For polygenic scores, we first compared those with AUD who had reported a suicide attempt to those with AUD who had not reported a suicide attempt across all PGSs, independently, using logistic regression in R (version 4.2.1). Second, to ensure that results within those with AUD were not biased by conditioning on AUD (Akimova et al., 2021), we also compared: 1) those with AUD who had a reported suicide attempt, 2) those with AUD who had not reported a suicide attempt, and 3) those without AUD who had a reported suicide attempt to those who neither reported a suicide attempt nor meet criteria for AUD (see Supplemental Table 1 for sample description) using a multinomial logistic regression model in the nnet package in R (Ripley & Venables, 2022). In both analyses, we included sex, age, the first six genetic principal components (PCs), genotype array, and birth cohort as covariates. To adjust for familial clustering, we used cluster robust standard errors (Colin Cameron et al., 2011; Davenport et al., 2011). We stratified analyses by genetic similarity and then meta-analyzed results (by
