Prediction of novel pre-microRNAs with high accuracy through boosting and SVM

High-throughput deep-sequencing technology has generated an unprecedented number of expressed short sequence reads, presenting not only an opportunity but also a challenge for prediction of novel microRNAs. To verify the existence of candidate microRNAs, we have to show that these short sequences can be processed from candidate pre-microRNAs. However, it is laborious and time consuming to verify these using existing experimental techniques. Therefore, here, we describe a new method, miRD, which is constructed using two feature selection strategies based on support vector machines (SVMs) and boosting method. It is a high-efficiency tool for novel pre-microRNA prediction with accuracy up to 94.0% among different species.

miRD is implemented in PHP/PERL+MySQL+R and can be freely accessed at http://mcg.ustc.edu.cn/rpg/mird/mird.php.