Transcription factor binding site prediction using deep learning

Abstract

A majority of the human genome consists of sequences that do not code for a particular protein, so called non-coding DNA. The non-coding region plays a major role in gene expression. These non-coding regions of the DNA contain cis-regulatory components such as promoters and enhancers, and can be bound by Transcription Factor (TF) proteins and thereby control the rate of transcription of DNA to messenger RNA. Next Generation Sequencing (NGS) techniques helps to identifying and studying the genomic factors such as Transcription Factor Binding Sites (TFBSs). Knowing the sequence specificities of DNA and RNA-binding proteins are essential for developing models of the regulatory processes in biological systems and for identifying causal disease variants. This research mainly focus on sequence specificities that can be ascertained from experimental data with deep learning techniques, which offer a scalable, flexible and unified computational approach for pattern discovery. Discovering the TFBS has immense significance in terms of developing techniques and evaluating regulatory processes in biological systems. The uniqueness of genetic sequences can be discovered with the TF. A genetic disease can be cured by determining the specificities of the gene sequences concerned. In molecular biology, identifying accurate TF binding sites is a challenge for researchers, and an efficient methodology is needed to recognize such binding sites. Currently, deep learning techniques have been yielding great results, especially in computational biology. In the present approach, the Convolutional Neural Network (CNN) and Artificial Neural Network (ANN) have been used. In the first experiment, CNN models were built to predict TFBSs from gene sequence datasets. newline newline newline