Identification of novel genomic loci for anxiety symptoms and extensive genetic overlap with psychiatric disorders

Abstract

Aims: Anxiety disorders are prevalent and anxiety symptoms (ANX) co-occur with many psychiatric disorders. We aimed to identify genomic loci associated with ANX, characterize its genetic architecture, and genetic overlap with psychiatric disorders.

Methods: We included a genome-wide association study of ANX (meta-analysis of UK Biobank and Million Veterans Program, n = 301,732), schizophrenia (SCZ), bipolar disorder (BIP), major depression (MD), attention-deficit/hyperactivity disorder (ADHD), and autism spectrum disorder (ASD), and validated the findings in the Norwegian Mother, Father, and Child Cohort (n = 95,841). We employed the bivariate causal mixture model and local analysis of covariant association to characterize the genetic architecture including overlap between the phenotypes. Conditional and conjunctional false discovery rate analyses were performed to boost the identification of loci associated with anxiety and shared with psychiatric disorders.

Results: Anxiety was polygenic with 12.9k genetic variants and overlapped extensively with psychiatric disorders (4.1k-11.4k variants) with predominantly positive genetic correlations between anxiety and psychiatric disorders. We identified 119 novel loci for anxiety by conditioning on the psychiatric disorders, and loci shared between anxiety and MD $\left(n=47\right)$ , BIP $\left(n=33\right)$ , SCZ $\left(n=71\right)$ , ADHD $\left(n=20\right)$ , and ASD $\left(n=5\right)$ . Genes annotated to anxiety loci exhibit enrichment for a broader range of biological pathways including cell adhesion and neurofibrillary tangle compared with genes annotated to the shared loci.

Conclusions: Anxiety is highly polygenic phenotype with extensive genetic overlap with psychiatric disorders, and we identified novel loci for anxiety implicating new molecular pathways. The shared genetic architecture may underlie the extensive cross-disorder comorbidity of anxiety, and the identified molecular underpinnings may lead to potential drug targets.

Keywords: anxiety; genetic loci; genetic overlap; psychiatric disorder.

Fig. 1

(a–d) Bivariate causal mixture model (MiXeR): genome‐wide genetic overlap between anxiety symptoms (ANX) and major depression (MD), bipolar disorder (BIP), schizophrenia (SCZ), and attention‐deficit/hyperactivity disorder (ADHD). r _g, genetic correlation. The numbers indicate estimates of trait‐influencing variants in the thousands. Colors: gray, shared variants; orange, variants unique to ANX; and blue, variants unique to MD, BIP, SCZ, or ADHD.

Fig. 2

(a–d) Local analysis of covariant association (LAVA): volcano plots of local genetic correlation coefficients (rho) with −log₁₀ P values for each locus. Dark red dots represent significantly correlated loci after Bonferroni correction. ADHD, attention‐deficit/hyperactivity disorder; ANX, anxiety symptoms; BIP, bipolar disorder; MD, major depression; SCZ, schizophrenia.

Fig. 3

Manhattan plot. Genomic risk loci associated with anxiety. Circled dots indicate the lead single nucleotide polymorphisms with genome‐wide significance.

Fig. 4

(a,b) Logistic regression: the association between polygenic risk scores (PRS) of various psychiatric disorders and anxiety disorders in MoBa (Mother, Father, and Child Cohort Study) parents. (a) Models for the PRS of anxiety symptoms (ANX), attention‐deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), bipolar disorder (BIP), major depression (MD), schizophrenia (SCZ), and covariates. (b) Nagelkerke R ² showing the difference in the percentage of prediction of anxiety traits by each PRS over a base model that includes age, sex, and genotype principal components. CI, confidence interval.