Identification of RNA-binding sites in proteins by integrating various sequence information

Twenty years of advances in prediction of nucleic acid-binding residues in protein sequences

Computational prediction of nucleic acid-binding residues in protein sequences is an active field of research, with over 80 methods that were released in the past 2 decades. We identify and discuss 87 sequence-based predictors that include dozens of recently published methods that are surveyed for the first time. We overview historical progress and examine multiple practical issues that include availability and impact of predictors, key features of their predictive models, and important aspects related to their training and assessment. We observe that the past decade has brought increased use of deep neural networks and protein language models, which contributed to substantial gains in the predictive performance. We also highlight advancements in vital and challenging issues that include cross-predictions between deoxyribonucleic acid (DNA)-binding and ribonucleic acid (RNA)-binding residues and targeting the two distinct sources of binding annotations, structure-based versus intrinsic disorder-based. The methods trained on the structure-annotated interactions tend to perform poorly on the disorder-annotated binding and vice versa, with only a few methods that target and perform well across both annotation types. The cross-predictions are a significant problem, with some predictors of DNA-binding or RNA-binding residues indiscriminately predicting interactions with both nucleic acid types. Moreover, we show that methods with web servers are cited substantially more than tools without implementation or with no longer working implementations, motivating the development and long-term maintenance of the web servers. We close by discussing future research directions that aim to drive further progress in this area.

EPDRNA: A Model for Identifying DNA-RNA Binding Sites in Disease-Related Proteins

Protein-DNA and protein-RNA interactions are involved in many biological processes and regulate many cellular functions. Moreover, they are related to many human diseases. To understand the molecular mechanism of protein-DNA binding and protein-RNA binding, it is important to identify which residues in the protein sequence bind to DNA and RNA. At present, there are few methods for specifically identifying the binding sites of disease-related protein-DNA and protein-RNA. In this study, so we combined four machine learning algorithms into an ensemble classifier (EPDRNA) to predict DNA and RNA binding sites in disease-related proteins. The dataset used in model was collated from UniProt and PDB database, and PSSM, physicochemical properties and amino acid type were used as features. The EPDRNA adopted soft voting and achieved the best AUC value of 0.73 at the DNA binding sites, and the best AUC value of 0.71 at the RNA binding sites in 10-fold cross validation in the training sets. In order to further verify the performance of the model, we assessed EPDRNA for the prediction of DNA-binding sites and the prediction of RNA-binding sites on the independent test dataset. The EPDRNA achieved 85% recall rate and 25% precision on the protein-DNA interaction independent test set, and achieved 82% recall rate and 27% precision on the protein-RNA interaction independent test set. The online EPDRNA webserver is freely available at http://www.s-bioinformatics.cn/epdrna .

A comprehensive review of protein-centric predictors for biomolecular interactions: from proteins to nucleic acids and beyond

Proteins interact with diverse ligands to perform a large number of biological functions, such as gene expression and signal transduction. Accurate identification of these protein-ligand interactions is crucial to the understanding of molecular mechanisms and the development of new drugs. However, traditional biological experiments are time-consuming and expensive. With the development of high-throughput technologies, an increasing amount of protein data is available. In the past decades, many computational methods have been developed to predict protein-ligand interactions. Here, we review a comprehensive set of over 160 protein-ligand interaction predictors, which cover protein-protein, protein-nucleic acid, protein-peptide and protein-other ligands (nucleotide, heme, ion) interactions. We have carried out a comprehensive analysis of the above four types of predictors from several significant perspectives, including their inputs, feature profiles, models, availability, etc. The current methods primarily rely on protein sequences, especially utilizing evolutionary information. The significant improvement in predictions is attributed to deep learning methods. Additionally, sequence-based pretrained models and structure-based approaches are emerging as new trends.

PRIP: A Protein-RNA Interface Predictor Based on Semantics of Sequences

RNA–protein interactions play an indispensable role in many biological processes. Growing evidence has indicated that aberration of the RNA–protein interaction is associated with many serious human diseases. The precise and quick detection of RNA–protein interactions is crucial to finding new functions and to uncovering the mechanism of interactions. Although many methods have been presented to recognize RNA-binding sites, there is much room left for the improvement of predictive accuracy. We present a sequence semantics-based method (called PRIP) for predicting RNA-binding interfaces. The PRIP extracted semantic embedding by pre-training the Word2vec with the corpus. Extreme gradient boosting was employed to train a classifier. The PRIP obtained a SN of 0.73 over the five-fold cross validation and a SN of 0.67 over the independent test, outperforming the state-of-the-art methods. Compared with other methods, this PRIP learned the hidden relations between words in the context. The analysis of the semantics relationship implied that the semantics of some words were specific to RNA-binding interfaces. This method is helpful to explore the mechanism of RNA–protein interactions from a semantics point of view.

