A polygenic score (PGS) is the weighted sum of single-nucleotide variant (SNV) risk alleles across the entire genome and has shown promise in evaluating disease risks.26,27,28,29,30,31,32,33 SNVs and their weights are derived from discovery datasets, and PGSs are calculated and tested in target datasets. In our previous work, we used concordant SNV (ie, SNVs with the same directions of effects in different study cohorts or populations) strategies to calculate PGSs.24,34,35 These strategies exclude large numbers of irrelevant SNVs while retaining disease-associated SNVs, thereby substantially increasing the estimability of PGSs. In a population of European ancestry, PGS estimability was comparable to family history of AUD and was associated with AUD severity and remission from AUD.24 However, in that study, PGS was calculated and tested in a family cohort primarily ascertained for individuals with AUD.24
