3/7 Psychiatric Genomics Consortium: Finding Actionable Variation

The goal of this study was to generate new therapeutic hypotheses by targeted sequencing of 161 genes most likely to harbor schizophrenia (SCZ) causal variants in multi-ancestry schizophrenia case-control cohorts. The gene panel was intended to contain putative SCZ risk genes, from within which, the majority of new discoveries from additional WES/WGS would come. To this end, we integrated current knowledge of the genetic architecture of SCZ (results from The Schizophrenia Exome Sequencing Meta-Analysis Consortium, or SCHEMA) and gene membership in gene sets that have been implicated in SCZ using a Bayesian framework. A total of 23,352 individuals selected to be non-overlapping with SCHEMA, as well as other previous and ongoing sequencing efforts in the field, were identified from over 20 individual study sites. Sequencing of the exonic regions of 161 prioritized genes was performed on the Ion Torrent platform in between June 2018 and April 2019. Sequencing plates were matched with respect to ancestry and case/control composition whenever possible. The average sequencing depth across all samples was 224x. Single sample calling was performed using Torrent Variant Caller v5.8.0, which is specially optimized to exploit the underlying flow signal information generated by the Ion Torrent sequencing. The schizophrenia case-control status was obtained from each original study site. ]]>
We designed a custom sequencing panel of 161 putative schizophrenia (SCZ) genes and applied it to case-control cohorts totaling 22,135 unrelated individuals from diverse ancestries. The exonic regions of these prioritized genes were sequenced on the Ion Torrent platform. Meta-analysis were performed with SCHEMA data and sequencing datasets for bipolar disorder and autism. We performed two broad types of analysis: (1) a global enrichment of all constrained genes on the custom panel (n = 80 genes) to investigate the overall role of rare disruptive variants in diverse ancestries and (2) gene-level burden tests to identify novel SCZ risk genes.To ensure compatibility with Psychiatric Genomics Consortium (PGC) definitions, cases were defined as those having a diagnosis of schizophrenia or schizoaffective disorders, and controls were individuals without a known diagnosis of a psychiatric disorder.]]>