Improving consensus contact prediction via server correlation reduction

QAUST: Protein Function Prediction Using Structure Similarity, Protein Interaction, and Functional Motifs

The number of available protein sequences in public databases is increasing exponentially. However, a significant percentage of these sequences lack functional annotation, which is essential for the understanding of how biological systems operate. Here, we propose a novel method, Quantitative Annotation of Unknown STructure (QAUST), to infer protein functions, specifically Gene Ontology (GO) terms and Enzyme Commission (EC) numbers. QAUST uses three sources of information: structure information encoded by global and local structure similarity search, biological network information inferred by protein–protein interaction data, and sequence information extracted from functionally discriminative sequence motifs. These three pieces of information are combined by consensus averaging to make the final prediction. Our approach has been tested on 500 protein targets from the Critical Assessment of Functional Annotation (CAFA) benchmark set. The results show that our method provides accurate functional annotation and outperforms other prediction methods based on sequence similarity search or threading. We further demonstrate that a previously unknown function of human tripartite motif-containing 22 (TRIM22) protein predicted by QAUST can be experimentally validated.

Contact prediction is hardest for the most informative contacts, but improves with the incorporation of contact potentials

Co-evolution between pairs of residues in a multiple sequence alignment (MSA) of homologous proteins has long been proposed as an indicator of structural contacts. Recently, several methods, such as direct-coupling analysis (DCA) and MetaPSICOV, have been shown to achieve impressive rates of contact prediction by taking advantage of considerable sequence data. In this paper, we show that prediction success rates are highly sensitive to the structural definition of a contact, with more permissive definitions (i.e., those classifying more pairs as true contacts) naturally leading to higher positive predictive rates, but at the expense of the amount of structural information contributed by each contact. Thus, the remaining limitations of contact prediction algorithms are most noticeable in conjunction with geometrically restrictive contacts—precisely those that contribute more information in structure prediction. We suggest that to improve prediction rates for such “informative” contacts one could combine co-evolution scores with additional indicators of contact likelihood. Specifically, we find that when a pair of co-varying positions in an MSA is occupied by residue pairs with favorable statistical contact energies, that pair is more likely to represent a true contact. We show that combining a contact potential metric with DCA or MetaPSICOV performs considerably better than DCA or MetaPSICOV alone, respectively. This is true regardless of contact definition, but especially true for stricter and more informative contact definitions. In summary, this work outlines some remaining challenges to be addressed in contact prediction and proposes and validates a promising direction towards improvement.

NeBcon: protein contact map prediction using neural network training coupled with naïve Bayes classifiers

Motivation

Recent CASP experiments have witnessed exciting progress on folding large-size non-humongous proteins with the assistance of co-evolution based contact predictions. The success is however anecdotal due to the requirement of the contact prediction methods for the high volume of sequence homologs that are not available to most of the non-humongous protein targets. Development of efficient methods that can generate balanced and reliable contact maps for different type of protein targets is essential to enhance the success rate of the ab initio protein structure prediction.

Results

We developed a new pipeline, NeBcon, which uses the naïve Bayes classifier (NBC) theorem to combine eight state of the art contact methods that are built from co-evolution and machine learning approaches. The posterior probabilities of the NBC model are then trained with intrinsic structural features through neural network learning for the final contact map prediction. NeBcon was tested on 98 non-redundant proteins, which improves the accuracy of the best co-evolution based meta-server predictor by 22%; the magnitude of the improvement increases to 45% for the hard targets that lack sequence and structural homologs in the databases. Detailed data analysis showed that the major contribution to the improvement is due to the optimized NBC combination of the complementary information from both co-evolution and machine learning predictions. The neural network training also helps to improve the coupling of the NBC posterior probability and the intrinsic structural features, which were found particularly important for the proteins that do not have sufficient number of homologous sequences to derive reliable co-evolution profiles.

Availiablity and Implementation

On-line server and standalone package of the program are available at http://zhanglab.ccmb.med.umich.edu/NeBcon/ .

Contact

zhng@umich.edu.

Supplementary information

Supplementary data are available at Bioinformatics online.

