Evaluating computational tools for analysis of ATAC-seq data and transcription factor footprinting.

Abstract

Assay for Transposase Accessible Chromatin combined with high-throughput sequencing (ATAC-seq) is a relatively new method for inspecting the accessibility of chromatin. ATAC-seq data can be used for discovering footprints of transcription factor binding sites. As the tools for these analyses are rapidly developing the aim of this thesis was to compare a new method (HMMRATAC) of peak calling to an older method (MACS2). Also, to test and compare two footprinting analysis software HINT-ATAC and TOBIAS. In total two datasets of varying quality were analyzed for possible transcription factor binding sites using both peak calling software combined with both footprinting software using their default parameters. Both peak calling software were able to call similar peaks, with some differences in peak width. However, the footprinting software HINT-ATAC seemed to be more user friendly by having a more flexible output format to be used in follow up analysis. In comparison TOBIAS used a different output format, which could affect the possible follow up analyses. All tested software were able to produce useable results. However, tweaking of analysis parameters should be considered as different data sets can behave differently.