Psychiatric disorders such as bipolar disorder and schizophrenia are known to have high heritability, but the molecular basis of their genetic factors has long been a mystery. Recent advances in genomics technology have made it possible to comprehensively analyze the information encoded in DNA, enabling us to elucidate genomic features associated with psychiatric disorders. The progress of genomic analysis technology has led to the discovery of rare variants associated with schizophrenia and autism with large effect sizes. However, the understanding of the genetic architecture of bipolar disorder, especially that of rare variants with large effect sizes, relatively lags. Thus, we focused on de novo mutations, which include highly rare variants, to detect rare variants associated with bipolar disorder with large effect sizes. We searched for de novo mutations from 354 trios with bipolar disorder, the largest sample size in a trio-based bipolar disorder study to date (Nishioka, et al., Nature Communications, 2021). In this study, we found the following results: (i) de novo mutations in bipolar disorder are characterized by loss-of-function mutations in genes susceptible to natural selection by loss-of-function mutations (high pLI genes, high probability of loss-of-function intolerance), and (ii) de novo mutations in bipolar disorder are characterized by deleterious mutations in synaptic and ion channel-related genes, (iii) there are two cases of deleterious mutations in XKR6 in bipolar disorder, whose deleterious mutations are significantly enriched in a broad spectrum of psychiatric/neurodevelopmental disorders, (iv) deleterious somatic mutations, including an LoF mutation in KMT2C, are enriched in the genes causative for developmental disorders if it existed as germline mutations, and two independent mutations in SRCAP were found from two unrelated patients with bipolar disorder. Genomic studies to date have clarified the characteristics of variants associated with bipolar disorder. At the same time, the heterogeneity of bipolar disorder and commonalities among bipolar disorder, schizophrenia, and neurodevelopmental disorders have also been clarified. We need to understand the relationship between genomic features and phenotypes with more detailed phenotypic information beyond the conventional diagnostic classifications, possibly with trans-diagnostic analysis. In this article, I describe our analysis of de novo mutation research in bipolar disorder with our study's background and prospects.
 Authors' abstract