Integrating case-control transcriptomic and genetic data in admixed individuals to identify disease genes for schizophrenia and bipolar disorder

PROJECT SUMMARY Genome-wide association studies (GWAS) have identified hundreds of genomic loci that increase disease susceptibility for schizophrenia (SCZ) and bipolar disorder (BP). We and others have shown that integration of functional molecular data, including transcript abundance levels, can be used to decipher causal insights from GWAS-identified allelic variant to gene to disease susceptibility. The identification of genetic risk factors and causal interpretations provide the greatest potential for the future of personalized genomic medicine. These studies and advances, however, have mostly been performed in individuals of European ancestry, thereby exacerbating health disparities of most vulnerable populations of diverse and admixed genetic ancestries. At this time, NIMH is leading the effort for genetic discoveries in underrepresented populations through funding of large- scale GWAS projects of SCZ and BP, but it is necessary to simultaneously increase the genetic diversity of functional molecular data for data integration and deciphering causal biological effects involved in SCZ and BP. We will tackle the lack of diversity in integrative transcriptomic studies for the deciphering of genetic susceptibility of SCZ and BP in Latino admixed populations, through deliberate integration of novel statistical method development with new transcriptomic data generation. We will generate transcriptome data in 1,500 Latino individuals (500 SCZ cases, 500 BP cases, 500 controls) as a public resource to generate new insights about the molecular mechanisms that contribute to risk of SCZ and BP. At the same time, we will develop new statistical methods that leverage the admixture process to improve the power of mapping genetic effects on expression in Latino individuals. We will incorporate the local and global genetic ancestries within admixed individuals to increase power of transcriptome-wide association (TWAS) and expression quantitative trait score (eQTS) studies to identify novel susceptibility genes. We will integrate new transcriptomic data in Latino individuals with ongoing Latino GWAS of SCZ and BP involving tens of thousands of Latino individuals both to generate new biological hypotheses and to validate our new statistical methods. The most promising causal variants and susceptibility genes will undergo functional validation. Our efforts will result in a large public gene expression resource, novel statistical tools, and new biological findings validated in functional assays. Our findings may yield clinically relevant applications for diagnosis and treatment for SCZ and BP particularly in Latino individuals that have been traditionally underserved in large-scale genomic studies.