PROJECT ABSTRACT Polycystic ovary syndrome (PCOS) affects up to 20% of women and is a leading cause of infertility, yet its etiology remains poorly understood. This is in part due to its complex architecture, which is driven by genetic and environmental risk factors that lead to symptom heterogeneity among PCOS patients. Currently, PCOS is only diagnosed by the presence of polycystic ovary morphology, hyperandrogenism, and oligo- or anovulation, despite the common presence of numerous metabolic comorbidities, insulin resistance, and cardiovascular diseases. Undiagnosed and untreated PCOS can have detrimental long-term health consequences, especially among high risk groups. This includes minority populations who experience a greater PCOS burden due to their increased risk of developing metabolic disorders that can be exacerbated by age, body mass index, and socioeconomic status. My project aims to characterize PCOS etiology and comorbidity patterns in underrepresented populations. We hypothesize that both population specific genetic factors and socioeconomic factors contribute to increased PCOS risk and severity in African American and Hispanic women. Our preliminary data show that the clinical spectrum of PCOS symptoms among diverse patients can be captured through electronic health record (EHR) phenotyping algorithms. Furthermore, the hormone profile of PCOS cases worsened as the stringency of the PCOS definitions increased, especially among African American and Hispanic women. We propose to test our hypothesis through two specific aims using EHR-based methods. Aim 1 will systematically identify PCOS comorbidity patterns associated with EHR-reported race and ethnicity among women with mild to severe PCOS symptoms. Aim 2 will evaluate the independent and combined effects of genetic, clinical, and socioeconomic risk factors on PCOS diagnosis and symptoms. This work will improve knowledge of PCOS etiology and comorbidity, thus creating a framework for precision reproductive medicine among minority women.