资 源 简 介
Few thousand of microRNAs (miRNAs) are actually experimentally validated in several species, but a lot remain to be discovered. In order to facilitate the research of new miRNAs, several bioinformatics prediction tools exist. Most of them predict the hairpin structure of a given pre-miRNA based on characteristics of the miRNAs precursors (pre-miRNAs), but they rarely take in account the case where the sequence of the miRNA is known. We have developed a new tool, miRdup, which, after the usual process of putative pre-miRNAs prediction, validate if the position of the miRNA on the hairpin is accurate. To this purpose it uses a random forest classifier trained with experimentally validated miRNAs from miRbase, with features that characterize the duplex of miRNA and its complementarity region. MiRdup has also the capacity to predict a miRNA given a pre-miRNA. Few tools already exist to fulfill this task, but are limited to the research of miRNAs with fixed length in specific lineages an
