Prediction of miRNA based on flexible neural tree

MicroRNA(miRNA) is a class of 20-24 nucleotides conserved non-coding small RNA. How to predict miRNA accurately is one of the difficulties in bioinformatics. A new predicting method has been proposed in this paper, i.e., particle swarm optimized flexible neural tree. We use 331 samples, each of which is comprised of 36 features to train the flexible neural tree model. When we get the optimized model it was used to test trainsets and get prediction accuracy up to 91.8%. So, the flexible neural tree methods are proved to be effectual. This indicates that our model can be used as a new direction to predict miRNA.