Computational Prediction of RNA-Binding Proteins and Binding Sites

Abstract

Proteins and RNA interaction have vital roles in many cellular processes such as protein synthesis, sequence encoding, RNA transfer, and gene regulation at the transcriptional and post-transcriptional levels. Approximately 6%–8% of all proteins are RNA-binding proteins (RBPs). Distinguishing these RBPs or their binding residues is a major aim of structural biology. Previously, a number of experimental methods were developed for the determination of protein–RNA interactions. However, these experimental methods are expensive, time-consuming, and labor-intensive. Alternatively, researchers have developed many computational approaches to predict RBPs and protein–RNA binding sites, by combining various machine learning methods and abundant sequence and/or structural features. There are three kinds of computational approaches, which are prediction from protein sequence, prediction from protein structure, and protein-RNA docking. In this paper, we review all existing studies of predictions of RNA-binding sites and RBPs and complexes, including data sets used in different approaches, sequence and structural features used in several predictors, prediction method classifications, performance comparisons, evaluation methods, and future directions.

SONAR Discovers RNA-Binding Proteins from Analysis of Large-Scale Protein-Protein Interactomes

“…Attempts to predict RBPs from primary sequence and protein structure have yielded only moderate success (Si et al, 2015), particularly since more than a third of RBPs have no prior RNA-binding related homology or annotation (Castello et al, 2013). To expand the catalog of RBPs beyond ones with sequence or structure homology, several experimental approaches have been undertaken.…”

De-novo protein function prediction using DNA binding and RNA binding proteins as a test case

“…Over the last fifteen years, dozens of computational methods of these two types have been introduced. Due to their number it is impossible to review all of them here but many of them are discussed in recent reviews1516. The success of these methods notwithstanding, they are limited in their scope and applicability: The first class of methods, namely methods for the discovery of proteins that bind NA, is usually limited to those proteins that have overall similarity to known NABPs.…”

Sequence-Based Prediction of RNA-Binding Residues in Proteins

“…There are more than 20 published approaches for predicting RNA-binding residues in proteins (for a recent compilation, see [50]), and a few methods are capable of predicting interfacial residues in both the protein and the RNA components of a protein–RNA complex (e.g., [52, 82]). Subheadings 3.2–3.5 below focus on three methods (RNABindRPlus, SNBRFinder, PS-PRIP) that are freely available on web-based servers and have been shown to perform well on benchmark datasets.…”

“…Computational approaches to predicting protein–RNA interfaces have been the topic of several recent reviews and benchmark comparisons [31, 47–50]. These approaches can be broadly classified into sequence- and structure-based methods [31, 47].…”

“…We also provide a brief guide to accessing and utilizing state-of-the-art computational methods, web servers and databases that provide information about interfaces in protein–RNA complexes and/or predictions of RNA-binding residues in proteins. For additional information, the reader is referred to two excellent reviews: a recent review by Si et al [50], which includes a comprehensive table of available sequence, structure and docking based methods; and a review by Tuszynska et al [59], which focuses on structural docking-based approaches which are not considered here.…”