SNB-PSSM: A spatial neighbor-based PSSM used for protein-RNA binding site prediction

Protein-RNA interactions play essential roles in a wide variety of biological processes. Recognition of RNA-binding residues on proteins has been a challenging problem. Most of methods utilize the position-specific scoring matrix (PSSM). It has been found that considering the evolutionary information of sequence neighboring residues can improve the prediction. In this work, we introduce a novel method SNB-PSSM (spatial neighbor-based PSSM) combined with the structure window scheme where the evolutionary information of spatially neighboring residues is considered. The results show our method consistently outperforms the standard and smoothed PSSM methods. Tested on multiple datasets, this approach shows an encouraging performance compared with RNABindRPlus, BindN+, PPRInt, xypan, Predict_RBP, SpaPF, PRNA, and KYG, although is inferior to RNAProSite, RBscore, and aaRNA. In addition, since our method is not sensitive to protein structure changes, it can be applied well on binding site predictions of modeled structures. Thus, the result also suggests the evolution of binding sites is spatially cooperative. The proposed method as an effective tool of considering evolutionary information can be widely used for the nucleic acid-/protein-binding site prediction and functional motif finding.

DBP-PSSM: Combination of Evolutionary Profiles with the XGBoost Algorithm to Improve the Identification of DNA-binding Proteins

BACKGROUND AND OBJECTIVE

DNA-binding proteins play important roles in a variety of biological processes, such as gene transcription and regulation, DNA replication and repair, DNA recombination and packaging, and the formation of chromatin and ribosomes. Therefore, it is urgent to develop a computational method to improve the recognition efficiency of DNA-binding proteins.

METHODS

We proposed a novel method, DBP-PSSM, which constructed the features from amino acid composition and evolutionary information of protein sequences. The maximum relevance, minimum redundancy (mRMR) was employed to select the optimal features for establishing the XGBoost classifier, therefore, the novel model of prediction DNA-binding proteins, DBP-PSSM, was established with 5-fold cross-validation on the training dataset.

RESULTS

DBP-PSSM achieved an accuracy of 81.18% and MCC of 0.657 in a test dataset, which outperformed the many existing methods. These results demonstrated that our method can effectively predict DNA-binding proteins.

CONCLUSION

The data and source code are provided at https://github.com/784221489/DNA-binding.

Comprehensive Survey and Comparative Assessment of RNA-Binding Residue Predictions with Analysis by RNA Type

With close to 30 sequence-based predictors of RNA-binding residues (RBRs), this comparative survey aims to help with understanding and selection of the appropriate tools. We discuss past reviews on this topic, survey a comprehensive collection of predictors, and comparatively assess six representative methods. We provide a novel and well-designed benchmark dataset and we are the first to report and compare protein-level and datasets-level results, and to contextualize performance to specific types of RNAs. The methods considered here are well-cited and rely on machine learning algorithms on occasion combined with homology-based prediction. Empirical tests reveal that they provide relatively accurate predictions. Virtually all methods perform well for the proteins that interact with rRNAs, some generate accurate predictions for mRNAs, snRNA, SRP and IRES, while proteins that bind tRNAs are predicted poorly. Moreover, except for DRNApred, they confuse DNA and RNA-binding residues. None of the six methods consistently outperforms the others when tested on individual proteins. This variable and complementary protein-level performance suggests that users should not rely on applying just the single best dataset-level predictor. We recommend that future work should focus on the development of approaches that facilitate protein-level selection of accurate predictors and the consensus-based prediction of RBRs.

A Brief Survey of Machine Learning Methods in Identification of Mitochondria Proteins in Malaria Parasite

The number of human deaths caused by malaria is increasing day-by-day. In fact, the mitochondrial proteins of the malaria parasite play vital roles in the organism. For developing effective drugs and vaccines against infection, it is necessary to accurately identify mitochondrial proteins of the malaria parasite. Although precise details for the mitochondrial proteins can be provided by biochemical experiments, they are expensive and time-consuming. In this review, we summarized the machine learning-based methods for mitochondrial proteins identification in the malaria parasite and compared the construction strategies of these computational methods. Finally, we also discussed the future development of mitochondrial proteins recognition with algorithms.

