used as in all other analyses (age, gender, generation, recording system, and the 10 PCs from EIGENSTRAT). Because LD can bias SNP heritability estimates upward (Speed, Hemani, Johnson, & Balding, 2012), we repeated these analyses after weighting SNPs by local LD patterns using LDAK software (http://dougspeed.com/ldak). Yang and colleagues have more recently recommended using the entire sample when it consists of closely related subjects, and estimating the magnitude of genetic influence while simultaneously modeling the environmental influences family members share (the C latent variable in biometric models). This provides an estimate of genetic influence unconfounded by shared environmental effects (as well as an estimate of such effects). In addition to this, we conducted the same analysis without modeling shared environmental influences (i.e., without any threshold of genetic relatedness). The difference between the two provides a simple indication of the magnitude of such effects (which is also estimated directly by GCTA).
