We performed a sensitivity analysis testing whether the association of the metaGRS with IS was affected by familial relatedness in the validation set. Relatedness analysis was done using KING53 v2.1.4, based on ~784,000 autosomal SNPs measured on the Axiom chip, identifying n = 336,643 participants in the UKB validation set with kinship more distant than that of 2nd degree. There was a negligible difference in the association between the metaGRS and stroke in the full UKB validation set and within this distantly related subset of individuals.
