Predicting the fate of microRNA target genes based on sequence features

MicroRNAs (miRNAs) are important post-transcriptional regulators that repress gene expression by binding to the 3′UTRs of their target mRNAs. There are two main outcomes for the transcripts targeted by miRNAs: mRNA degradation and translational repression. It is still unclear what factors determine whether a target transcript is degraded or translationally repressed. In this study, we collected two classes of genes that are targeted by miR-1, miR-155, miR-16, miR-30a, and let-7b and built new computational models with machine-learning methods to predict the fates of target genes based on sequence features. The prediction results indicate that the sequence context of the miRNA binding site at the 3′UTR of a target gene plays an important role in determining how an miRNA regulates the expression of its target. Further analysis shows that four out of the five studied miRNAs probably share similar regulatory mechanisms on their target genes.