Residual Neural Network for Predicting Super-enhancers on Genome Scale

Abstract

Residual neural network (ResNet) is a Deep Learning model introduced by He et al. [13] in 2015 to enhance traditional Convolutional neural networks for computer vision problems. It uses skip connections over some layer blocks to avoid vanishing gradient problem. Currently, many researches are focused to test and prove the efficiency of the ResNet on different domains such as genomics. In this paper, we propose a new ResNet model for predicting super-enhancers on genome scale. In fact, the prediction of super-enhancers (SEs) has prominent roles in biological and pathological processes; especially that related to the detection and progression of tumors. The obtained results are very promising and they proved the performance of our proposal compared to the CNN results.