Tomkova Group Research Overview
Genomes of cancer patients carry many changes – mutations in the DNA itself, as well as changes in the epigenetic marks on and around the DNA. It is important to understand (1) what causes these changes in the first place, (2) which of these changes causally contribute to carcinogenesis, (3) how do the genetic and epigenetic changes interact to cause cancer, (4) and how can we use this knowledge for prevention (changing our lifestyle to avoid these changes), early cancer detection (development of biomarkers), and treatment (based on the discovered vulnerabilities). We address these questions using highly interdisciplinary approaches, large quantities of genomic data, and collaborative research.
Our past and ongoing research themes include:
Which mutations are due to dna polymerase errors?
Despite replication errors being a well-known source of DNA mutations, it is unclear what proportion of the mutations observed in different cancers, or even normal cells and germline, are due to DNA polymerase errors. We have shown that DNA replication leaves a footprint in majority of the known cancer mutational signatures – in the form of replication strand asymmetry or replication timing bias (Tomkova et al., Genome Biology, 2018). However, this footprint can be due to both direct and indirect mechanisms. We have also shown that replication errors may play a role even in mutational processes previously explained by other mechanisms, such as mutational signature 1 (Tomkova et al., DNA Repair, 2018). In ongoing projects, we use a combination of novel experimental sequencing techniques, computational methods, and integration of large human, mouse, and cell-line datasets, to disentangle the contribution of individual DNA polymerases to the different mutational processes that operate in human cells and contribute to cancer.
How can we use tissue-specific epigenome to decipher non-coding cancer drivers?
Finding mutations that drive cancer is a key goal in cancer research, important for therapy and understanding tumorigenesis. The vast majority of known cancer drivers lie in the protein-coding genome. However, the coding regions form at most ca. 2% of our genome. What happens in the remaining 98% of genome? We developed a computational framework Dr.Nod for Discovery of Regulatory NOn-coding Drivers and showed the importance of using tissue-specific epigenome for non-coding driver discovery (Tomkova et al., NAR, 2023).
How do epigenetic marks influence mutagenesis?
The mutagenic role of 5-methylcytosine, the major DNA modification (epigenetic mark) in our genomes, is well known. We investigated the mutagenic roles of 5-hydroxymethylcytosine, the second major DNA modification in human genome, showing that CpGs marked with 5-hydroxymethylcytosine are largely protected from this mutagenicity (Tomkova et al., 2016).
How do epigenetic marks interact to govern gene expression and contribute to cancer development?
Different epigenetic marks have been associated with activated vs. repressed genes. However, a lot of our knowledge is based on correlations, and determining the order of events and the causal players is not straightforward. We collaborate with Dr. O’Geen and Dr. Segal (University of California Davis) to decipher which epigenetic marks are essential for repression or activation of different genes and cell types using epigenome editing and genomic data analyses. For example, we showed that absence of promoter H3K27ac occupancy is a key determinant of heritable silencing and CRISPR-engineering DNA methylation (O’Geen, Nucleic Acids Research, 2022).
Importantly, many epigenetic marks are dramatically altered in cancer (and other diseases), and we are interested in understanding which alterations have causally contributed to the disease development and in which way. We study the epigenome development in cancer and its pre-cancerous lesions, with the aim to understand the contribution of different epigenetic aberrations to carcinogenesis and to develop biomarkers for early detection and disease progression.