Genetic contributions to bipolar disorder: current status and future directions

Abstract

Bipolar disorder (BD) is a highly heritable mental disorder and is estimated to affect about 50 million people worldwide. Our understanding of the genetic etiology of BD has greatly increased in recent years with advances in technology and methodology as well as the adoption of international consortiums and large population-based biobanks. It is clear that BD is also highly heterogeneous and polygenic and shows substantial genetic overlap with other psychiatric disorders. Genetic studies of BD suggest that the number of associated loci is expected to substantially increase in larger future studies and with it, improved genetic prediction of the disorder. Still, a number of challenges remain to fully characterize the genetic architecture of BD. First among these is the need to incorporate ancestrally-diverse samples to move research away from a Eurocentric bias that has the potential to exacerbate health disparities already seen in BD. Furthermore, incorporation of population biobanks, registry data, and electronic health records will be required to increase the sample size necessary for continued genetic discovery, while increased deep phenotyping is necessary to elucidate subtypes within BD. Lastly, the role of rare variation in BD remains to be determined. Meeting these challenges will enable improved identification of causal variants for the disorder and also allow for equitable future clinical applications of both genetic risk prediction and therapeutic interventions.

Keywords: Mood disorder; affective disorder; psychiatric genetics.

Fig. 1.

Polarity of symptoms for bipolar disorder subtypes. Bipolar disorder type I is characterized by at least one manic episode. Bipolar disorder type II is characterized by at least one depressive and one hypomanic episode during the lifetime, but no manic episodes. Major depressive disorder does not include episodes of hypomania or mania.

Fig. 2.

Estimated heritability of psychiatric, behavioral, and neurological disorders. Mean twin-based and SNP-based (on liability scale) heritability for different psychiatric (BIP, bipolar disorder; SCZ, schizophrenia; ADHD, attention-deficit/hyperactivity disorder; MD, major depression; ANX, generalized anxiety disorder), behavioral (AN, anorexia nervosa; AUD, alcohol use disorder; CUD, cannabis use disorder), or neurological (ASD, autism spectrum disorder; AD, Alzheimer's disorder; OCD, obsessive-compulsive disorder; TS, Tourette's syndrome) disorders. Error bars are shown for SNP-based estimates from LDSC.

Fig. 3.

Statistical power calculations for current and future GWAS. The variance explained by genome-wide significant variants (y-axis) is calculated for increasing GWAS sample sizes (x-axis) using the univariate causal mixture model (Holland et al., 2020). The legend describes the estimated GWAS sample sizes (SE) needed to capture 50% of the genetic variance (horizontal dashed line) associated with each trait. Stars indicate the sample sizes of currently available GWAS, and circles indicate the estimated sample sizes needed to capture 50% of the genetic variance for each trait. Traits include attention-deficit/hyperactivity disorder (ADHD) (Demontis et al., 2019b), autism spectrum disorder (ASD) (Grove et al., 2019), bipolar disorder (BD) (Mullins et al., 2020), depression (MDD) (Howard et al., 2019), and schizophrenia (SCZ) (Pardiñas et al., 2018). Height is included as a somatic control (no genetic correlation exists between height and bipolar disorder) (Yengo et al., 2018). s.e., standard error.