Weighted burden analysis in 200,000 exome-sequenced subjects characterises rare variant effects on risk of type 2 diabetes

Type 2 diabetes (T2D) is a disease for which both common genetic variants and environmental factors influence risk. A few genes have been identified in which very rare variants have large effects on risk and here we carry out a weighted burden analysis of rare variants in a sample of over 200,000 exome-sequenced participants in the UK Biobank project, of whom over 13,000 have T2D. Variant weights were allocated based on allele frequency and predicted effect, as informed by a previous analysis of hyperlipidaemia. There was an exome-wide significant increased burden of rare, functional variants in three genes, GCK, HNF4A and GIGYF1. GIGYF1 has not previously been identified as a diabetes risk gene but its product is plausibly involved in the modification of insulin signalling. A number of other genes did not attain exome-wide significance but were highly ranked and potentially of interest, including ALAD, PPARG, GYG1 and GHRL. Loss of function (LOF) variants were associated with T2D in GCK and GIGYF1 whereas nonsynonymous variants annotated as probably damaging were associated in GCK and HNF4A. Overall, fewer than 1% of T2D cases carried one of these variants. In two genes previously implicated in diabetes aetiology, HNF1A and HNF1B, there was an excess of LOF variants among cases but the small numbers of these fell well short of statistical significance, suggesting that even larger datasets will be helpful for more fully elucidating the contribution of rare genetic variants to T2D risk. This research has been conducted using the UK Biobank Resource.