Decoding MicroRNA Motifs: A Time Series Approach using Hidden Markov Models

MicroRNAs (miRNAs) are short, RNAs that modulate stability, processing, and activity of microRNA after they are transcribed from the DNA template. With a length of approximately 22 nucleotides, miRNAs have the ability to regulate most of the human protein-coding genes. The identification and functional prediction of miRNAs is a critical task in bioinformatics. Discovering Motif is a well-established and significant approaches for determining the function of miRNAs.In this research, we propose a new approach to motif discovery in microRNA using a Hidden Markov Model (HMM). By modeling each sequence as a time series, we are able to use HMMs to effectively detect motifs. We have implemented our model and compared it with some of the best related models. Evaluation results reveal superiority of HMM from motif discovery point of view.