Identifying multiple stem-loops pre-miRNA using support vector machine

Those pre-miRNAs with multiple loops are usually excluded in the most existing prediction methods. But as more and more miRNA have been identified, amount of miRNA precursor with multiple loops have been found. Therefore, determining how to effectively predict real pre-miRNA with multiple loops from those large of pseudo pre-miRNAs with multiple loops is an imperative problem. Some features of main branch are extracted to describe pre-miRNA intrinsic features, and SVM classifier is implemented to recognize real pre-miRNA with multiple stem-loops. Training and testing on dataset from miRBase12.0, SVM classifier achieves sensitivity of 75.76% and specificity of 95.16% on human test set, and when being applied to pre-miRNAs of all other species, it correctly identifies 86.71% of them. The proposed method in this work provides a powerful predicting method to recognize the real pre-miRNA with multiple stem-loops.