RNApin: Identification of Protein-Interacting Nucleotides in an RNA Sequence Using Composition Profile of Tri-Nucleotides

RNApin: Identification of Protein-Interacting Nucleotides in an RNA Sequence Using Composition Profile of Tri-Nucleotides RNApin is a computational web server developed for predicting protein-interacting nucleotides in RNA sequences. RNA-protein interactions play important roles in translation, gene expression, gene regulation, RNA stability, and many cellular processes. RNApin helps identify nucleotide positions in RNA that are likely to interact with proteins using machine learning-based prediction models. Web Server: https://webs.iiitd.edu.in/raghava/rnapin Citation Panwar, B., and Raghava, G. P. S. Identification of protein-interacting nucleotides in a RNA sequence using composition profile of tri-nucleotides. Genomics, 105(4), 197-203, 2015. https://doi.org/10.1016/j.ygeno.2015.01.005 This tool and dataset is also available on Zenodo at About the Research RNA-protein interactions are essential for several biological processes, including protein translation, gene expression, RNA processing, and gene regulation. RNA-binding proteins interact with specific nucleotide regions in RNA molecules, and these interactions are important for normal cellular function. Failure or disruption of RNA-protein interactions is associated with several human genetic diseases, including fragile X syndrome, spinal muscular atrophy, myotonic dystrophy, paraneoplastic neurologic syndromes, and fragile X tremor ataxia syndrome. Although many methods exist for predicting RNA-interacting residues in proteins, limited methods were available for identifying protein-interacting nucleotides in RNA sequences. RNApin was developed to address this gap. Data Compilation: Protein-interacting RNA chains were collected from the PRIDB database. A total of 1546 protein-interacting RNA chains were retrieved, and a 25% non-redundant dataset of 208 RNA chains was created using BLASTCLUST. Using a 5 Å distance cutoff, 10,198 protein-interacting nucleotides and 36,384 non-interacting nucleotides were identified. Methodology: RNApin uses sliding window-based patterns and machine learning models to classify nucleotides as protein-interacting or non-interacting. Different feature approaches were tested, including binary profile of patterns, mono-nucleotide composition, di-nucleotide composition, and tri-nucleotide composition.