Identification of key regulators from ATAC-seq using neural networks
Title: Identification of key regulators from ATAC-seq using neural networks
SNIC Project: SNIC 2020/13-14
Project Type: SNIC Small Compute
Principal Investigator: Julia Åkesson <julak640@student.liu.se>
Affiliation: Linköpings universitet
Duration: 2020-02-11 – 2021-03-01
Classification: 10610
Keywords:

Abstract

Transcription factors (TFs) regulating many target genes play important roles as disease regulators. These key regulators could be identified using ATAC-seq, which is a high-throughput sequencing technology to measure open chromatin sites in the genome. The opening of the chromatin is a requisite for TFs to bind to the DNA and regulate the expression of their target genes. However, current methods to identify TF binding sites results in many false positive results. The use of a neural network approach serve as a possible way to extract more reliable predictions of TF-target interactions from ATAC-seq. A trained neural network could be used to make personalized gene regulatory network from ATAC-seq data and to identify key regulators of disease. The ultimate goal is to apply the trained neural network to identify key regulators of the disease Multiple sclerosis.