Computational Prediction of RNA Binding Proteins: Features and Models

Abstract

To date, two major layers of gene regulation have been identified during RNA life cycle. These include: transcription regulation by transcription and epigenetic factors and post-transcription regulation by various classes of small RNA (sRNAs). Besides these two layers, there is another layer of gene regulation which is completely regulated by RNA-binding proteins (RBPs). RBPs play crucial roles in gene expression and regulation in both transcriptional and post-transcriptional levels. RBPs account for approximately 6–8% of all proteins. A key objective of computational biology is to identify these RBPs. Although a variety of experimental methods for RBP identification have been developed, these techniques are costly, time-consuming, and labour-intensive. Alternatively, researchers have developed multiple computational approaches for predicting RBPs by integrating multiple machine learning and deep learning methods with numerical RBP features. In this chapter, we discuss the computational methods, datasets, and features for computational recognition of RBPs. We believe that this chapter will provide valuable information as far as computational prediction of RBPs is concerned.