25% accuracy is achieved in multi-functional enzyme prediction, which is often ignored in other enzyme function prediction studies. The online prediction server and datasets can be accessed from the link http://server.malab.cn/MEC/.     

Accurate prediction of protein structural class

Because of the increasing gap between the data from sequencing and structural genomics, the accurate prediction of the structural class of a protein domain solely from the primary sequence has remained a challenging problem in structural biology. Traditional sequence-based predictors generally select several sequence features and then feed them directly into a classification program to identify the structural class. The current best sequence-based predictor achieved an overall accuracy of 74.1% when tested on a widely used, non-homologous benchmark dataset 25PDB. In the present work, we built a multiple linear regression (MLR) model to convert the 440-dimensional (440D) sequence feature vector extracted from the Position Specific Scoring Matrix (PSSM) of a protein domain to a 4-dimensinal (4D) structural feature vector, which could then be used to predict the four major structural classes. We performed 10-fold cross-validation and jackknife tests of the method on a large non-homologous dataset containing 8,244 domains distributed among the four major classes. The performance of our approach outperformed all of the existing sequence-based methods and had an overall accuracy of 83.1%, which is even higher than the results of those predicted secondary structure-based methods.     

Annotating proteins by mining protein interaction networks

MOTIVATION: In general, most accurate gene/protein annotations are provided by curators. Despite having lesser evidence strengths, it is inevitable to use computational methods for fast and a priori discovery of protein function annotations. This paper considers the problem of assigning Gene Ontology (GO) annotations to partially annotated or newly discovered proteins. RESULTS: We present a data mining technique that computes the probabilistic relationships between GO annotations of proteins on protein-protein interaction data, and assigns highly correlated GO terms of annotated proteins to non-annotated proteins in the target set. In comparison with other techniques, probabilistic suffix tree and correlation mining techniques produce the highest prediction accuracy of 81% precision with the recall at 45%. AVAILABILITY: Code is available upon request. Results and used materials are available online at http://kirac.case.edu/PROTAN.