Using machine learning and high-throughput RNA sequencing to classify the precursors of small non-coding RNAs

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• High-throughput sequencing enables detailed transcriptome analysis.
• Small RNA sequencing (sRNA-seq) profiles regulatory RNAs like microRNAs and siRNAs.
• Cells contain diverse small RNAs (<50nt) beyond regulatory RNAs, including fragments from snoRNAs, snRNAs, scRNAs, tRNAs, and transposons.

Purpose of the Study:

• To introduce CoRAL (Classification of RNAs by Analysis of Length), a computational method for discriminating RNA classes from sRNA-seq data.
• To provide a user's guide for the CoRAL tool.
• To enable detailed analysis of small RNA populations and their biogenesis pathways.

Main Methods:

• Development of CoRAL, a computational tool utilizing length-based features for RNA classification.
• Application of CoRAL to high-throughput small RNA sequencing data.
• Analysis of features relevant to small RNA biogenesis pathways.

Main Results:

• CoRAL accurately distinguishes between different classes of small RNAs.
• The method leverages features indicative of RNA biogenesis pathways.
• CoRAL provides insights into variations in RNA processing across different tissues, conditions, and species.

Conclusions:

• CoRAL is an effective tool for classifying diverse small RNAs identified through high-throughput sequencing.
• The classification approach offers valuable insights into RNA processing and biogenesis.
• CoRAL facilitates comparative studies of RNA populations across biological contexts.