This project aims to identify predictive markers of symptom change using assessments readily deployable in real-world settings, including EHRs, traditional clinical measures, computational behavioral tasks, short gamified measures of mood and reward, spoken narrative data for natural language processing analyses, and transdiagnostic outcome measures. Longitudinal data will be collected over two years from 2,400 individuals, enriched for psychopathology across a broad spectrum of traditional diagnoses, and analyses will seek to identify a minimum set of measures with maximum added clinical value, examine longitudinal clinical trajectories, and identify subgroups with different clinical outcomes over 2-years.