Skip to main content

Big Data Analytics continue to impress cancer research

All News

(Vienna, 20 February 2019) Cells in healthy tissues are programmed to divide, replicate and enter natural cell death (apoptosis) based on various gene regulation and repair mechanisms. This is also true for cancerous tumors. However, certain gene mutations hamper apoptosis in tumor cells, thus allowing them to proliferate uncontrollably.

One specifically well-researched example are the BRCA1/2 genes, which suppress tumor formation in the breast and are involved in double strand DNA repair. Cells with mutated BRCA1/BRCA2 genes have to rely on other mechanisms to ensure error-free DNA repair and consequent cell division and replication. One such alternative mechanism involves the use of Poly(ADP-ribose) polymerase (PARP) proteins. In reference to fruit fly experiments carried out in the 1940s, researchers have found that if a cell with BRCA1/2 mutation additionally loses its ability to use PARP proteins to repair its DNA, it will become nonviable. However, the sole incapacitation of just PARP proteins in non-mutated cells does not result in cell death. “We call this gene interplay synthetic lethality”, explains Michael Krainer from the Department of Medicine I at Medical University of Vienna/Vienna General Hospital.

In recent years, scientists have exploited the BRCA1/2 and PARP synthetic lethal pathway to develop PARP-inhibiting compounds that selectively destroy not only breast tumor cells, but also cancerous cells in other tumor entities. For example, in prostate cancer, mutations in BRCA or other members of the double strand repair pathway have been targeted to develop various individualized and selective therapies. Krainer has played an instrumental role in this field - starting with the first ever investigation of BRCA germline mutations in the general population in the 1990s, to being principal investigator in international clinical trials testing various PARP-inhibitors in prostate cancer today. “In general, PARP inhibitors are well tolerated as compared to chemotherapy”, says Krainer.

During the past decades researchers have employed yeast as genetic screening tools to experiment and discover further, novel synthetically lethal gene combinations and deposited these findings in public databases. Big Data Analysis and Artificial Intelligence make it possible to use these data through the homology of yeast genes to human genes to identify clinically meaningful synthetic lethal interactions in humans. “The advances in machine-learning and network biology (in silico methods) enables the use of these databases and provide clear paths for most efficient experimental testing”, says Krainer.

Starting in November 2018, Krainer’s multidisciplinary oncology research group published exciting results. Widening the field of Systems Biology, their experiments conceptualized around synthetic lethality, demonstrate how machine learning combined with in-vitro yeast experiments can be harnessed to verify the efficacy of known low-toxicity breast cancer- treating drugs. In collaboration with the Vienna-based Computational Biology firm Emergentec Biodevelopment GmbH, a computational workflow embedded in the “e.valuation” technology platform suggested novel, low toxicity drug combinations which were observed to cause preferential death of tumorous cells. “Information on gene targets in yeast is taken to the human genome level, followed by inference on synthetic lethal interactions. This is ground-breaking not only for cancer therapeutics, but also in the fields of computational methodology, says Maximilian Marhold, first author of the study, proudly. Last week, the research group published similar work targeting ovarian cancer. “Big data promises the cost efficient development of novel, highly targeted drug combinations with favorable toxicity profiles not only in breast and ovarian cancer, but also in other tumors.”


Original Publications:
Marhold M, Tomasich E, Schwarz M, Udovica S, Heinzel A, Mayer P, Horak P, Perco P, Krainer M (2018). Synthetic lethal combinations of low-toxicity drugs for breast
cancer identified in silico by genetic screens in yeast - – Oncotarget - 9(91):36379-36391.

Heinzel, A., Marhold, M., Mayer, P., Schwarz, M., Tomasich E., Lukas, A., Krainer, M., Perco, P. (2019). Synthetic Lethality guiding selection of drug combinations in ovarian cancer – PLOS-One. Doi: