Computational Prediction of MicroRNAs Encoded in Viral and Other Genomes

摘要

We present an overview of selected computational methods formicroRNA prediction. It is especially aimed at viral miRNAdetection. As the number of microRNAs increases and the range ofgenomes encoding miRNAs expands, it seems that these smallregulators have a more important role than has been previouslythought. Most microRNAs have been detected by cloning and Northernblotting, but experimental methods are biased towards abundantmicroRNAs as well as being time-consuming. Computational detectionmethods must therefore be refined to serve as a faster, better,and more affordable method for microRNA detection. We also presentdata from a small study investigating the problems ofcomputational miRNA prediction. Our findings suggest that theprediction of microRNA precursor candidates is fairly easy, whileexcluding false positives as well as exact prediction of themature microRNA is hard. Finally, we discuss possible improvementsto computational microRNA detection.