Leveraging large-scale biobanks for therapeutic target discovery

Area of Science:

• Genetics
• Pharmacology
• Bioinformatics

Background:

• Large-scale biobanks like the Million Veteran Program (MVP), UK Biobank, and FinnGen offer genetic data for over a million individuals.
• Understanding gene-trait associations is crucial for identifying novel drug targets and elucidating mechanisms of existing ones.

Purpose of the Study:

• To harmonize data from major biobanks and perform two-sample Mendelian randomization (MR) to identify causal gene-trait relationships.
• To evaluate the potential of identified gene-trait pairs as drug targets and develop a predictive model for drug development success.

Main Methods:

• Harmonized genetic association results from over 1 million individuals across multiple biobanks.
• Conducted two-sample Mendelian randomization (MR) using gene expression (GTEx, eQTLGen) and plasma protein levels (ARIC, Fenland, deCODE) as proxies for target modulation across 2,003 phenotypes.
• Developed a predictive ranking model trained on approved drug targets (ChEMBL 34) and biological annotations.

Main Results:

• Identified 69,669 gene-trait pairs with evidence for causal effects (p ≤ 1.6 × 10^-9), including 6,447 genes with strong evidence for at least one trait.
• Gene-trait pairs identified were significantly associated with higher odds of being approved drug targets and indications, with 9% of approved targets rediscovered.
• The predictive model achieved a precision-recall area under the receiver operating characteristic curve of 0.79 for predicting drug development success and clinical indication.

Conclusions:

• This study provides a comprehensive resource of gene-trait causal relationships, significantly enhancing the identification of potential drug targets.
• The findings validate the utility of large biobanks and MR in drug discovery and provide a predictive tool to prioritize targets for pharmaceutical development.
• The publicly available results in CIPHER facilitate further research and accelerate the translation of genetic discoveries into clinical applications.