Many factors have been proposed as contributors to risk of alcohol abuse, but quantifying their influence has been difficult; here a longitudinal study of a large sample of adolescents and machine learning are used to generate models of predictors of current and future alcohol abuse, assessing the relative contribution of many factors, including life history, individual personality differences, brain structure and genotype. Many factors have been identified as contributors to risk of alcohol abuse but their relative importance has been difficult to quantify. Robert Whelan et al. constructed models of current and future adolescent binge drinking using data from the IMAGEN project, a study of risk-taking behaviour in more than 2,000 teenagers recruited at age 14 from the United Kingdom, Ireland, France and Germany. The authors used machine learning to generate models of predictors of current and future alcohol abuse, assessing the contribution of many factors including life history, individual personality differences, brain structure and genotype. A key finding of the study was that personality factors were, surprisingly, not particularly useful predictors of future alcohol misuse. In contrast, neurodevelopmental immaturity, certain structural and functional indicators in the brain, sexual experience and prenatal alcohol exposure were associated with current and future binge drinking. A comprehensive account of the causes of alcohol misuse must accommodate individual differences in biology, psychology and environment, and must disentangle cause and effect. Animal models1 can demonstrate the effects of neurotoxic substances; however, they provide limited insight into the psycho-social and higher cognitive factors involved in the initiation of substance use and progression to misuse. One can search for pre-existing risk factors by testing for endophenotypic biomarkers2 in non-using relatives; however, these relatives may have personality or neural resilience factors that protect them from developing dependence3. A longitudinal study has potential to identify predictors of adolescent substance misuse, particularly if it can incorporate a wide range of potential causal factors, both proximal and distal, and their influence on numerous social, psychological and biological mechanisms4. Here we apply machine learning to a wide range of data from a large sample of adolescents (n = 692) to generate models of current and future adolescent alcohol misuse that incorporate brain structure and function, individual personality and cognitive differences, environmental factors (including gestational cigarette and alcohol exposure), life experiences, and candidate genes. These models were accurate and generalized to novel data, and point to life experiences, neurobiological differences and personality as important antecedents of binge drinking. By identifying the vulnerability factors underlying individual differences in alcohol misuse, these models shed light on the aetiology of alcohol misuse and suggest targets for prevention.