%0 Journal Article
%T A Contig-Based Strategy for the Genome-Wide Discovery of MicroRNAs without Complete Genome Resources
%A Jun-Zhi Wen
%A Jian-You Liao
%A Ling-Ling Zheng
%A Hui Xu
%A Jian-Hua Yang
%A Dao-Gang Guan
%A Si-Min Zhang
%A Hui Zhou
%A Liang-Hu Qu
%J PLOS ONE
%D 2014
%I Public Library of Science (PLoS)
%R 10.1371/journal.pone.0088179
%X MicroRNAs (miRNAs) are important regulators of many cellular processes and exist in a wide range of eukaryotes. High-throughput sequencing is a mainstream method of miRNA identification through which it is possible to obtain the complete small RNA profile of an organism. Currently, most approaches to miRNA identification rely on a reference genome for the prediction of hairpin structures. However, many species of economic and phylogenetic importance are non-model organisms without complete genome sequences, and this limits miRNA discovery. Here, to overcome this limitation, we have developed a contig-based miRNA identification strategy. We applied this method to a triploid species of edible banana (GCTCV-119, Musa spp. AAA group) and identified 180 pre-miRNAs and 314 mature miRNAs, which is three times more than those were predicted by the available dataset-based methods (represented by EST+GSS). Based on the recently published miRNA data set of Musa acuminate, the recall rate and precision of our strategy are estimated to be 70.6% and 92.2%, respectively, significantly better than those of EST+GSS-based strategy (10.2% and 50.0%, respectively). Our novel, efficient and cost-effective strategy facilitates the study of the functional and evolutionary role of miRNAs, as well as miRNA-based molecular breeding, in non-model species of economic or evolutionary interest.
%U http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0088179