Because no GWS findings were identified in our primary analysis of African ancestry we performed cross ancestry lookups in the summary statistics of European ancestry. Of 223 GWS SNPs from the European ancestry meta-analysis, 206 were available in African ancestry, 61% (n=125) had the same effect direction, 20 were nominally significant (p<0.05), and 1 SNP survived Bonferroni correction (Figure 5). This SNP that survived multiple testing correction (rs1950829 EUR p=7.24Ex10−19, AFR p=9.34×10−6), is in an intron of the “Leucine rich repeat fibronectin type III domain containing 5” (LRFN5) gene. This gene was previously detected in genome-wide gene- and pathway based analyses of depressive symptom burden conducted in three cohorts from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), the Health and Retirement Study (HRS), and the Indiana Memory and Aging Study (IMAS).37 As larger samples are collected for more diverse ancestry groups we expect to see more novel loci identified for non-European populations. Finally, we conducted a transancestral meta-analysis by combining studies of African and European ancestries in 1,213,867 participants, thereby identifying 233 independent SNPs and 183 risk loci. For now, transancestral analysis is a way to leverage results from understudied populations.
