This proposed project aims to use genetics to help develop an approach for classifying severe mental illness (SMI) that has a stronger scientific foundation than the systems currently used in both research and clinical practice. These classification systems have, for more than a century, divided the bulk of SMI into dichotomous diagnostic categories: psychotic disorders (including schizophrenia [SCZ]) and mood disorders (including bipolar disorder [BP] and major depressive disorder [MDD]). However the overlap of symptomatology across mood and psychotic disorders, and growing evidence for the genetic correlation between these categories, demonstrate that they imprecisely represent the biological underpinning of SMI. It has been proposed that frameworks based on symptom-level and dimensional (quantitative) information, such as the NIMH Research Domain Criteria (RDoC), would better reflect the genetic contribution to SMI and would therefore provide a more useful framework for their classification. However the evidence supporting this hypothesis remains sparse, in large part because we lack the right datasets to test it. In this project we will generate a unique SMI dataset, using electronic health records to ascertain individuals who have received inpatient treatment at a single psychiatric hospital that serves the entire 1 million inhabitants of the state of Caldas, Colombia. All of the individuals whom we will investigate are members of the ?Paisa?, a genetically and culturally homogeneous population that comprises the majority in this region of Colombia. By recruiting 8,000 participants across the full range of severe mood and psychotic disorders (as well as 2,000 demographically-matched controls); performing uniform phenotyping of these 10,000 individuals using diagnostic and quantitative assessments; and genome wide genotyping, we will establish dimensional phenotypes that index core deficits of SMI and that reference multiple RDoC domains. We will then conduct genetic analyses of symptom-level and quantitative phenotypes, evaluating their relationship to known SMI loci and to polygenic risk scores (PRS) that represent the overall contribution of common genetic variation to these disorders; the SCZ, BP, and MDD workgroups of the Psychiatric Genomics Consortium (PGC) will provide us with up-to-date genetic data for each diagnosis. Additionally, we will conduct genome wide association analyses of the quantitative traits, including meta-analyses for traits that have been assessed in other study populations. We will also contribute our data (including genotypes available to us for an additional 6,000 Paisa controls) to the case-control meta-analyses of the PGC workgroups, contributing to the diversity of their datasets by adding a substantial number of samples from a previously underrepresented (Hispanic) population. .
Current classification systems split severe mental illness (SMI) into distinct diagnostic categories, although both clinical and genetic evidence suggest that a dimensional classification system would rest on a stronger scientific foundation. We will analyze common genetic variation in a Colombian sample of 8,000 extensively phenotyped SMI individuals, ascertained across diagnostic categories, and 2,000 demographically matched controls. By doing so, we will attempt to identify symptom-level and quantitative phenotypes that demonstrate a stronger genetic relationship to SMI than do the diagnostic categories themselves.
