Genomic architecture of bipolar disorder in Japan: Insights from genomic structural equation modeling

作者：Hiroki Kimura, Yutaka Nagasaki, Sawako Furukawa, Shiori Ogawa, Takeo Saito, Chikashi Terao, Nakao Iwata, Masashi Ikeda · 发表于：Psychiatry and Clinical Neurosciences · 年份：2025 · DOI：10.1111/pcn.13906 · 被引用次数：3 · 研究领域：Genetic Associations and Epidemiology、Bipolar Disorder and Treatment、Genetic Syndromes and Imprinting

Bipolar disorder (BD) is a genetically and clinically heterogeneous condition that shares part of its polygenic architecture with schizophrenia (SCZ) and depression.1 We previously found that while European BD subtype I (BD1) shows a stronger genetic correlation with SCZ, Japanese BD1 is more strongly correlated with major depressive disorder (MDD),2 suggesting that the genetic architecture of BD differs among populations. However, genetic correlations only quantify the association between traits; they do not assess causality or multivariate genetic architecture. By contrast, genomic structural equation modeling (GSEM) models the multivariate genetic architecture of a constellation of traits,3 and recent GSEM applications have identified common underlying factors across psychiatric disorders (e.g. a psychotic dimension, an internalizing dimension) that help explain the shared genetic architecture among these conditions.4 Therefore, to examine the multivariate genetic architecture of BD in the Japanese population, we performed GSEM using genome-wide association study (GWAS) summary statistics for SCZ, BD, and MDD from both European and East Asian populations (nine traits in total). These included updated public data as well as the Japanese BD cohort identical to that reported in our previous study (Table S1).2 Detailed methods and results are provided in the Supplementary Information. First, we estimated pairwise genetic correlations among these nine psychiatric traits using l...