Recent Advances in Machine Learning Based Prediction of RNA-protein Interactions

The interactions between RNAs and proteins play critical roles in many biological processes. Therefore, characterizing these interactions becomes critical for mechanistic, biomedical, and clinical studies. Many experimental methods can be used to determine RNA-protein interactions in multiple aspects. However, due to the facts that RNA-protein interactions are tissuespecific and condition-specific, as well as these interactions are weak and frequently compete with each other, those experimental techniques can not be made full use of to discover the complete spectrum of RNA-protein interactions. To moderate these issues, continuous efforts have been devoted to developing high quality computational techniques to study the interactions between RNAs and proteins. Many important progresses have been achieved with the application of novel techniques and strategies, such as machine learning techniques. Especially, with the development and application of CLIP techniques, more and more experimental data on RNA-protein interaction under specific biological conditions are available. These CLIP data altogether provide a rich source for developing advanced machine learning predictors. In this review, recent progresses on computational predictors for RNA-protein interaction were summarized in the following aspects: dataset, prediction strategies, and input features. Possible future developments were also discussed at the end of the review.

In Silico Prediction and Validation of Novel RNA Binding Proteins and Residues in the Human Proteome

Deciphering a complete landscape of protein-RNA interactions in the human proteome remains an elusive challenge. We computationally elucidate RNA binding proteins (RBPs) using an approach that complements previous efforts. We employ two modern complementary sequence-based methods that provide accurate predictions from the structured and the intrinsically disordered sequences, even in the absence of sequence similarity to the known RBPs. We generate and analyze putative RNA binding residues on the whole proteome scale. Using a conservative setting that ensures low, 5% false positive rate, we identify 1511 putative RBPs that include 281 known RBPs and 166 RBPs that were previously predicted. We empirically demonstrate that these overlaps are statistically significant. We also validate the putative RBPs based on two major hallmarks of their RNA binding residues: high levels of evolutionary conservation and enrichment in charged amino acids. Moreover, we show that the novel RBPs are significantly under-annotated functionally which coincides with the fact that they were not yet found to interact with RNAs. We provide two examples of our novel putative RBPs for which there is recent evidence of their interactions with RNAs. The dataset of novel putative RBPs and RNA binding residues for the future hypothesis generation is provided in the Supporting Information.

Comprehensive review and empirical analysis of hallmarks of DNA-, RNA- and protein-binding residues in protein chains

Proteins interact with a variety of molecules including proteins and nucleic acids. We review a comprehensive collection of over 50 studies that analyze and/or predict these interactions. While majority of these studies address either solely protein–DNA or protein–RNA binding, only a few have a wider scope that covers both protein–protein and protein–nucleic acid binding. Our analysis reveals that binding residues are typically characterized with three hallmarks: relative solvent accessibility (RSA), evolutionary conservation and propensity of amino acids (AAs) for binding. Motivated by drawbacks of the prior studies, we perform a large-scale analysis to quantify and contrast the three hallmarks for residues that bind DNA-, RNA-, protein- and (for the first time) multi-ligand-binding residues that interact with DNA and proteins, and with RNA and proteins. Results generated on a well-annotated data set of over 23 000 proteins show that conservation of binding residues is higher for nucleic acid- than protein-binding residues. Multi-ligand-binding residues are more conserved and have higher RSA than single-ligand-binding residues. We empirically show that each hallmark discriminates between binding and nonbinding residues, even predicted RSA, and that combining them improves discriminatory power for each of the five types of interactions. Linear scoring functions that combine these hallmarks offer good predictive performance of residue-level propensity for binding and provide intuitive interpretation of predictions. Better understanding of these residue-level interactions will facilitate development of methods that accurately predict binding in the exponentially growing databases of protein sequences.

