Transcriptome-Wide Prediction Of lncRNA-RNA Interactions By A Thermodynamics Algorithm

Abstract

MotivationThe discovery of thousands of long noncoding RNAs (lncRNAs) in mammals raises a question about their functionality. It has been shown that some of them function post-transcriptionally via formation of inter-molecular duplexes. Sequence alignment tools are frequently used for transcriptome-wide prediction of RNA-RNA interactions. However, such approaches have poor prediction accuracy since they ignore RNA secondary structure and interaction energy. On the other hand, application of the thermodynamics-based algorithms to long transcripts is not computationally feasible on a large scale.\n\nResultsHere we describe a new computational pipeline ASSA that combines sequence alignment and thermodynamics tools for efficient prediction of RNA-RNA interactions between long transcripts. ASSA outperforms four other tools in terms of the Area Under the Curve. ASSA predictions for the lncRNA HOTAIR confirm that it binds to the chromatin through hybridization with the nascent transcripts. Analysis of the 49 murine lncRNA knockdown experiments reveals one transcript that may regulate its targets via RNA-RNA interactions.\n\nAvailabilityASSA is available at http://assa.sourceforge.net/.\n\nContactivan.antonov@gatech.edu