200 Novel Microrna Prediction in AMPV Genomeg

In this paper, we successfully predicted 4 microRNAs in genome of Avian metapneumovirus by using BNMir-P program, and further validity it in second round prediction using miPred software which based on machine-learned algorithm termed random forest (RF). By BLAST against reported mature sequences, these 4 AMPV microRNAs were regarded as novel microRNAs. Target prediction results showed total 23 target genes were found in chicken UTR database. Based their function, the targets could be divided in to 4 groups, including immunity system genes, lipid metabolism genes, ion-channel genes involving in Ca²⁺ and K⁺ transporting and gene expression regulators. This finding connected the AMPV infection mechanism and its harm effects with the disturbed gene expression and enzyme function, giving deep understand of the AMPV infection and providing novel clues for further research. Taken together, the result demostrated the prediction flow was accurate, sensitive and species indenpendent, and was suitable for microRNA prediction of other animal virus.