Genetic data integration for drug target prioritisation

Although >90% of new drug candidates fail in clinical trials, drug targets with human genetic evidence are 2x as likely to succeed. However, the rate of genetic discoveries exceeds our ability to prioritise drug targets. The main challenges are threefold: we struggle to distinguish shared causal signals from spurious overlaps; we cannot effectively link individual signals to specific targets and characterise their function; and we lack the tools to assess if desired modulation of the target is achievable to have an impact on disease. To address these challenges, we will develop a novel colocalisation method to systematically identify all shared genetic signals at scale. We will annotate these signals using state-of-the-art genetic variant effect prediction models. Finally, we will develop a machine learning model to predict the mode-of-action of genetic variants. These advances will significantly reduce the time and effort required to perform genetic drug target prioritisation.