RNAcon: Prediction and classification of ncRNAs using structural information

RNAcon: Prediction and Classification of ncRNAs Using Structural Information RNAcon is a computational tool developed for the prediction and classification of non-coding RNAs, also known as ncRNAs. The tool first discriminates coding RNA sequences from non-coding RNA sequences and then classifies non-coding RNAs into different functional classes using structural information. RNAcon uses sequence composition, RNA secondary structure prediction, graph properties, and machine learning models to support large-scale ncRNA analysis. Web Server: https://webs.iiitd.edu.in/raghava/rnacon/ Citation Panwar, B., Arora, A., and Raghava, G. P. S. Prediction and classification of ncRNAs using structural information. BMC Genomics, 15, 127, 2014. https://doi.org/10.1186/1471-2164-15-127 About the Research Non-coding RNAs are RNA transcripts that do not encode proteins but play important roles in cellular processes such as gene regulation, RNA processing, RNA modification, chromosome stability, protein stability, and developmental regulation. With the growth of high-throughput sequencing data, there is a major need for computational methods that can identify whether a transcript is coding or non-coding and further classify ncRNAs into their respective functional families. RNAcon was developed to perform both tasks: prediction of ncRNAs and classification of ncRNAs into different classes. Data Compilation: For ncRNA prediction, non-coding RNA sequences were collected from the Rfam database and coding RNA sequences were collected from the RefSeq database. For ncRNA classification, the study used datasets from 18 different ncRNA classes originally used in the GraPPLE study. Methodology: RNAcon uses an SVM-based model with tri-nucleotide composition for discriminating coding and non-coding RNA sequences. For classifying ncRNAs, RNA secondary structures were predicted using IPknot, graph properties were calculated using the igraph R package, and machine learning classifiers were used to assign ncRNAs into functional classes.