title: A computational multiomics method for quantifying activity and regulatory mode of transcription factors and its application in leukemia
creator: Berest, Ivan
subject: ddc-500
subject: 500 Natural sciences and mathematics
subject: ddc-570
subject: 570 Life sciences
description: Recent breakthroughs in sequencing technologies allowed researchers to generate extensive amounts of data characterizing cellular regulation at many levels. Consequently, this boosted our understanding of gene regulatory networks responsible for different biological processes and highlighted the overall importance of transcription factors (TFs). TFs are dynamic mediators that react to both intra- and extracellular changes in order to ultimately transmit signals and execute genetically inherited gene regulatory programs in a time- and location-specific manner. However, it is still challenging to quantify in vivo TF specific binding occupancy and dynamics due to the high complexity of the regulatory part of the genome. Modern technologies measuring chromatin changes (e.g., chromatin accessibility, DNA methylation, histone  modifications) can now generate testable hypotheses about the effects of TF binding on gene regulation.  In this thesis, I mainly describe the novel computational tool diffTF, a multiomics data integration tool for globally assessing differential TF activity and classifying TFs into transcriptional activators and repressors (by integrating chromatin accessibility and gene expression data). We applied it to a recently published ATAC-seq dataset from a cohort of chronic lymphocytic leukemia (CLL) patients and identified dozens of differential active TFs representing two different CLL subtypes that are inherently  linked to tumour progression. In addition, we integrated gene expression data from corresponding RNA-seq and were able to globally predict an activating or repressive role for 40% of the expressed TFs. We validated the approach on an independent CLL dataset and showed that the majority of TFs does not change their mode of action upon genetic or environmental perturbations. Finally, we extensively tested  and benchmarked diffTF to validate its technical robustness.  We also applied diffTF to a multiomics dataset from the mouse hematopoietic differentiation system and targeted potential TFs that are disturbed upon epigenetic dysregulation driven by a Tet methylcytosine dioxygenase 2 (TET2) knockout in acute myeloid leukemia (AML). TET2 plays an essential role in the cellular DNA methylation balance and is known to be frequently mutated in leukemia. We used  the first high-quality TET2 binding map to identify TF families that can facilitate TET2 binding in the genome.  In summary, we developed a novel hypothesis-generation computational tool that can, in a data-driven way, identify key regulators of cellular biological processes based on chromatin and expression data.
date: 2020
type: Dissertation
type: info:eu-repo/semantics/doctoralThesis
type: NonPeerReviewed
format: application/pdf
identifier: https://archiv.ub.uni-heidelberg.de/volltextserverhttps://archiv.ub.uni-heidelberg.de/volltextserver/28844/1/PhDthesis_IvanBerest.pdf
identifier: DOI:10.11588/heidok.00028844
identifier: urn:nbn:de:bsz:16-heidok-288445
identifier:   Berest, Ivan  (2020) A computational multiomics method for quantifying activity and regulatory mode of transcription factors and its application in leukemia.  [Dissertation]     
relation: https://archiv.ub.uni-heidelberg.de/volltextserver/28844/
rights: info:eu-repo/semantics/openAccess
rights: Please see front page of the work (Sorry, Dublin Core plugin does not recognise license id)
language: eng