Schizophrenia (SCZ) and bipolar disorder (BP) are the major adult psychotic disorders;uncertainty about their relationship is a central issue in psychiatry. By discovering variants with a high impact on either or both disorders (or on their endophenotypes) we can transform our understanding of their biology. This multisite project, from investigators with a track record of successful collaboration, will leverage exceptional, extensively phenotyped pedigree and population samples and integrate a combination of bioinformatics and experimental genomics approaches to identify such variants and demonstrate their relationship to these diseases. We will proceed through three steps: (1) Detect novel variants by whole genome sequencing (WGS) of discovery samples from ethnically homogenous populations;(2) Prioritize genome regions for further studies of these variants, using bioinformatics and functional genomics analyses;(3) Identify disease related variants through (a) imputation-based association analyses in very large case/control samples from the same populations and (b) validation of associated and predicted-deleterious variants using novel high-throughput functional assays. The WGS discovery sample includes SCZ, BP, and control individuals from Finland (FIN) and the Netherlands (NL). We will analyze their variants using a pipeline already implemented by our group to analyze the completed WGS of BP pedigrees from recently bottlenecked Latin American founder populations. The majority of our WGS samples derive from such founder populations;as we have shown previously this creates a high probability of detecting deleterious variants in these samples. WGS will detect a huge number of variants, most of unknown functional significance. We will thus narrow our focus to regions (coding and non-coding) most likely relevant to SCZ and BP, using two approaches: (1) Bioinformatics to prioritize regions previously implicated in these disorders (e.g. GWAS or CNV loci, or variants highlighted in prior sequencing studies including that of our BP pedigrees) and candidate regulatory regions (through analyses of ENCODE or other reference data);(2) Sequencing-based assessments of transcriptomic and epigenomic variation that we will conduct in two unique samples relevant to our phenotypes: fibroblasts from affected and unaffected members of our BP pedigrees;and neuronal and neuronal progenitor cells from reference individuals. We will then leverage very large, genotyped case/control samples available from the same populations as our discovery samples (or closely related ones) to accurately impute the novel variants from our prioritized regions;this will enable us to identify associations with SCZ, BP, and/or endophenotypes. Finally, in an initial validation study, we will use a novel high-throughput sequence-based reporter assay to evaluate the function of putative regulatory variants highlighted by our bioinformatics pipeline or by our association analyses.
Uncertainty about the genetic relationship between bipolar disorder and schizophrenia has been a major obstacle to understanding the biological underpinnings of these conditions. Through whole genome sequencing of cases and controls, genome-level experimental studies of gene function, and a series of bioinformatic and statistical analyses, this project will elucidate this relationship by identifying genetic variants with a high impact on one or both disorders.