Genetic Colocalisation Across Diseases to Identify Drug Repurposing Candidates and Risk of Patient Co-Morbidities

The process of drug discovery is notoriously expensive and time-intensive, with a high rate of failure. In this context, drug repurposing-finding new therapeutic uses for existing approved medications-presents a promising, faster, and more cost-effective alternative. This research aims to identify shared genetic mechanisms across various diseases, thereby uncovering potential therapeutic targets and candidates for drug repurposing.

By integrating genetic association results with gene expression and proteomics data, the project seeks to improve drug repurposing.

The project will leverage large-scale genomic datasets, including those from the UK Biobank and the 'All of Us' Biobank. By employing genome-wide association studies (GWAS) and Bayesian colocalization analyses, the research will identify shared causal variants across diseases. These variants will then be correlated with gene expression data (eQTL, pQTL) to elucidate the biological mechanisms underlying these conditions, highlighting potential targets for new therapies.

In addition to identifying drug targets, this research will incorporate machine learning models to stratify patients based on their genetic predisposition to co-morbidities. By creating polygenic risk scores (PRS) from shared genetic variants, the models will predict the likelihood of patients developing multiple interconnected conditions. This approach enables more precise patient stratification, facilitating better-targeted treatments and enhancing outcomes in personalized medicine.

Ultimately, the goal is to not only repurpose existing drugs more effectively but also to contribute to a more tailored approach to patient care, improving the management of co-morbidities through insights derived from genomic data.