Of the one-third of individuals with AUD who attempt to quit drinking every year, only 25% are successful in reducing their consumption a year later. Therefore, understanding the parameters that can set an optimal initial state (including biomarkers, demographic, and medications) can greatly affect the success of remission from AUD. Using multimodal, multi-featured machine learning applications with the COGA longitudinal dataset, we uncovered these parameters, uniquely characterized per sex and ancestry. This is the first study to formulate a multimodal-based prediction model to determine AUD individuals who are going to be in remission from AUD. Results confirmed previous results showing that the combined feature model (e.g., EEG, PRS, medication, and demographic information) achieved a higher prediction score than models based on single domain suggesting that genetics prediction models will improve from the addition of phenotypes to the calculation. Intriguingly, results indicate higher accuracy scores for EA and AA males, when the ancestry was defined by genetics than by self-report for models with only PRS features. Several discriminative features were identified for each of the models revealing novel predisposition sex and
