Quantifying the relationship between drug targets and diseases using human genetics evidence

Quantifying the relationship between drug targets and diseases using human genetics evidence

Lay summary

This aim of this research project is to answer a question asked by every organization making new medicines: given a fixed budget, for which diseases and drug targets should we initiate a drug discovery program?

The UK Biobank provides a unique resource to help us answer this question by assimilating genetic information from more than 500,000 individuals together with the measurements of many traits of these same individuals.

For more than a decade a research design called a genome-wide association study (GWAS) has been used to identify associations between genetic mutations and a particular human trait, including disease-associated traits. A related research design, the phenome-wide association study (PheWAS), allows researchers to identify which traits are associated with a particular genetic mutation. Genes which are found to have many mutations that are associated with a disease in GWAS studies may be good targets for a drug discovery program, especially if those mutations are not observed to impact other diseases in PheWAS studies.

To translate the associations observed in GWAS and PheWAS studies of UK Biobank data into information that can be used to inform drug discovery, additional analyses must be performed to incorporate data sets from other sources. Other sources of data include quantitative summaries of results of GWAS and PheWAS studies performed on other large cohorts as well as data about the molecular, cellular, and tissue-level impacts of genetic mutations. These additional data sources will allow our researchers to move from an observed association between a genetic mutation and a disease to a therapeutic hypothesis of how we can make a medicine to treat that disease.

The scientific rationale for using human genetic evidence to inform drug discovery program selection derives from several published studies from academic and industry researchers. These researchers have examined the rate of approval of drug programs initiated with human genetic evidence versus those without and have found that drug programs are far more likely to succeed when they are supported by human genetic evidence.

The public health impact of our work will be primarily achieved through the development of new medicines using evidence we derive from our analysis of UK Biobank data. Additionally, we plan to make much of our work open source and publish our results in open access scientific journals, which may have an indirect impact on public health by making other researchers more productive in their quest to make new medicines.