To assess the utility of PRS in prediction of developing comorbid disorders, we performed Cox’s regression analysis on the depression case group with and without additional diagnosis of anxiety, bipolar disorder, schizophrenia and/or SUD. In addition to PRSs for the individual disorders, we included 4 aggregate scores, one for each disorder, combining all PRSs into a single score weighted by their association with anxiety, bipolar disorder, schizophrenia, and SUD respectively (see Methods).
