Genome-wide characteristics of de novo mutations in autism

Abstract

De novo mutations (DNMs) are important in Autism Spectrum Disorder (ASD), but so far analyses have mainly been on the ~1.5% of the genome encoding genes. Here, we performed whole genome sequencing (WGS) of 200 ASD parent-child trios and characterized germline and somatic DNMs. We confirmed that the majority of germline DNMs (75.6%) originated from the father, and these increased significantly with paternal age only (p=4.2×10^-10). However, when clustered DNMs (those within 20kb) were found in ASD, not only did they mostly originate from the mother (p=7.7×10^-13), but they could also be found adjacent to de novo copy number variations (CNVs) where the mutation rate was significantly elevated (p=2.4×10^-24). By comparing DNMs detected in controls, we found a significant enrichment of predicted damaging DNMs in ASD cases (p=8.0×10^-9; OR=1.84), of which 15.6% (p=4.3×10^-3) and 22.5% (p=7.0×10^-5) were in the non-coding or genic non-coding, respectively. The non-coding elements most enriched for DNM were untranslated regions of genes, boundaries involved in exon-skipping and DNase I hypersensitive regions. Using microarrays and a novel outlier detection test, we also found aberrant methylation profiles in 2/185 (1.1%) of ASD cases. These same individuals carried independently identified DNMs in the ASD risk- and epigenetic- genes DNMT3A and ADNP. Our data begins to characterize different genome-wide DNMs, and highlight the contribution of non-coding variants, to the etiology of ASD.

Figure 1

Sequence context of regions with de novo mutations. (a) Transition (Ti) to transversion (Tv) ratio of different kinds of de novo mutations found in: germline of ASD, germline of Dutch population controls, somatic events of ASD and lymphocyte-derived cell line (LCL) of ASD. (b) Sequence context of the base substitution mutation spectra for different de novo mutations. Each of the 96 mutated trinucleotides (mutated position at centre) from each cohort is represented in a heatmap (intensity of colour correspond to frequency of each mutation). The 5′ base to the mutated site is shown on the vertical axis, while the 3′ base is shown on the horizontal axis.

Figure 2

Origins of de novo mutations in ASD. (a) Parent-of-origin of germline and somatic variants. Number of germline de novo SNVs and de novo indels derived from the father was significantly higher than that from the mother. On the other hand, there are significantly more clustered (within 20 kb) germline de novo mutations originating from the mother than from the father, while somatic mutations can be found in similar proportion from both parents. (b) Number of de novo SNVs found on the paternal allele increases with the age of father, but there is no correlation between the number of de novo SNVs found on the maternal allele and the age of the mother. (c) Distance between de novo mutations is shorter than expected for a subset of de novo mutations both between and within individuals. (d) Mutation rate is significantly higher than the background within 100 kb flanking the de novo CNVs.

Figure 3

Functional impact of genome-wide damaging de novo mutations. (a) Damaging de novo mutations are significantly enriched in both coding (missense) and non-coding regions (splicingNeg, UTR3 and UTR5) in ASD compared with population controls. Definition of damaging tiers can be found in the Materials and methods. LOF, loss-of-function mutations; missense, missense mutations; splicingNeg, exon-skipping mutations predicted by SPANR; UTR, untranslated regions. Number of variants is indicated above each bar. Solid horizontal line indicates OR=1, and dash horizontal line represents P=0.05. (b) Non-coding de novo mutations in non-genic regions are significantly enriched in DNase I hypersensitive regions (DHS). Damaging de novo mutations predicted by ‘Deepbind loss; PCons>0’ are significantly enriched in ASD in general (All regions), but further enriched in DNase I hypersensitive regions and regions proximal to genes. DHS, DNase I hypersensitive sites; PCons, PhastCons; Tx, transcript. Number of variants is indicated above each bar. Solid horizontal line indicates OR=1, and dash horizontal line represents P=0.05. (c) Damaging de novo mutations are significantly enriched (false discovery rate ⩽0.25) in Gene Ontology defined pathways that are related to chromatin organisation, RNA processing and translation, synaptic transmission and others. Genes involved in the pathways are listed. Genes with DNMs in coding region are bolded. Asterisk represents gene that is found in more than one gene pathway.

Figure 4

Sample outliers identified by the Methylation Outlier Sample Test (MOST). (a) Sample 2-1276-003, carrying a de novo damaging heterozygous mutation at DNMT3A (p.R635W), was identified as an outlier sample in principle component (PC) 9 and 13. (b) Sample 2-0028-003, which carries a de novo frameshift mutation at ADNP (p.Q345fs), was identified as an outlier sample in PC9 and PC11 (and PC14 not shown). (c) Sample 2-1280-003 was identified as an outlier sample in PC 12. No de novo mutation in known epigene was found, but there is a maternal inherited rare damaging missense mutation at KMT5C (p.R205Q). (d) Functional enrichment of genes involved in the PC9 responsible for the sample outliers. Functions from negative loadings are in blue and that from positive loadings are in red.