RPI-Bind: a structure-based method for accurate identification of RNA-protein binding sites

RNA and protein interactions play crucial roles in multiple biological processes, while these interactions are significantly influenced by the structures and sequences of protein and RNA molecules. In this study, we first performed an analysis of RNA-protein interacting complexes, and identified interface properties of sequences and structures, which reveal the diverse nature of the binding sites. With the observations, we built a three-step prediction model, namely RPI-Bind, for the identification of RNA-protein binding regions using the sequences and structures of both proteins and RNAs. The three steps include 1) the prediction of RNA binding regions on protein, 2) the prediction of protein binding regions on RNA, and 3) the prediction of interacting regions on both RNA and protein simultaneously, with the results from steps 1) and 2). Compared with existing methods, most of which employ only sequences, our model significantly improves the prediction accuracy at each of the three steps. Especially, our model outperforms the catRAPID by >20% at the 3^rd step. All of these results indicate the importance of structures in RNA-protein interactions, and suggest that the RPI-Bind model is a powerful theoretical framework for studying RNA-protein interactions.

A combinatorial scoring function for protein-RNA docking

Protein-RNA docking is still an open question. One of the main challenges is to develop an effective scoring function that can discriminate near-native structures from the incorrect ones. To solve the problem, we have constructed a knowledge-based residue-nucleotide pairwise potential with secondary structure information considered for nonribosomal protein-RNA docking. Here we developed a weighted combined scoring function RpveScore that consists of the pairwise potential and six physics-based energy terms. The weights were optimized using the multiple linear regression method by fitting the scoring function to L_rmsd for the bound docking decoys from Benchmark II. The scoring functions were tested on 35 unbound docking cases. The results show that the scoring function RpveScore including all terms performs best. Also RpveScore was compared with the statistical mechanics-based method derived potential ITScore-PR, and the united atom-based statistical potentials QUASI-RNP and DARS-RNP. The success rate of RpveScore is 71.6% for the top 1000 structures and the number of cases where a near-native structure is ranked in top 30 is 25 out of 35 cases. For 32 systems (91.4%), RpveScore can find the binding mode in top 5 that has no lower than 50% native interface residues on protein and nucleotides on RNA. Additionally, it was found that the long-range electrostatic attractive energy plays an important role in distinguishing near-native structures from the incorrect ones. This work can be helpful for the development of protein-RNA docking methods and for the understanding of protein-RNA interactions. RpveScore program is available to the public at http://life.bjut.edu.cn/kxyj/kycg/2017116/14845362285362368_1.html Proteins 2017; 85:741-752. © 2016 Wiley Periodicals, Inc.

Prediction of redox-sensitive cysteines using sequential distance and other sequence-based features

Background

Reactive oxygen species can modify the structure and function of proteins and may also act as important signaling molecules in various cellular processes. Cysteine thiol groups of proteins are particularly susceptible to oxidation. Meanwhile, their reversible oxidation is of critical roles for redox regulation and signaling. Recently, several computational tools have been developed for predicting redox-sensitive cysteines; however, those methods either only focus on catalytic redox-sensitive cysteines in thiol oxidoreductases, or heavily depend on protein structural data, thus cannot be widely used.

Results

In this study, we analyzed various sequence-based features potentially related to cysteine redox-sensitivity, and identified three types of features for efficient computational prediction of redox-sensitive cysteines. These features are: sequential distance to the nearby cysteines, PSSM profile and predicted secondary structure of flanking residues. After further feature selection using SVM-RFE, we developed Redox-Sensitive Cysteine Predictor (RSCP), a SVM based classifier for redox-sensitive cysteine prediction using primary sequence only. Using 10-fold cross-validation on RSC758 dataset, the accuracy, sensitivity, specificity, MCC and AUC were estimated as 0.679, 0.602, 0.756, 0.362 and 0.727, respectively. When evaluated using 10-fold cross-validation with BALOSCTdb dataset which has structure information, the model achieved performance comparable to current structure-based method. Further validation using an independent dataset indicates it is robust and of relatively better accuracy for predicting redox-sensitive cysteines from non-enzyme proteins.

Conclusions

In this study, we developed a sequence-based classifier for predicting redox-sensitive cysteines. The major advantage of this method is that it does not rely on protein structure data, which ensures more extensive application compared to other current implementations. Accurate prediction of redox-sensitive cysteines not only enhances our understanding about the redox sensitivity of cysteine, it may also complement the proteomics approach and facilitate further experimental investigation of important redox-sensitive cysteines.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-016-1185-4) contains supplementary material, which is available to authorized users.

Accurate prediction of RNA-binding protein residues with two discriminative structural descriptors

BACKGROUND

RNA-binding proteins participate in many important biological processes concerning RNA-mediated gene regulation, and several computational methods have been recently developed to predict the protein-RNA interactions of RNA-binding proteins. Newly developed discriminative descriptors will help to improve the prediction accuracy of these prediction methods and provide further meaningful information for researchers.

RESULTS

In this work, we designed two structural features (residue electrostatic surface potential and triplet interface propensity) and according to the statistical and structural analysis of protein-RNA complexes, the two features were powerful for identifying RNA-binding protein residues. Using these two features and other excellent structure- and sequence-based features, a random forest classifier was constructed to predict RNA-binding residues. The area under the receiver operating characteristic curve (AUC) of five-fold cross-validation for our method on training set RBP195 was 0.900, and when applied to the test set RBP68, the prediction accuracy (ACC) was 0.868, and the F-score was 0.631.

CONCLUSIONS

The good prediction performance of our method revealed that the two newly designed descriptors could be discriminative for inferring protein residues interacting with RNAs. To facilitate the use of our method, a web-server called RNAProSite, which implements the proposed method, was constructed and is freely available at http://lilab.ecust.edu.cn/NABind .

A Large-Scale Assessment of Nucleic Acids Binding Site Prediction Programs

Computational prediction of nucleic acid binding sites in proteins are necessary to disentangle functional mechanisms in most biological processes and to explore the binding mechanisms. Several strategies have been proposed, but the state-of-the-art approaches display a great diversity in i) the definition of nucleic acid binding sites; ii) the training and test datasets; iii) the algorithmic methods for the prediction strategies; iv) the performance measures and v) the distribution and availability of the prediction programs. Here we report a large-scale assessment of 19 web servers and 3 stand-alone programs on 41 datasets including more than 5000 proteins derived from 3D structures of protein-nucleic acid complexes. Well-defined binary assessment criteria (specificity, sensitivity, precision, accuracy…) are applied. We found that i) the tools have been greatly improved over the years; ii) some of the approaches suffer from theoretical defects and there is still room for sorting out the essential mechanisms of binding; iii) RNA binding and DNA binding appear to follow similar driving forces and iv) dataset bias may exist in some methods.

DSPMP: Discriminating secretory proteins of malaria parasite by hybridizing different descriptors of Chou's pseudo amino acid patterns

Identification of the proteins secreted by the malaria parasite is important for developing effective drugs and vaccines against infection. Therefore, we developed an improved predictor called "DSPMP" (Discriminating Secretory Proteins of Malaria Parasite) to identify the secretory proteins of the malaria parasite by integrating several vector features using support vector machine-based methods. DSPMP achieved an overall predictive accuracy of 98.61%, which is superior to that of the existing predictors in this field. We show that our method is capable of identifying the secretory proteins of the malaria parasite and found that the amino acid composition for buried and exposed sequences, denoted by AAC(b/e), was the most important feature for constructing the predictor. This article not only introduces a novel method for detecting the important features of sample proteins related to the malaria parasite but also provides a useful tool for tackling general protein-related problems. The DSPMP webserver is freely available at http://202.207.14.87:8032/fuwu/DSPMP/index.asp.

Different motif requirements for the localization zipcode element of β-actin mRNA binding by HuD and ZBP1

Interactions of RNA-binding proteins (RBPs) with their target transcripts are essential for regulating gene expression at the posttranscriptional level including mRNA export/localization, stability, and translation. ZBP1 and HuD are RBPs that play pivotal roles in mRNA transport and local translational control in neuronal processes. While HuD possesses three RNA recognition motifs (RRMs), ZBP1 contains two RRMs and four K homology (KH) domains that either increase target specificity or provide a multi-target binding capability. Here we used isolated cis-element sequences of the target mRNA to examine directly protein-RNA interactions in cell-free systems. We found that both ZBP1 and HuD bind the zipcode element in rat β-actin mRNA's 3' UTR. Differences between HuD and ZBP1 were observed in their binding preference to the element. HuD showed a binding preference for U-rich sequence. In contrast, ZBP1 binding to the zipcode RNA depended more on the structural level, as it required the proper spatial organization of a stem-loop that is mainly determined by the U-rich element juxtaposed to the 3' end of a 5'-ACACCC-3' motif. On the basis of this work, we propose that ZBP1 and HuD bind to overlapping sites in the β-actin zipcode, but they recognize different features of this target sequence.

RNABindRPlus: a predictor that combines machine learning and sequence homology-based methods to improve the reliability of predicted RNA-binding residues in proteins

Protein-RNA interactions are central to essential cellular processes such as protein synthesis and regulation of gene expression and play roles in human infectious and genetic diseases. Reliable identification of protein-RNA interfaces is critical for understanding the structural bases and functional implications of such interactions and for developing effective approaches to rational drug design. Sequence-based computational methods offer a viable, cost-effective way to identify putative RNA-binding residues in RNA-binding proteins. Here we report two novel approaches: (i) HomPRIP, a sequence homology-based method for predicting RNA-binding sites in proteins; (ii) RNABindRPlus, a new method that combines predictions from HomPRIP with those from an optimized Support Vector Machine (SVM) classifier trained on a benchmark dataset of 198 RNA-binding proteins. Although highly reliable, HomPRIP cannot make predictions for the unaligned parts of query proteins and its coverage is limited by the availability of close sequence homologs of the query protein with experimentally determined RNA-binding sites. RNABindRPlus overcomes these limitations. We compared the performance of HomPRIP and RNABindRPlus with that of several state-of-the-art predictors on two test sets, RB44 and RB111. On a subset of proteins for which homologs with experimentally determined interfaces could be reliably identified, HomPRIP outperformed all other methods achieving an MCC of 0.63 on RB44 and 0.83 on RB111. RNABindRPlus was able to predict RNA-binding residues of all proteins in both test sets, achieving an MCC of 0.55 and 0.37, respectively, and outperforming all other methods, including those that make use of structure-derived features of proteins. More importantly, RNABindRPlus outperforms all other methods for any choice of tradeoff between precision and recall. An important advantage of both HomPRIP and RNABindRPlus is that they rely on readily available sequence and sequence-derived features of RNA-binding proteins. A webserver implementation of both methods is freely available at http://einstein.cs.iastate.edu/RNABindRPlus/.

Protein-specific prediction of mRNA binding using RNA sequences, binding motifs and predicted secondary structures

Background

RNA-binding proteins interact with specific RNA molecules to regulate important cellular processes. It is therefore necessary to identify the RNA interaction partners in order to understand the precise functions of such proteins. Protein-RNA interactions are typically characterized using in vivo and in vitro experiments but these may not detect all binding partners. Therefore, computational methods that capture the protein-dependent nature of such binding interactions could help to predict potential binding partners in silico.

Results

We have developed three methods to predict whether an RNA can interact with a particular RNA-binding protein using support vector machines and different features based on the sequence (the Oli method), the motif score (the OliMo method) and the secondary structure (the OliMoSS method). We applied these approaches to different experimentally-derived datasets and compared the predictions with RNAcontext and RPISeq. Oli outperformed OliMoSS and RPISeq, confirming our protein-specific predictions and suggesting that tetranucleotide frequencies are appropriate discriminative features. Oli and RNAcontext were the most competitive methods in terms of the area under curve. A precision-recall curve analysis achieved higher precision values for Oli. On a second experimental dataset including real negative binding information, Oli outperformed RNAcontext with a precision of 0.73 vs. 0.59.

Conclusions

Our experiments showed that features based on primary sequence information are sufficiently discriminating to predict specific RNA-protein interactions. Sequence motifs and secondary structure information were not necessary to improve these predictions. Finally we confirmed that protein-specific experimental data concerning RNA-protein interactions are valuable sources of information that can be used for the efficient training of models for in silico predictions. The scripts are available upon request to the corresponding author.

Annotating the protein-RNA interaction sites in proteins using evolutionary information and protein backbone structure

RNA-protein interactions play important roles in various biological processes. The precise detection of RNA-protein interaction sites is very important for understanding essential biological processes and annotating the function of the proteins. In this study, based on various features from amino acid sequence and structure, including evolutionary information, solvent accessible surface area and torsion angles (φ, ψ) in the backbone structure of the polypeptide chain, a computational method for predicting RNA-binding sites in proteins is proposed. When the method is applied to predict RNA-binding sites in three datasets: RBP86 containing 86 protein chains, RBP107 containing 107 proteins chains and RBP109 containing 109 proteins chains, better sensitivities and specificities are obtained compared to previously published methods in five-fold cross-validation tests. In order to make further examination for the efficiency of our method, the RBP107 dataset is used as training set, RBP86 and RBP109 datasets are used as the independent test sets. In addition, as examples of our prediction, RNA-binding sites in a few proteins are presented. The annotated results are consistent with the PDB annotation. These results show that our method is useful for annotating RNA binding sites of novel proteins.

Protein-RNA interface residue prediction using machine learning: an assessment of the state of the art

BACKGROUND

RNA molecules play diverse functional and structural roles in cells. They function as messengers for transferring genetic information from DNA to proteins, as the primary genetic material in many viruses, as catalysts (ribozymes) important for protein synthesis and RNA processing, and as essential and ubiquitous regulators of gene expression in living organisms. Many of these functions depend on precisely orchestrated interactions between RNA molecules and specific proteins in cells. Understanding the molecular mechanisms by which proteins recognize and bind RNA is essential for comprehending the functional implications of these interactions, but the recognition 'code' that mediates interactions between proteins and RNA is not yet understood. Success in deciphering this code would dramatically impact the development of new therapeutic strategies for intervening in devastating diseases such as AIDS and cancer. Because of the high cost of experimental determination of protein-RNA interfaces, there is an increasing reliance on statistical machine learning methods for training predictors of RNA-binding residues in proteins. However, because of differences in the choice of datasets, performance measures, and data representations used, it has been difficult to obtain an accurate assessment of the current state of the art in protein-RNA interface prediction.

RESULTS

We provide a review of published approaches for predicting RNA-binding residues in proteins and a systematic comparison and critical assessment of protein-RNA interface residue predictors trained using these approaches on three carefully curated non-redundant datasets. We directly compare two widely used machine learning algorithms (Naïve Bayes (NB) and Support Vector Machine (SVM)) using three different data representations in which features are encoded using either sequence- or structure-based windows. Our results show that (i) Sequence-based classifiers that use a position-specific scoring matrix (PSSM)-based representation (PSSMSeq) outperform those that use an amino acid identity based representation (IDSeq) or a smoothed PSSM (SmoPSSMSeq); (ii) Structure-based classifiers that use smoothed PSSM representation (SmoPSSMStr) outperform those that use PSSM (PSSMStr) as well as sequence identity based representation (IDStr). PSSMSeq classifiers, when tested on an independent test set of 44 proteins, achieve performance that is comparable to that of three state-of-the-art structure-based predictors (including those that exploit geometric features) in terms of Matthews Correlation Coefficient (MCC), although the structure-based methods achieve substantially higher Specificity (albeit at the expense of Sensitivity) compared to sequence-based methods. We also find that the expected performance of the classifiers on a residue level can be markedly different from that on a protein level. Our experiments show that the classifiers trained on three different non-redundant protein-RNA interface datasets achieve comparable cross-validation performance. However, we find that the results are significantly affected by differences in the distance threshold used to define interface residues.

CONCLUSIONS

Our results demonstrate that protein-RNA interface residue predictors that use a PSSM-based encoding of sequence windows outperform classifiers that use other encodings of sequence windows. While structure-based methods that exploit geometric features can yield significant increases in the Specificity of protein-RNA interface residue predictions, such increases are offset by decreases in Sensitivity. These results underscore the importance of comparing alternative methods using rigorous statistical procedures, multiple performance measures, and datasets that are constructed based on several alternative definitions of interface residues and redundancy cutoffs as well as including evaluations on independent test sets into the comparisons.

PRIDB: a Protein-RNA interface database

The Protein-RNA Interface Database (PRIDB) is a comprehensive database of protein-RNA interfaces extracted from complexes in the Protein Data Bank (PDB). It is designed to facilitate detailed analyses of individual protein-RNA complexes and their interfaces, in addition to automated generation of user-defined data sets of protein-RNA interfaces for statistical analyses and machine learning applications. For any chosen PDB complex or list of complexes, PRIDB rapidly displays interfacial amino acids and ribonucleotides within the primary sequences of the interacting protein and RNA chains. PRIDB also identifies ProSite motifs in protein chains and FR3D motifs in RNA chains and provides links to these external databases, as well as to structure files in the PDB. An integrated JMol applet is provided for visualization of interacting atoms and residues in the context of the 3D complex structures. The current version of PRIDB contains structural information regarding 926 protein-RNA complexes available in the PDB (as of 10 October 2010). Atomic- and residue-level contact information for the entire data set can be downloaded in a simple machine-readable format. Also, several non-redundant benchmark data sets of protein-RNA complexes are provided. The PRIDB database is freely available online at http://bindr.gdcb.iastate.edu/PRIDB.