Improving pre-miRNA prediction with complexity measures of the mature and deep learning

L. BUGNON, C. YONES, J. RAAD, D.H. MILONE, G. STEGMAYER

Mendoza

Congreso; X Congreso Argentino de Bioinformática y Biología Computacional; 2019

A2B2C

The miRNAs are small RNAmolecules that regulate gene expression in animal and plant cellsthrough post-transcriptional control. They are stored insideprecursors of 100 bases long approximately called pre-miRNAs, whichhave a stem-loop structure. Several experimental methods fordetecting pre-miRNAs can be used , such as qPCR, microarray and deepsequencing. However, these techniques present some practicaldifficulties when evaluating a very large number of candidatesequences in a genome. Due to these technical and practicaldifficulties, computational methods play an increasingly importantrole for their prediction . In order to find new candidates forpre-miRNA, many different features sets have been proposed, whichmostly describe information of the structure of the pre-miRNAinspired by the action of Drosha. However, the specificity of thesubsequent processes impose restrictions on those hairpins that willbecome mature miRNA. Given that this important information iscodified in the mature region, the secondary structure of theprecursor by itself might not be sufficient to differentiate a truepre-miRNA from other hairpins.