Improving Polygenic Prediction for Psychiatric Disorders for Clinical Use

The Lancet Commission on global mental health highlighted the importance of understanding the nature integrating genetics, neuroscience, and clinical disciplines in mental health research. The proposal is highly relevant.

Using a novel approach integrating multiple types of genomic data from large-scale cohorts in the US, UK, and Sweden, this project aims to better understand the genetic architecture of psychiatric disorders (primarily schizophrenia and major depressive disorder, MDD) in service of improving polygenic prediction to inform clinical practice.

Specifically, 1) to develop a variant prioritization strategy to identify genetic variants associated with MDD that are most likely functional and explicitly taking admixed genetic ancestries into consideration by leveraging genomic and multi-omics data from a diverse cohort in the US. 2)applying the variant prioritization strategy to construct novel polygenic risk scores (PRS) for MDD and identify the most predictive PRS, using genotype and whole genome sequencing data from UK Biobank. 3) to assess the role of the enhanced schizophrenia and MDD PRS, respectively, on illness prediction and sub-clinical group identification in Swedish cohorts.

Results from this project will reduce health disparities and improve the quality of PRS to be introduced into clinical practice for MDD and schizophrenia which could further aid diagnostic refinement, clinical trajectory prediction, and novel therapeutic target discovery.