Detecting global influence of transcription factor interactions on gene expression in lymphoblastoid cells using neural network models

Abstract

Background Transcription factor(TF) interactions are known to regulate target gene(TG) expression in eukaryotes via TF regulatory modules(TRMs). Such interactions can be formed due to colocalizing TFs binding proximally to each other in the DNA sequence or over long distances between distally binding TFs via chromatin looping. While the former type of interaction has been characterized extensively, long distance TF interactions are still largely understudied. Furthermore, most prior approaches have focused on characterizing physical TF interactions without accounting for their effects on TG expression regulation. Understanding TRM based TG expression regulation could aid in understanding diseases caused by disruptions to these mechanisms. In this paper, we present a novel neural network based TRM detection approach that consists of using multi-omics TF based regulatory mechanism information to generate features for building non-linear multilayer perceptron TG expression prediction models in the GM12878 immortalized lymphoblastoid cells.