Collectively, these analyses demonstrate that much of the polygenic liability to alcohol use and misuse remains unaccounted for – we anticipate that much larger GWAS of phenotypes representing disordered drinking, in conjunction with other risk indices (e.g., PRS for negative affect, a hallmark of later stages of AUD) will be necessary to explain additional variance. Importantly, as the use of PRS becomes increasingly common in attempts to dissect the transitions from experimentation, to regular or problem use and further, to AUD, differences between the discovery sample and target sample(s) in age, ascertainment, and other characteristics will need to be considered in the interpretation of findings. In addition, our findings highlight the high degree of heterogeneity and polygenicity underlying alcohol use and misuse – not only is there no “gene for” alcoholism, there is currently no robust “polygenic indicator for” clinical prediction of liability to problem drinking. Furthermore, while we focused on contrasting the genetic relationship of each individual AUDIT subscale with the variety of alcohol-related measures across our different samples, it should be noted that the proportion of variance explained
