Scientists at the Broad Institute and Harvard Medical School (HMS) have published results from an exome-sequencing study of 20,791 patients with type 2 diabetes (T2D) and 24,440 non-diabetic controls representing five ancestries. This study found gene-level associations of rare variants (having minor allele frequencies of less than 0.5%) in four genes including a series of more than 30 SLC30A8 alleles that appear to be protective against T2D.
The results of this potentially pathbreaking study were reported in Nature on May 22, 2019. The authors describe it as “one of the largest analyses of exome-sequenced cases for T2D, specifically, and for any disease, more generally.”
Finding variants associated with complex diseases such as T2D is extremely challenging. Common-variant array-based genome-wide association studies (GWAS) have suggested thousands of genomic loci are possibly linked to hundreds of human traits. However, to date it appears that heritability of most complex traits arises from common regulatory variants with modest effects, and some of which need to occur together. Uncovering such variants is a priority for those seeking to identify patients at risk for T2D as well as those searching for new drug targets.
One of the main challenges for scientists doing such studies is getting large enough cohorts. The largest studies until now have included less than 10,000 disease cases. That falls well short of the number needed to uncover rare disease-associated variants.
For T2D, the strongest gene-level signals for rare variants found to date explain at most 25% of the heritability of the disease. In this latest study, the Broad/HMS team observed gene-level effect sizes they estimate would require 75,000–185,000 sequenced cases to achieve exome-wide significance. They also propose a method to interpret modest rare-variant associations and to incorporate them into target- or gene-prioritization efforts.
A key part of this field of research will be understanding how exactly variants impact risk. The Broad/HMS authors explain that “Protein-coding variants with strong effects on protein function or disease can offer molecular ‘probes’ into the pathological relevance of a gene and potentially establish a direct causal link between gene gain- or loss-of-function and disease risk—especially when there is evidence of multiple independent variant associations (an ‘allelic series’) within a gene.”
This research should help advance the use of exome-sequencing studies to identify significant rare-variant associations for complex diseases. Uncovering these variants in T2D is particularly important, since the number of people with diabetes has quadrupled in the last three decades, making it the ninth major cause of death globally. Of the people diagnosed with diabetes, 90% have T2D.