A Sequence-Based Dynamic Ensemble Learning System for Protein Ligand-Binding Site Prediction

BACKGROUND

Proteins have the fundamental ability to selectively bind to other molecules and perform specific functions through such interactions, such as protein-ligand binding. Accurate prediction of protein residues that physically bind to ligands is important for drug design and protein docking studies. Most of the successful protein-ligand binding predictions were based on known structures. However, structural information is not largely available in practice due to the huge gap between the number of known protein sequences and that of experimentally solved structures.

RESULTS

This paper proposes a dynamic ensemble approach to identify protein-ligand binding residues by using sequence information only. To avoid problems resulting from highly imbalanced samples between the ligand-binding sites and non ligand-binding sites, we constructed several balanced data sets and we trained a random forest classifier for each of them. We dynamically selected a subset of classifiers according to the similarity between the target protein and the proteins in the training data set. The combination of the predictions of the classifier subset to each query protein target yielded the final predictions. The ensemble of these classifiers formed a sequence-based predictor to identify protein-ligand binding sites.

CONCLUSIONS

Experimental results on two Critical Assessment of protein Structure Prediction datasets and the ccPDB dataset demonstrated that of our proposed method compared favorably with the state-of-the-art.

AVAILABILITY

http://www2.ahu.edu.cn/pchen/web/LigandDSES.htm.

CMsearch: simultaneous exploration of protein sequence space and structure space improves not only protein homology detection but also protein structure prediction

MOTIVATION

Protein homology detection, a fundamental problem in computational biology, is an indispensable step toward predicting protein structures and understanding protein functions. Despite the advances in recent decades on sequence alignment, threading and alignment-free methods, protein homology detection remains a challenging open problem. Recently, network methods that try to find transitive paths in the protein structure space demonstrate the importance of incorporating network information of the structure space. Yet, current methods merge the sequence space and the structure space into a single space, and thus introduce inconsistency in combining different sources of information.

METHOD

We present a novel network-based protein homology detection method, CMsearch, based on cross-modal learning. Instead of exploring a single network built from the mixture of sequence and structure space information, CMsearch builds two separate networks to represent the sequence space and the structure space. It then learns sequence-structure correlation by simultaneously taking sequence information, structure information, sequence space information and structure space information into consideration.

RESULTS

We tested CMsearch on two challenging tasks, protein homology detection and protein structure prediction, by querying all 8332 PDB40 proteins. Our results demonstrate that CMsearch is insensitive to the similarity metrics used to define the sequence and the structure spaces. By using HMM-HMM alignment as the sequence similarity metric, CMsearch clearly outperforms state-of-the-art homology detection methods and the CASP-winning template-based protein structure prediction methods.

AVAILABILITY AND IMPLEMENTATION

Our program is freely available for download from http://sfb.kaust.edu.sa/Pages/Software.aspx

CONTACT

: xin.gao@kaust.edu.sa

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

LigandRFs: random forest ensemble to identify ligand-binding residues from sequence information alone

Background

Protein-ligand binding is important for some proteins to perform their functions. Protein-ligand binding sites are the residues of proteins that physically bind to ligands. Despite of the recent advances in computational prediction for protein-ligand binding sites, the state-of-the-art methods search for similar, known structures of the query and predict the binding sites based on the solved structures. However, such structural information is not commonly available.

Results

In this paper, we propose a sequence-based approach to identify protein-ligand binding residues. We propose a combination technique to reduce the effects of different sliding residue windows in the process of encoding input feature vectors. Moreover, due to the highly imbalanced samples between the ligand-binding sites and non ligand-binding sites, we construct several balanced data sets, for each of which a random forest (RF)-based classifier is trained. The ensemble of these RF classifiers forms a sequence-based protein-ligand binding site predictor.

Conclusions

Experimental results on CASP9 and CASP8 data sets demonstrate that our method compares favorably with the state-of-the-art protein-ligand binding site prediction methods.

