AIRBP: Accurate identification of RNA-binding proteins using machine learning techniques

Abstract

11Motivation: Identification of RNA-binding proteins (RBPs) that bind to ribonucleic acid molecules, is an 12 important problem in Computational Biology and Bioinformatics. It becomes indispensable to identify 13 RBPs as they play crucial roles in post-transcriptional control of RNAs and RNA metabolism as well as 14 have diverse roles in various biological processes such as splicing, mRNA stabilization, mRNA 15 localization, and translation, RNA synthesis, folding-unfolding, modification, processing, and degradation. 16The existing experimental techniques for identifying RBPs are time-consuming and expensive. Therefore, 17identifying RBPs directly from the sequence using computational methods can be useful to efficiently 18 annotate RBPs and assist the experimental design. In this work, we present a method, called AIRBP, which 19 is designed using an advanced machine learning technique, called stacking, to effectively predict RBPs by 20 utilizing features extracted from evolutionary information, physiochemical properties, and disordered 2 21 properties. Moreover, our method, AIRBP is trained on the useful feature-subset identified by the 22 evolutionary algorithm (EA). 23 Results: The results show that AIRBP attains Accuracy (ACC), F1-score, and MCC of 95.38%, 0.917, and 24 0.885, respectively, based on the benchmark dataset, using 10-fold cross-validation (CV). Further 25 evaluation of AIRBP on independent test set reveals that it achieves ACC, F1-score, and MCC of 93.04%, 26 0.943, and 0.855, for Human test set; 91.60%, 0.942 and 0.789 for S. cerevisiae test set; and 91.67%, 0.953 27 and 0.594 for A. thaliana test set, respectively. These results indicate that AIRBP outperforms the current 28 state-of-the-art method. Therefore, the proposed top-performing AIRBP can be useful for accurate 29 identification and annotation of RBPs directly from the sequence and help gain valuable insight to treat 30 critical diseases. 50 and study of RBPs and RNP complexes date back to almost half a century ago where experimental methods 51 such as purification of mRNPs from in vitro UV-irradiated polysomal fractions (Greenberg, 1979), from 52 UV-irradiated intact cells (Wagenmakers, et al., 1980) and untreated cells (Lindberg and Sundquist, 1974) 53 revealed the association of a specific set of proteins with mRNA (Baltz, et al., 2012). Recently, cutting-54 edge experimental approaches are developed to recognize numerous RBPs, which include identification of 55 860 RBPs in human HeLa cells (Castello, et al., 2012) using UV crosslinking methods, 797 RBPs in human 56 embryonic kidney cell line (Baltz, et al., 2012) using photoreactive nucleotide-enhanced UV crosslinking 4 57 and oligo(dT) purification approach, 555 mRNA-binding proteins from mouse embryonic stem cells 58 (Kwon, et al., 2013) using UV crosslinking, oligo(dT) and Mass Spectrometry and 120 RBPs from S. 59 cerevisiae cells (Mitchell, et al., 2013) using UV crosslinking and purification methods. These experiments 60for identifying and analyzing of RBPs, have broadened ou...