An analytical framework for whole-genome sequence association studies and its implications for autism spectrum disorder

Abstract

Genomic association studies of common or rare protein-coding variation have established robust statistical approaches to account for multiple testing. Here, we present a comparable framework to evaluate rare and de novo noncoding single nucleotide variants, insertion/deletions, and all classes of structural variation from whole-genome sequencing (WGS). Integrating genomic annotations at the level of nucleotides, genes, and regulatory regions, we define 51,801 annotation categories. Analyses of 519 autism spectrum disorder families did not identify association with any categories after correction for 4,123 effective tests. Without appropriate correction, biologically plausible associations are observed in both cases and controls. Despite excluding previously identified gene-disrupting mutations, coding regions still exhibited the strongest associations. Thus, in autism the contribution of de novo noncoding variation is probably modest compared to de novo coding variants. Robust results from future WGS studies will require large cohorts and comprehensive analytical strategies that consider the substantial multiple testing burden.

De novo variants in population constrained fetal brain enhancers and intellectual disability

“…Aggregation of a minority of disease-causing variants with the majority of benign regulatory variants nullifies any signal from disease-causing variants in non proteincoding genomic regions in disease cohorts. It is noteworthy that in protein coding regions of the genome only protein-truncating variants, but not other protein-coding variants, show significant enrichment in neurodevelopmental disorders 11,67 . The analysis of DNVs in selected monogenic phenotypes provides a powerful analysis instrument, because it can focus on a relatively small number of variants that have increased likelihood of being disease-causing.…”

Unified inference of missense variant effects and gene constraints in the human genome

“…Comparison with alternative methods in predicting de novo mutations associated with developmental disorders We downloaded de novo mutations identified in 4,293 individuals affected by developmental disorders [31] and 2,278 healthy individuals from denovo-db (version v1.6.1). The healthy controls included individuals enrolled in previous studies of autism spectrum disorder [80,81,82,83,84], severe non-syndromic sporadic intellectual disability [85], schizophrenia [86], and healthy populations [87, 88, 89, 90]. We removed redundant variants and any variants presented in the gnomAD exome sequencing data.…”

Estimating contribution of rare non‐coding variants to neuropsychiatric disorders

“…However, nominal significance was indeed observed in the control group, further suggesting need of a comprehensive, unbiased, and appropriately stringent approach, such as CWAS. Key findings and the points of difference in these two studies are summarized in Table …”

“…Considering possible sources of many differences between the two studies, first, each study used a different statistical procedure to calculate P ‐values: one‐tailed Fisher's exact tests without covariation for possible confounding factors, such as paternal age, were used in Turner et al significance was calculated by case/control label‐swapping permutations (at least 10 000 times) that took paternal age and sequencing quality metrics into account. Given that significant overall enrichment of DNM in ASD disappeared when adjusting for paternal age and sequencing quality metrics by linear regression, the analysis by Werling et al .…”

“…; while in Werling et al ., significance was calculated by case/control label‐swapping permutations (at least 10 000 times) that took paternal age and sequencing quality metrics into account. Given that significant overall enrichment of DNM in ASD disappeared when adjusting for paternal age and sequencing quality metrics by linear regression, the analysis by Werling et al . seems to be more conservative and would be more accurate.…”