Investigating Direct and Indirect Genetic Effects in Attention-Deficit/Hyperactivity Disorder Using Parent-Offspring Trios

Abstract

Background: Attention-deficit/hyperactivity disorder (ADHD) is highly heritable, but little is known about the relative effects of transmitted (i.e., direct) and nontransmitted (i.e., indirect) common variant risks. Using parent-offspring trios, we tested whether polygenic liability for neurodevelopmental and psychiatric disorders and lower cognitive ability is overtransmitted to ADHD probands. We also tested for indirect or genetic nurture effects by examining whether nontransmitted ADHD polygenic liability is elevated. Finally, we examined whether complete trios are representative of the clinical ADHD population.

Methods: Polygenic risk scores (PRSs) for ADHD, anxiety, autism, bipolar disorder, depression, obsessive-compulsive disorder, schizophrenia, Tourette syndrome, and cognitive ability were calculated in UK control subjects (n = 5081), UK probands with ADHD (n = 857), their biological parents (n = 328 trios), and also a replication sample of 844 ADHD trios.

Results: ADHD PRSs were overtransmitted and cognitive ability and obsessive-compulsive disorder PRSs were undertransmitted. These results were independently replicated. Overtransmission of polygenic liability was not observed for other disorders. Nontransmitted alleles were not enriched for ADHD liability compared with control subjects. Probands from incomplete trios had more hyperactive-impulsive and conduct disorder symptoms, lower IQ, and lower socioeconomic status than complete trios. PRS did not vary by trio status.

Conclusions: The results support direct transmission of polygenic liability for ADHD and cognitive ability from parents to offspring, but not for other neurodevelopmental/psychiatric disorders. They also suggest that nontransmitted neurodevelopmental/psychiatric parental alleles do not contribute indirectly to ADHD via genetic nurture. Furthermore, ascertainment of complete ADHD trios may be nonrandom, in terms of demographic and clinical factors.

Keywords: ADHD; Case-control design; Genetic nurture; Missing data; Polygenic risk scores; Trio design.

Figure 1:

Mean deviation of proband polygenic risk scores from the mid-parent distribution (i.e. standard deviations away from the mid-parent distribution) in ADHD trios, using the a) primary sample (N=328 trios) and the b) replication sample (N=844 trios). ADHD: attention deficit hyperactivity disorder; ANX: anxiety disorders; ASD: autism spectrum disorder; BD: bipolar disorder; COG: cognitive ability; MDD: major depressive disorder; OCD: obsessive-compulsive disorder; pTDT: polygenic transmission disequilibrium test; SCZ: schizophrenia; TS: Tourette’s syndrome. P-values indicate the probability that the mean of the pTDT deviation distribution is 0 (two-sided, one-sample t test). Error bars indicate standard errors. * p<0.05; ** p<0.01; *** p<0.001. P-values shown are corrected for multiple tests for primary analyses and raw p-values are shown for the replication analyses. See Table S1 for detailed results.

Figure 2:

Mean polygenic risk scores in ADHD complete trios: a) primary sample (N=328) and b) replication sample (N=616), in proband (P), fathers (F), mothers (M), and non-transmitted parental alleles (NT), relative to the control population sample (bold horizontal line at y=0). ADHD: attention deficit hyperactivity disorder; ANX: anxiety disorders; ASD: autism spectrum disorder; COG: cognitive ability; SCZ: schizophrenia; OCD: obsessive-compulsive disorder; TS: Tourette’s syndrome. Note: Bipolar disorder and major depressive disorder could not be examined due to inclusion of the control population sample in the discovery genetic studies for those disorders. Error bars indicate standard errors. * p<0.05; ** p<0.01; *** p<0.001. P-values shown are corrected for multiple tests for primary analyses and raw p-values are shown for the replication analyses. See Table S2 for detailed results.