MiRA: adaptable novel miRNA identification in plants using small RNA sequencing data. Issue 1 (December 2015)

• Record Type: Journal Article
• Authors: Evers, Maurits; Huttner, Michael; Dueck, Anne; Meister, Gunter; Engelmann, Julia
• Is Part Of: BMC bioinformatics. Volume 16:Issue 1(2015)
• Page Start: 1
• Page End: 10
• Publication Date: 2015-12
• Subjects: Sequencing data -- miRNA identification -- RNA secondary structure -- Chlamydomonas reinhardtii -- Small RNA sequencing -- Next generation sequencing; Bioinformatics -- Periodicals; Computational biology -- Periodicals; 570.285
• DOI: 10.1186/s12859-015-0798-3
• Languages: English
• ISSNs: 1471-2105

Abstract

Abstract Background MicroRNAs (miRNAs) are short regulatory RNAs derived from longer precursor RNAs. miRNA biogenesis has been studied in animals and plants, recently elucidating more complex aspects, such as non-conserved, species-specific, and heterogeneous miRNA precursor populations. Small RNA sequencing data can help in computationally identifying genomic loci of miRNA precursors. The challenge is to predict a valid miRNA precursor from inhomogeneous read coverage from a complex RNA library: while the mature miRNA typically produces many sequence reads, the remaining part of the precursor is covered very sparsely. As recent results suggest, alternative miRNA biogenesis pathways may lead to a more diverse miRNA precursor population than previously assumed. In plants, the latter manifests itself in e.g. complex secondary structures and expression from multiple loci within precursors. Current miRNA identification algorithms often depend on already existing gene annotation, and/or make use of specific miRNA precursor features such as precursor lengths, secondary structures etc. Consequently and in view of the emerging new understanding of a more complex miRNA biogenesis in plants, current tools may fail to characterise organism-specific and heterogeneous miRNA populations. Results miRA is a new tool to identify miRNA precursors in plants, allowing for heterogeneous and complex precursor populations. miRA requires small RNA sequencing data and a corresponding reference genome, and evaluates precursor secondary structures and precursor processing accuracy; key parameters can be adapted based on the specific organism under investigation. We show that miRA outperforms the currently best plant miRNA prediction tools both in sensitivity and specificity, for data involvingArabidopsis thaliana and the Volvocine algaeChlamydomonas reinhardtii ; the latter organism has been shown to exhibit a heterogeneous and complex precursor population with little cross-species miRNA sequence conservation, and therefore constitutes an ideal model organism. Furthermore we identify novel miRNAs in the Chlamydomonas-related organismVolvox carteri . Conclusions We propose miRA, a new plant miRNA identification tool that is well adapted to complex precursor populations. miRA is particularly suited for organisms with no existing miRNA annotation, or without a known related organism with well characterized miRNAs. Moreover, miRA has proven its ability to identify species-specific miRNAs. miRA is flexible in its parameter settings, and produces user-friendly output files in various formats (pdf, csv, genome-browser-suitable annotation files, etc.). It is freely available athttps://github.com/mhuttner/miRA .