DP-miRNA: An improved prediction of precursor microRNA using deep learning model

MicroRNA (miRNA) are small non-coding RNAs regulating gene expression at the post-transcriptional level. Detecting miRNA in a genome is challenging experimentally and results vary depending on their cellular environment. These limitations inspire the development of knowledge-based prediction method. This paper proposes a deep learning based classification model for predicting precursor miRNA sequence that contains the miRNA sequence. The feature set consists of sequence features, folding measures, stem-loop features and statistical features. We evaluate the performance of the proposed method on human dataset. The deep neural network based classification outperformed support vector machine, neural network, naive Bayes classifiers, k-nearest neighbors, random forests as well as hybrid systems combining SVM and genetic algorithm.