Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Oct 3;93(4):607-19.
doi: 10.1016/j.ajhg.2013.09.001.

Identification of small exonic CNV from whole-exome sequence data and application to autism spectrum disorder

Christopher S Poultney  1 , Arthur P Goldberg, Elodie Drapeau, Yan Kou, Hala Harony-Nicolas, Yuji Kajiwara, Silvia De Rubeis, Simon Durand, Christine Stevens, Karola Rehnström, Aarno Palotie, Mark J Daly, Avi Ma'ayan, Menachem Fromer, Joseph D Buxbaum
Affiliations

Identification of small exonic CNV from whole-exome sequence data and application to autism spectrum disorder

Christopher S Poultney et al. Am J Hum Genet. .

Abstract

Copy number variation (CNV) is an important determinant of human diversity and plays important roles in susceptibility to disease. Most studies of CNV carried out to date have made use of chromosome microarray and have had a lower size limit for detection of about 30 kilobases (kb). With the emergence of whole-exome sequencing studies, we asked whether such data could be used to reliably call rare exonic CNV in the size range of 1-30 kilobases (kb), making use of the eXome Hidden Markov Model (XHMM) program. By using both transmission information and validation by molecular methods, we confirmed that small CNV encompassing as few as three exons can be reliably called from whole-exome data. We applied this approach to an autism case-control sample (n = 811, mean per-target read depth = 161) and observed a significant increase in the burden of rare (MAF ≤1%) 1-30 kb CNV, 1-30 kb deletions, and 1-10 kb deletions in ASD. CNV in the 1-30 kb range frequently hit just a single gene, and we were therefore able to carry out enrichment and pathway analyses, where we observed enrichment for disruption of genes in cytoskeletal and autophagy pathways in ASD. In summary, our results showed that XHMM provided an effective means to assess small exonic CNV from whole-exome data, indicated that rare 1-30 kb exonic deletions could contribute to risk in up to 7% of individuals with ASD, and implicated a candidate pathway in developmental delay syndromes.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Transmission Analysis Transmission rate of CNV from parents to child is shown as a function of XHMM quality score threshold (SQ) for CNV of size 1–30 kb (left), 1–10 kb (center), and 10–30 kb (right). At each SQ threshold, transmission was calculated and aggregated per family. The red line indicates the median per-family transmission rate at each threshold, with gray bars indicating the interquartile range at each threshold. A minor allele frequency (MAF) filter of 5% was applied to the entire set of CNV before size stratification and transmission analysis.
Figure 2
Figure 2
Enrichment of Small Deletions in Autism (A) Case/control comparisons for 1–30 kb CNV (deletions and duplications) in the top row, and 1–30 kb deletions in the bottom row. (B) Case/control comparisons for 1–10 kb CNV (deletions and duplications) in the top row, and 1–10 kb deletions in the bottom row. Error bars represent SEM.
Figure 3
Figure 3
Molecular Validation of XHMM Calls Each panel shows ten examples of qPCR-validated deletions in differing size ranges, as indicated in the panel. Each deletion is represented by a pair of bars representing the dosage of DNA relative to control as assayed by using qPCR for a probe placed within the called deletion. The left bar shows the result, normalized to 1.0, for the control probe; the right bar shows the normalized dosage for the probe in the sample with the called deletion. Error bars represent SEM. Note that the rightmost sample in the upper panel (04C38268A, gene ATP8B3) is likely a homozygous deletion.
Figure 4
Figure 4
Breakpoint Determination via Sanger Sequencing Each panel shows the results from Sanger sequencing of one of five validated deletions. In each panel, exons and transcripts are shown on the top, followed by the extent of the called and validated deletions, and finally the sequence surrounding deletion start and end points. Regions of surrounding sequence that differ from control are shown in red.
Figure 5
Figure 5
DAPPLE Network Derived from Genes Hit by 1–30 kb Deletions The subnetwork of direct connections is shown for genes disrupted in ASD cases (left) or controls (right). For case genes, there were 21 connections between 17 genes and the network was significantly enriched for direct connections between input genes (p = 0.006) and mean number of direct connections per input gene (p = 0.002). For control genes, there were four connections between eight genes and the network showed no enrichment for direct connections between input genes (p = 0.106) or mean number of direct connections per input gene (p = 0.855).
Figure 6
Figure 6
Interaction of Genes Hit by 1–30 kb Deletions in Cases with Autophagy Genes We made use of a carefully curated PPI network to explore the autophagy pathway and its relation with genes hit by small deletions in ASD subjects. Circular nodes represent genes hit by CNV in ASD subjects. Green nodes participate in the autophagy pathway according to the NCBI Gene Database and green rectangular nodes are categorized as hub proteins in the pathway.
Figure 7
Figure 7
Networks of ASD Genes Disrupted by Small CNV or De Novo Loss-of-Function Mutations We made use of a carefully curated PPI network to explore the relationship of genes disrupted by small CNV (this study) or de novo loss-of-function mutations. All nodes were sized based on connectivity degree. Green nodes denote small CNV case genes, blue nodes represent LoF genes, and the orange- to brown-color nodes are intermediate proteins where the shade is based on p value computed by using a proportion test, with darker color indicating smaller p value.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Iafrate A.J., Feuk L., Rivera M.N., Listewnik M.L., Donahoe P.K., Qi Y., Scherer S.W., Lee C. Detection of large-scale variation in the human genome. Nat. Genet. 2004;36:949–951. - PubMed
    1. Sanders S.J., Ercan-Sencicek A.G., Hus V., Luo R., Murtha M.T., Moreno-De-Luca D., Chu S.H., Moreau M.P., Gupta A.R., Thomson S.A. Multiple recurrent de novo CNVs, including duplications of the 7q11.23 Williams syndrome region, are strongly associated with autism. Neuron. 2011;70:863–885. - PMC - PubMed
    1. Sebat J., Lakshmi B., Troge J., Alexander J., Young J., Lundin P., Månér S., Massa H., Walker M., Chi M. Large-scale copy number polymorphism in the human genome. Science. 2004;305:525–528. - PubMed
    1. Cooper G.M., Coe B.P., Girirajan S., