Partner-specific prediction of RNA-binding residues in proteins: A critical assessment

RNA-protein interactions play essential roles in regulating gene expression. While some RNA-protein interactions are “specific”, that is, the RNA-binding proteins preferentially bind to particular RNA sequence or structural motifs, others are “non-RNA specific.” Deciphering the protein-RNA recognition code is essential for comprehending the functional implications of these interactions and for developing new therapies for many diseases. Because of the high cost of experimental determination of protein-RNA interfaces, there is a need for computational methods to identify RNA-binding residues in proteins. While most of the existing computational methods for predicting RNA-binding residues in RNA-binding proteins are oblivious to the characteristics of the partner RNA, there is growing interest in methods for partner-specific prediction of RNA binding sites in proteins. In this work, we assess the performance of two recently published partner-specific protein-RNA interface prediction tools, PS-PRIP, and PRIdictor, along with our own new tools. Specifically, we introduce a novel metric, RNA-specificity metric (RSM), for quantifying the RNA-specificity of the RNA binding residues predicted by such tools. Our results show that the RNA-binding residues predicted by previously published methods are oblivious to the characteristics of the putative RNA binding partner. Moreover, when evaluated using partner-agnostic metrics, RNA partner-specific methods are outperformed by the state-of-the-art partner-agnostic methods. We conjecture that either (a) the protein-RNA complexes in PDB are not representative of the protein-RNA interactions in nature, or (b) the current methods for partner-specific prediction of RNA-binding residues in proteins fail to account for the differences in RNA partner-specific versus partner-agnostic protein-RNA interactions, or both.     

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

ABSTRACT: 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 (Naive 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.     

A computational model for predicting protein interactions based on multidomain collaboration

Recently, several domain-based computational models for predicting protein-protein interactions (PPIs) have been proposed. The conventional methods usually infer domain or domain combination (DC) interactions from already known interacting sets of proteins, and then predict PPIs using the information. However, the majority of these models often have limitations in providing detailed information on which domain pair (single domain interaction) or DC pair (multidomain interaction) will actually interact for the predicted protein interaction. Therefore, a more comprehensive and concrete computational model for the prediction of PPIs is needed. We developed a computational model to predict PPIs using the information of intraprotein domain cohesion and interprotein DC coupling interaction. A method of identifying the primary interacting DC pair was also incorporated into the model in order to infer actual participants in a predicted interaction. Our method made an apparent improvement in the PPI prediction accuracy, and the primary interacting DC pair identification was valid specifically in predicting multidomain protein interactions. In this paper, we demonstrate that 1) the intraprotein domain cohesion is meaningful in improving the accuracy of domain-based PPI prediction, 2) a prediction model incorporating the intradomain cohesion enables us to identify the primary interacting DC pair, and 3) a hybrid approach using the intra/interdomain interaction information can lead to a more accurate prediction.     

A knowledge-based potential function predicts the specificity and relative binding energy of RNA-binding proteins

RNA-protein interactions are fundamental to gene expression. Thus, the molecular basis for the sequence dependence of protein-RNA recognition has been extensively studied experimentally. However, there have been very few computational studies of this problem, and no sustained attempt has been made towards using computational methods to predict or alter the sequence-specificity of these proteins. In the present study, we provide a distance-dependent statistical potential function derived from our previous work on protein-DNA interactions. This potential function discriminates native structures from decoys, successfully predicts the native sequences recognized by sequence-specific RNA-binding proteins, and recapitulates experimentally determined relative changes in binding energy due to mutations of individual amino acids at protein-RNA interfaces. Thus, this work demonstrates that statistical models allow the quantitative analysis of protein-RNA recognition based on their structure and can be applied to modeling protein-RNA interfaces for prediction and design purposes.     

Deciphering protein-protein interactions. Part II. Computational methods to predict protein and domain interaction partners

Recent advances in high-throughput experimental methods for the identification of protein interactions have resulted in a large amount of diverse data that are somewhat incomplete and contradictory. As valuable as they are, such experimental approaches studying protein interactomes have certain limitations that can be complemented by the computational methods for predicting protein interactions. In this review we describe different approaches to predict protein interaction partners as well as highlight recent achievements in the prediction of specific domains mediating protein-protein interactions. We discuss the applicability of computational methods to different types of prediction problems and point out limitations common to all of them.     

Predicting and validating protein interactions using network structure

Protein interactions play a vital part in the function of a cell. As experimental techniques for detection and validation of protein interactions are time consuming, there is a need for computational methods for this task. Protein interactions appear to form a network with a relatively high degree of local clustering. In this paper we exploit this clustering by suggesting a score based on triplets of observed protein interactions. The score utilises both protein characteristics and network properties. Our score based on triplets is shown to complement existing techniques for predicting protein interactions, outperforming them on data sets which display a high degree of clustering. The predicted interactions score highly against test measures for accuracy. Compared to a similar score derived from pairwise interactions only, the triplet score displays higher sensitivity and specificity. By looking at specific examples, we show how an experimental set of interactions can be enriched and validated. As part of this work we also examine the effect of different prior databases upon the accuracy of prediction and find that the interactions from the same kingdom give better results than from across kingdoms, suggesting that there may be fundamental differences between the networks. These results all emphasize that network structure is important and helps in the accurate prediction of protein interactions. The protein interaction data set and the program used in our analysis, and a list of predictions and validations, are available at http://www.stats.ox.ac.uk/bioinfo/resources/PredictingInteractions.     

Predicting subcellular localization of proteins using machine-learned classifiers

MOTIVATION: Identifying the destination or localization of proteins is key to understanding their function and facilitating their purification. A number of existing computational prediction methods are based on sequence analysis. However, these methods are limited in scope, accuracy and most particularly breadth of coverage. Rather than using sequence information alone, we have explored the use of database text annotations from homologs and machine learning to substantially improve the prediction of subcellular location. RESULTS: We have constructed five machine-learning classifiers for predicting subcellular localization of proteins from animals, plants, fungi, Gram-negative bacteria and Gram-positive bacteria, which are 81% accurate for fungi and 92-94% accurate for the other four categories. These are the most accurate subcellular predictors across the widest set of organisms ever published. Our predictors are part of the Proteome Analyst web-service.     

PaleAle 5.0: prediction of protein relative solvent accessibility by deep learning

Predicting the three-dimensional structure of proteins is a long-standing challenge of computational biology, as the structure (or lack of a rigid structure) is well known to determine a protein’s function. Predicting relative solvent accessibility (RSA) of amino acids within a protein is a significant step towards resolving the protein structure prediction challenge especially in cases in which structural information about a protein is not available by homology transfer. Today, arguably the core of the most powerful prediction methods for predicting RSA and other structural features of proteins is some form of deep learning, and all the state-of-the-art protein structure prediction tools rely on some machine learning algorithm. In this article we present a deep neural network architecture composed of stacks of bidirectional recurrent neural networks and convolutional layers which is capable of mining information from long-range interactions within a protein sequence and apply it to the prediction of protein RSA using a novel encoding method that we shall call “clipped”. The final system we present, PaleAle 5.0, which is available as a public server, predicts RSA into two, three and four classes at an accuracy exceeding 80% in two classes, surpassing the performances of all the other predictors we have benchmarked.

     

A knowledge-based scoring function for protein-RNA interactions derived from a statistical mechanics-based iterative method

Protein-RNA interactions play important roles in many biological processes. Given the high cost and technique difficulties in experimental methods, computationally predicting the binding complexes from individual protein and RNA structures is pressingly needed, in which a reliable scoring function is one of the critical components. Here, we have developed a knowledge-based scoring function, referred to as ITScore-PR, for protein-RNA binding mode prediction by using a statistical mechanics-based iterative method. The pairwise distance-dependent atomic interaction potentials of ITScore-PR were derived from experimentally determined protein–RNA complex structures. For validation, we have compared ITScore-PR with 10 other scoring methods on four diverse test sets. For bound docking, ITScore-PR achieved a success rate of up to 86% if the top prediction was considered and up to 94% if the top 10 predictions were considered, respectively. For truly unbound docking, the respective success rates of ITScore-PR were up to 24 and 46%. ITScore-PR can be used stand-alone or easily implemented in other docking programs for protein–RNA recognition.     