Protein structural model selection by combining consensus and single scoring methods

Quality assessment (QA) for predicted protein structural models is an important and challenging research problem in protein structure prediction. Consensus Global Distance Test (CGDT) methods assess each decoy (predicted structural model) based on its structural similarity to all others in a decoy set and has been proved to work well when good decoys are in a majority cluster. Scoring functions evaluate each single decoy based on its structural properties. Both methods have their merits and limitations. In this paper, we present a novel method called PWCom, which consists of two neural networks sequentially to combine CGDT and single model scoring methods such as RW, DDFire and OPUS-Ca. Specifically, for every pair of decoys, the difference of the corresponding feature vectors is input to the first neural network which enables one to predict whether the decoy-pair are significantly different in terms of their GDT scores to the native. If yes, the second neural network is used to decide which one of the two is closer to the native structure. The quality score for each decoy in the pool is based on the number of winning times during the pairwise comparisons. Test results on three benchmark datasets from different model generation methods showed that PWCom significantly improves over consensus GDT and single scoring methods. The QA server (MUFOLD-Server) applying this method in CASP 10 QA category was ranked the second place in terms of Pearson and Spearman correlation performance.

eThread: a highly optimized machine learning-based approach to meta-threading and the modeling of protein tertiary structures

Template-based modeling that employs various meta-threading techniques is currently the most accurate, and consequently the most commonly used, approach for protein structure prediction. Despite the evident progress in this field, accurate structure models cannot be constructed for a significant fraction of gene products, thus the development of new algorithms is required. Here, we describe the development, optimization and large-scale benchmarking of eThread, a highly accurate meta-threading procedure for the identification of structural templates and the construction of corresponding target-to-template alignments. eThread integrates ten state-of-the-art threading/fold recognition algorithms in a local environment and extensively uses various machine learning techniques to carry out fully automated template-based protein structure modeling. Tertiary structure prediction employs two protocols based on widely used modeling algorithms: Modeller and TASSER-Lite. As a part of eThread, we also developed eContact, which is a Bayesian classifier for the prediction of inter-residue contacts and eRank, which effectively ranks generated multiple protein models and provides reliable confidence estimates as structure quality assessment. Excluding closely related templates from the modeling process, eThread generates models, which are correct at the fold level, for >80% of the targets; 40–50% of the constructed models are of a very high quality, which would be considered accurate at the family level. Furthermore, in large-scale benchmarking, we compare the performance of eThread to several alternative methods commonly used in protein structure prediction. Finally, we estimate the upper bound for this type of approach and discuss the directions towards further improvements.

Sampling multiple scoring functions can improve protein loop structure prediction accuracy

Accurately predicting loop structures is important for understanding functions of many proteins. In order to obtain loop models with high accuracy, efficiently sampling the loop conformation space to discover reasonable structures is a critical step. In loop conformation sampling, coarse-grain energy (scoring) functions coupling with reduced protein representations are often used to reduce the number of degrees of freedom as well as sampling computational time. However, due to implicitly considering many factors by reduced representations, the coarse-grain scoring functions may have potential insensitivity and inaccuracy, which can mislead the sampling process and consequently ignore important loop conformations. In this paper, we present a new computational sampling approach to obtain reasonable loop backbone models, so-called the Pareto optimal sampling (POS) method. The rationale of the POS method is to sample the function space of multiple, carefully selected scoring functions to discover an ensemble of diversified structures yielding Pareto optimality to all sampled conformations. The POS method can efficiently tolerate insensitivity and inaccuracy in individual scoring functions and thereby lead to significant accuracy improvement in loop structure prediction. We apply the POS method to a set of 4-12-residue loop targets using a function space composed of backbone-only Rosetta and distance-scale finite ideal-gas reference (DFIRE) and a triplet backbone dihedral potential developed in our lab. Our computational results show that in 501 out of 502 targets, the model sets generated by POS contain structure models are within subangstrom resolution. Moreover, the top-ranked models have a root mean square deviation (rmsd) less than 1 A in 96.8, 84.1, and 72.2% of the short (4-6 residues), medium (7-9 residues), and long (10-12 residues) targets, respectively, when the all-atom models are generated by local optimization from the backbone models and are ranked by our recently developed Pareto optimal consensus (POC) method. Similar sampling effectiveness can also be found in a set of 13-residue loop targets.

