example, in the largest study of alcohol consumption (21), the alcohol consumption PRS accounted for only ~2.5% of the variance in alcohol use in two independent datasets. Recent work suggested that predictions may improve by incorporating functional genomic information. For example, McCartney et al (58) showed that, compared to conventional PRS, risk scores that took into account DNA methylation were better predictors of alcohol consumption (12.5% vs PRS 0.7%; but see (59)). Nonetheless, the way in which such methods can be used for prevention or treatments of AUD has yet to be established. Lastly, it remains to be determined the nature of these associations. Mendelian randomization analyses can serve to further understand and explore the correlations between alcohol use behaviors and comorbid traits (see Supplemental 1).
