The Role of GWAS and GRS in T1D

Learning Objectives

• Discuss how genome-wide association studies can inform polygenetic risk scores

Summary

This is a background polygenic risk summary slide highlighting that for several decades we've been aggregating cohorts of thousands of cases and thousands of controls and testing the associations of many common genetic variants with disease.

The example on the left-hand side here is a Manhattan plot where you see these peaks across the genome, which essentially are significant associations of genetic loci with disease.

This has been interesting in terms of identifying potential mechanistic targets and validating mechanistic findings from other types of scientific study.

However, those individual association peaks don't really go with strong genetic risk on their own for a condition, they only contribute a small amount of risk. So, if you want to use these to identify people at high risk you have to add them all together, create a polygenic risk score and then place a population of people on a distribution from low to medium to high. You can see high-risk people for cardiovascular disease being in the minority up here.

Type 1 diabetes is not like most common complex traits. It has genetic associations across the genome which are important and mechanistically important to new drug discovery and understanding the mechanism of type 1 diabetes.

But the majority of risk is accounted for in the HLA Class 2 region by a small number of HLA Class 2 alleles. You can see that from that enormous association peak. This is handy because we can measure this really quite simply and sometimes cheaply using a small number of single nucleotide polymorphisms.

MAT-GLB-2407833-1.0 - 12/2024