Computational prediction of novel miRNAs from genome-wide data

KAMENETZKY, L.; YONES C; MILONE, D.; STEGMAYER, G.; MACCHIAROLI, N.

Functional Genomics

Springer

Año: 2016; p. 1 - 13

The computational prediction of novel microRNAs (miRNAs) within afull genome involves identifying sequences having the highest chance of being bonafide miRNA precursors (pre-miRNAs). These sequences are usually named candi-dates to miRNA. The well-known pre-miRNAs are usually only a few in comparisonto the hundreds of thousands of potential candidates to miRNA that have to be an-alyzed. Although the selection of positive labeled examples is straightforward, it isvery difficult to build a set of negative examples in order to obtain a good set oftraining samples for a supervised method. In this chapter we describe an approachto this problem, based on the unsupervised clustering of unlabeled sequences fromgenome-wide data, and the well-known miRNA precursors for the organism understudy. Therefore, the protocol developed allows for quick identification of the bestcandidates to miRNA as those sequences clustered together with known precursors.