The (i, j) element of K is the excess allele sharing for alleles drawn from individuals i and j, beyond that expected for unrelated individuals given the allele fraction estimates [37]. The eigenvectors of K are the principal components, used to diagnose and correct for population structure. The first principal component (PC1) showed some dispersion of individuals away from the main cluster (Figure 4(b)), which may correspond to admixture. This was the only one of the first five PCs that was significantly associated with neuroticism (p<10−4), explaining 0.5% of the phenotypic variance. From K we also identified 54 pairs of individuals with kinship estimate >0.05, which corresponds approximately to at least a cousin relationship in an outbred population (Figure 4(a)). Of these 54 kin pairs, 19 had kinships around 0.25, consistent with a first-degree relationship (sibling, or parent/offspring).
