Genome-wide association study of MRI markers of cerebral small vessel disease in 42,310 participants

Abstract

Cerebral small vessel disease is a major cause of stroke and dementia, but its genetic basis is incompletely understood. We perform a genetic study of three MRI markers of the disease in UK Biobank imaging data and other sources: white matter hyperintensities (N = 42,310), fractional anisotropy (N = 17,663) and mean diffusivity (N = 17,467). Our aim is to better understand the disease pathophysiology. Across the three traits, we identify 31 loci, of which 21 were previously unreported. We perform a transcriptome-wide association study to identify associations with gene expression in relevant tissues, identifying 66 associated genes across the three traits. This genetic study provides insights into the understanding of the biological mechanisms underlying small vessel disease.

Fig. 1

Genome-wide association Manhattan plots for WMH, FA and MD. Manhattan plots are shown for each of the phenotypes: WMH (top), FA (middle), and MD (bottom).

Fig. 2. Comparison of association signals across WMH, FA and MD.

Gene annotation was perfomed with PhenoScanner. ^*The lead SNP and/or proxies lie in an intergenic region. PP is the posterior probability of colocalization.

Fig. 3. Genetic correlations between WMH, FA and MD traits and other traits.

The symbols * and † indicate significant correlated traits by Bonferroni and FDR multiple testing correction respectively.

Fig. 4. PhenoScanner disease/trait annotation for WMH, FA and MD loci.

Only loci with a disease/trait annotation appear on this figure. For better visibility, we assigned each trait to one category with a specific color.

Fig. 5. TWAS and COLOC results for WMH, FA and MD.

COLOC.PP4 is the posterior probability of the hypothesis 4 in colocalization analysis, meaning there is association with the imaging biomarker trait and the gene expression and with one shared SNP.