Genomic dissection of sleep archetypes in a large autism cohort

Abstract

Poor sleep is a major concern among individuals with autism and their caregivers. To better characterize the genetic and phenotypic heterogeneity of poor sleep in autism, we recruited 5,686 families from SPARK, a nationwide genetic study of autism, who described their sleep experiences using the Children's Sleep Health Questionnaire (CSHQ) and other self-report items. The collective experiences from this large sample allowed us to discover eight distinct archetypes of sleep in autism. Membership in some of these archetypes showed significant SNP-heritability (0.50 - 0.65, 95% confidence interval = 0.08 - 1), and polygenic estimates of educational attainment, BMI, and ADHD risk contributed extensively to the genetic signatures of these sleep archetypes. Surprisingly, polygenic estimates of general population sleep phenotypes showed sparser and more modest associations, perhaps suggesting that the genetic drivers of disordered sleep in autism may be distinct from those encountered in the general population. GWAS on archetype membership yielded no genome-wide significant loci, however, the most significant gene for the most severe archetype was the nitric oxide (NO) signaling gene NOS1AP, which was previously linked to sleep disruption in schizophrenia. Finally, the eight sleep archetypes showed specific signatures of treatment response across five major categories of sleep aid, pointing to the potential of treatment plans that are tailored to the nature of the sleep problem. These findings provide critical new insight into the comorbidities, subtypes, and genetic risk factors associated with disordered sleep in autism.

Figure 1:

SPARK cohort CSHQ associations. Correlation values (R squared) of comorbidity, diagnostic, and parental sleep questionnaire data with proband total CSHQ score are shown for SPARK participants (n=5686) split by age group (A). The association between total CSHQ score and major diagnostic and demographic factors are shown (B-I).

Figure 2:

Associations between total CSHQ score and polygenic risk scores (PRS) in ASD. A total of 50 z-scaled PRS scores (SNP p-value threshold = 0.01) were tested for association with total CSHQ score using a linear model. The x-axis represents the coefficients from this model fit. Bolded, dark orange PRS are associated at FDR < 0.05, while lighter orange PRS are associated at FDR < 0.10.

Figure 3:

ASD sleep archetypes. Archetypal analysis was performed using CSHQ questionnaire items (A). Characteristic patterns across items of the CSHQ distinguished the sleep archetypes (B). Individuals have a coefficient for each archetype, with some fitting unambiguously (see methods) into a single sleep archetype (C), while most individuals appeared as a mixture of archetypes. Archetypes do not vary drastically in age, sex, or race (D). Sleep archetypes display specific patterns of PRS association (E), with normal sleep (A2) showing depleted risk for high BMI and ADHD coupled with high propensity for educational attainment, and broadly affected (A1) showing the inverse pattern. Sleep archetypes showed specific patterns of comorbidities, most not explicitly related to sleep (F).

Figure 4:

Heritability, GWAS, and MAGMA analysis of sleep archetypes. GCTA-GREML heritability estimates of the sleep archetypes (A). GWAS on sleep archetype coefficients with SNP color representing the archetype with the most significant p-value (B). MAGMA gene-level analysis of archetype GWAS results, with the same coloring scheme as the SNP plot (C). The top gene for each archetype is shown. Points below significance level of p=0.001 are shaded light grey.

Figure 5:

Response to sleep aid varies by sleep archetype. Number of individuals who have taken each category of sleep aid (A). Some archetype coefficients are associated with decreasing sleep aid response, e.g., melatonin (B), while others are associated with increasing response (C). We performed this analysis over all combinations of archetype and sleep aid (D). Increased responsiveness is indicated with a blue-green shade, while decreased responsiveness is indicated in orange.