Multiple interests in structural models of DARC transmembrane protein

Duffy Antigen Receptor for Chemokines (DARC) is an unusual transmembrane chemokine receptor which (i) binds the two main chemokine families and (ii) does not transduct any signal as it lacks the DRY consensus sequence. It is considered as silent chemokine receptor, a tank useful for chemiotactism. DARC had been particularly studied as a major actor of malaria infection by Plasmodium vivax. It is also implicated in multiple chemokine inflammation, inflammatory diseases, in cancer and might play a role in HIV infection and AIDS. In this review, we focus on the interest to build structural model of DARC to understand more precisely its abilities to bind its physiological ligand CXCL8 and its malaria ligand. We also present innovative development on VHHs able to bind DARC protein. We underline difficulties and limitations of such bioinformatics approaches and highlight the crucial importance of biological data to conduct these kinds of researches.

Improving predicted protein loop structure ranking using a Pareto-optimality consensus method

Background

Accurate protein loop structure models are important to understand functions of many proteins. Identifying the native or near-native models by distinguishing them from the misfolded ones is a critical step in protein loop structure prediction.

Results

We have developed a Pareto Optimal Consensus (POC) method, which is a consensus model ranking approach to integrate multiple knowledge- or physics-based scoring functions. The procedure of identifying the models of best quality in a model set includes: 1) identifying the models at the Pareto optimal front with respect to a set of scoring functions, and 2) ranking them based on the fuzzy dominance relationship to the rest of the models. We apply the POC method to a large number of decoy sets for loops of 4- to 12-residue in length using a functional space composed of several carefully-selected scoring functions: Rosetta, DOPE, DDFIRE, OPLS-AA, and a triplet backbone dihedral potential developed in our lab. Our computational results show that the sets of Pareto-optimal decoys, which are typically composed of ~20% or less of the overall decoys in a set, have a good coverage of the best or near-best decoys in more than 99% of the loop targets. Compared to the individual scoring function yielding best selection accuracy in the decoy sets, the POC method yields 23%, 37%, and 64% less false positives in distinguishing the native conformation, indentifying a near-native model (RMSD < 0.5A from the native) as top-ranked, and selecting at least one near-native model in the top-5-ranked models, respectively. Similar effectiveness of the POC method is also found in the decoy sets from membrane protein loops. Furthermore, the POC method outperforms the other popularly-used consensus strategies in model ranking, such as rank-by-number, rank-by-rank, rank-by-vote, and regression-based methods.

Conclusions

By integrating multiple knowledge- and physics-based scoring functions based on Pareto optimality and fuzzy dominance, the POC method is effective in distinguishing the best loop models from the other ones within a loop model set.

Distant homologs of anti-apoptotic factor HAX1 encode parvalbumin-like calcium binding proteins

Background

Apoptosis is a highly ordered and orchestrated multiphase process controlled by the numerous cellular and extra-cellular signals, which executes the programmed cell death via release of cytochrome c alterations in calcium signaling, caspase-dependent limited proteolysis and DNA fragmentation. Besides the general modifiers of apoptosis, several tissue-specific regulators of this process were identified including HAX1 (HS-1 associated protein X-1) - an anti-apoptotic factor active in myeloid cells. Although HAX1 was the subject of various experimental studies, the mechanisms of its action and a functional link connected with the regulation of apoptosis still remains highly speculative.

Findings

Here we provide the data which suggests that HAX1 may act as a regulator or as a sensor of calcium. On the basis of iterative similarity searches, we identified a set of distant homologs of HAX1 in insects. The applied fold recognition protocol gives us strong evidence that the distant insects' homologs of HAX1 are novel parvalbumin-like calcium binding proteins. Although the whole three EF-hands fold is not preserved in vertebrate our analysis suggests that there is an existence of a potential single EF-hand calcium binding site in HAX1. The molecular mechanism of its action remains to be identified, but the risen hypothesis easily translates into previously reported lines of various data on the HAX1 biology as well as, provides us a direct link to the regulation of apoptosis. Moreover, we also report that other family of myeloid specific apoptosis regulators - myeloid leukemia factors (MLF1, MLF2) share the homologous C-terminal domain and taxonomic distribution with HAX1.

Conclusions

Performed structural and active sites analyses gave new insights into mechanisms of HAX1 and MLF families in apoptosis process and suggested possible role of HAX1 in calcium-binding, still the analyses require further experimental verification.