Kinreich S, Meyers JL, Maron-Katz A, et al. - Researchers sought for predictive models that may assist in identifying the individuals who are prone to develop Alcohol use Disorder (AUD) and for the biomarkers indicating a predisposition to AUD. From the Collaborative Study of the Genetics of Alcoholism (COGA), they included 656 participants including offspring and non-offspring of European American (EA) and African American (AA) ancestry who were recruited as early as age 12 and were unaffected at first assessment and reassessed years later as AUD (DSM-5) (n = 328) or unaffected (n = 328). They performed machine learning analysis for 220 EEG measures, 149 alcohol-related single nucleotide polymorphisms (SNPs) from a recent large Genome-wide Association Study of alcohol use/misuse and two family history (mother DSM-5 AUD and father DSM-5 AUD) features. The AA compared with the EA prediction models had significant and higher accuracy rates and females vs males had a higher model accuracy trend for both ancestries. Combined EEG and SNP features model was identified as superior to models based on only EEG features or only SNP features for both EA and AA samples. In both ancestry samples, accuracy score was higher for the youngest age group vs the two other older age groups. The model’s accuracy enhanced with integration of maternal AUD in both ancestries’ samples. They identified several discriminative EEG measures and SNPs features, including lower posterior gamma, higher slow wave connectivity (delta, theta, alpha), higher frontal gamma ratio, higher beta correlation in the parietal area, and 5 SNPs: rs4780836, rs2605140, rs11690265, rs692854, and rs13380649. Outcomes thereby emphasize that sampling uniformity followed by stratified (eg, ancestry, gender, developmental period) analysis, and wider selection of features, are significant to develop better prediction scores that may allow a more accurate estimation of AUD development.