Statistical ‘omics approaches to understanding autoimmune diseases

A DNA spelling mistake that raises the risk of juvenile arthritis or type 1 diabetes often leaves no clue about which gene or cell type it actually harms. Researchers have mapped hundreds of these risk variants across the genome, but the chain of cause and effect from a single letter change to a malfunctioning immune system remains mostly invisible. This project uses statistical methods to bridge that gap. By combining genetic data from ten different autoimmune diseases with maps of gene activity in specific cell types, the researcher aims to pinpoint which genes and which cells are the real drivers of disease. The same approach will then group risk variants into the biological pathways they disrupt, and test whether patients can be sorted into subgroups based on which pathways are most affected. If such subgroups exist, they open the door to stratified medicine—matching a patient to a drug based on the specific genetic pathway driving their disease, rather than treating all patients with the same diagnosis identically. This is fundamental science with a clear translational target: turning statistical signals into clinically actionable patient categories.