Skip to navigation Skip to main content Skip to footer

Approved Research

Protein Protein Interaction Networks Systematically Derived from Large Scale Human Genetic Data

Principal Investigator: Dr William Chen
Approved Research ID: 72821
Approval date: November 30th 2021

Lay summary

Drug discovery relies on experimental models which often don't translate well into efficacious medicines.  UK Biobank makes possible the building of a robust human basis for drug discovery by focusing attention on human genes and disease traits.  As part of doing so, we need decipher the "wiring diagram" of genes which specifies the way they interact and lead to disease.  The wiring diagram is complex because it involves tens of thousands of genes interacting with each other.  One way to make sense of the wiring diagram is to identify smaller organizational units that control subsets of traits.  We will combine the UK Biobank with other types of data to identify these organizational units and their related traits.

We will use the rich genetic data in the UK Biobank to mark important regions of the genome as it relates to disease-related traits including those of autoimmune, neurological, metabolic, and cardiovascular origin.  Because volunteers in the UK Biobank have provided other kinds of measurements that relate to disease-related traits such as lab measurements and survey data, we are able to use genetic information from these traits to divide the wiring diagram of the disease into smaller organizational groups.  We will combine the genetic information with wiring information obtained from another set of experiments.  Importantly once the map is made, we will use another data sources, the loss-of-function genetics from UK Biobank and animal knockout data, to test whether our findings make sense. 

The project relies on three elements: the curation of very large human genetic and disease data sets in the UK Biobank (6 months), application of advanced computational algorithms to human data to generate trait-specific gene wiring diagrams (12 months), and validation of the maps with other informatics and experimental data in the public domain (6 months).

The maps generated from this study increase our understanding of human biology by clarifying the way groups of genes work together leading to disease.  The new knowledge reduces our reliance on animal models for explaining human disease and replaces such models with human data.  Historical analyses show that drug discovery aided with such human genetics data see increased rates of success in making an impact on patient outcomes.