Predicting domain-domain interaction based on domain profiles with feature selection and support vector machines

Background  

Protein-protein interaction (PPI) plays essential roles in cellular functions. The cost, time and other limitations associated with the current experimental methods have motivated the development of computational methods for predicting PPIs. As protein interactions generally occur via domains instead of the whole molecules, predicting domain-domain interaction (DDI) is an important step toward PPI prediction. Computational methods developed so far have utilized information from various sources at different levels, from primary sequences, to molecular structures, to evolutionary profiles.     

DomPep--a general method for predicting modular domain-mediated protein-protein interactions

Protein-protein interactions (PPIs) are frequently mediated by the binding of a modular domain in one protein to a short, linear peptide motif in its partner. The advent of proteomic methods such as peptide and protein arrays has led to the accumulation of a wealth of interaction data for modular interaction domains. Although several computational programs have been developed to predict modular domain-mediated PPI events, they are often restricted to a given domain type. We describe DomPep, a method that can potentially be used to predict PPIs mediated by any modular domains. DomPep combines proteomic data with sequence information to achieve high accuracy and high coverage in PPI prediction. Proteomic binding data were employed to determine a simple yet novel parameter Ligand-Binding Similarity which, in turn, is used to calibrate Domain Sequence Identity and Position-Weighted-Matrix distance, two parameters that are used in constructing prediction models. Moreover, DomPep can be used to predict PPIs for both domains with experimental binding data and those without. Using the PDZ and SH2 domain families as test cases, we show that DomPep can predict PPIs with accuracies superior to existing methods. To evaluate DomPep as a discovery tool, we deployed DomPep to identify interactions mediated by three human PDZ domains. Subsequent in-solution binding assays validated the high accuracy of DomPep in predicting authentic PPIs at the proteome scale. Because DomPep makes use of only interaction data and the primary sequence of a domain, it can be readily expanded to include other types of modular domains.     

Analysis of CLIP and iCLIP methods for nucleotide-resolution studies of protein-RNA interactions

ABSTRACT: UV cross-linking and immunoprecipitation (CLIP) and individual-nucleotide resolution CLIP (iCLIP) are methods to study protein-RNA interactions in untreated cells and tissues. Here, we analyzed six published and two novel data sets to confirm that both methods identify protein-RNA cross-link sites, and to identify a slight uridine preference of UV-C-induced cross-linking. Comparing Nova CLIP and iCLIP data revealed that cDNA deletions have a preference for TTT motifs, whereas iCLIP cDNA truncations are more likely to identify clusters of YCAY motifs as the primary Nova binding sites. In conclusion, we demonstrate how each method impacts the analysis of protein-RNA binding specificity.     

MHCPred 2.0

The accurate computational prediction of T-cell epitopes can greatly reduce the experimental overhead implicit in candidate epitope identification within genomic sequences. In this article we present MHCPred 2.0, an enhanced version of our online, quantitative T-cell epitope prediction server. The previous version of MHCPred included mostly alleles from the human leukocyte antigen A (HLA-A) locus. In MHCPred 2.0, mouse models are added and computational constraints removed. Currently the server includes 11 human HLA class I, three human HLA class II, and three mouse class I models. Additionally, a binding model for the human transporter associated with antigen processing (TAP) is incorporated into the new MHCPred. A tool for the design of heteroclitic peptides is also included within the server. To refine the veracity of binding affinities prediction, a confidence percentage is also now calculated for each peptide predicted. AVAILABILITY: As previously, MHCPred 2.0 is freely available at the URL http://www.jenner.ac.uk/MHCPred/ CONTACT: Darren R. Flower (darren.flower@jenner.ac.uk).     

Recent advances in predicting protein–protein interactions with the aid of artificial intelligence algorithms

Protein–protein interactions (PPIs) are essential in the regulation of biological functions and cell events, therefore understanding PPIs have become a key issue to understanding the molecular mechanism and investigating the design of drugs. Here we highlight the major developments in computational methods developed for predicting PPIs by using types of artificial intelligence algorithms. The first part introduces the source of experimental PPI data. The second part is devoted to the PPI prediction methods based on sequential information. The third part covers representative methods using structural information as the input feature. The last part is methods designed by combining different types of features. For each part, the state-of-the-art computational PPI prediction methods are reviewed in an inclusive view. Finally, we discuss the flaws existing in this area and future directions of next-generation algorithms.