Back

Disentangling the shared genetics of ADHD, cannabis use disorder and cannabis use and prediction of cannabis use disorder in ADHD

Nielsen, T. T.; Duan, J.; Levey, D. F.; Walters, G. B.; Johnson, E. C.; Thorgeirsson, T.; VA Million Veteran Program, ; Werge, T.; Mortensen, P. B.; Stefansson, H.; Stefansson, K.; Hougaard, D. M.; Agrawal, A.; Gelernter, J.; Grove, J.; Borglum, A. D.; Demontis, D.

2024-02-24 genetic and genomic medicine
10.1101/2024.02.22.24303124 medRxiv
Show abstract

Cannabis use disorder (CUD) and cannabis use (CU) are prevalent conditions cooccurring with ADHD, but not much is known about the underlying shared genetics. Here we perform cross-disorder GWAS meta-analyses of ADHD and CUD or CU to identify pleiotropic risk loci and evaluate differences in the genetics of ADHD-CUD and ADHD-CU, and subsequently we dissect the polygenic architecture of CUD comorbidity in ADHD in the iPSYCH cohort. There was a higher genetic overlap of ADHD and CUD than observed for ADHD and CU and we found a significant direct effect of ADHD genetic risk on CUD with only a minor part (12%) mediated by the genetics of CU. We identified 36 genome-wide significant loci for ADHD-CUD and 10 loci for ADHD-CU, with concordant direction of effect on the phenotypes. Three different approaches identified DRD2, which encodes the dopamine 2 receptor, as a risk gene for ADHD-CUD and, overall, ADHD-CUD risk genes were associated with high expression across several brain tissues and brain developmental stages, which was not observed for ADHD-CU genes. ADHD-CUD and ADHD-CU demonstrated similar genetic correlations with substance use phenotypes, while they differed significantly with respect to substance use disorder (SUD) phenotypes. ADHD-CUD individuals had significantly increased polygenic score (PGS) for psychiatric disorders compared to ADHD without CUD and increased burden of rare deleterious variants. Stratifying individuals with ADHD by their CUD-PGS revealed an absolute risk of 22% for comorbid CUD among the 20% of cases with the highest CUD-PGS, which was strikingly higher than the absolute risk of 1.6% observed among the 20% of controls with the highest CUD-PGS. Sex-specific analyses identified substantial differences in the absolute risk of comorbid CUD between males and females with ADHD, with a [~]10% higher CUD risk among males than females in the high-risk CUD-PGS group (24% risk for males and 14% risk for females).

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.