Research

A prime challenge in the development of precision health care strategies is to understand how genetic variants or chemical perturbations that individually have only a modest contribution to disease risk or treatment response may lead to strong synergistic effects on disease progression, treatment efficacy or toxicity when combined. While such epistatic or synergistic interactions play a role in many diseases, including cardiometabolic diseases and cancer treatment, their systematic identification has remained difficult due to complex networks underlying genotype-phenotype relationships.

Research strategy

Our research group has expertise in network-centric and machine learning-based approaches to modeling and predicting complex relationships between genetic and functional dependencies and medical phenotypes such as susceptibility to diseases and responses to treatments. We are using both 'reverse-genetic' approaches, including CRISPR-Cas9 and drug screening, as well as 'forward-genetic' approaches, such as next-generation sequencing. We believe that combining functional and genetic profiling will provide a more comprehensive network view of the mechanisms behind disease processes and enable accurate predictions of system-level phenotypic responses to genetic and chemical perturbations.

In the, we are implementing mathematical models to identify target addictions and other druggable vulnerabilities in individual cancer patients. Using a systems pharmacology approach, we make use of polypharmacological effects of drugs to inhibit cancer survival pathways and sub-clones in terms of multi-target synergistic and synthetic lethal interactions. Machine learning modelling offers a unique potential to identify key players and their interaction partners in disease networks, as well as to suggest more effective and selective targets for personalized therapies. Toward clinical translation, these approaches will improve both safety and efficacy when implementing the precision health care strategies.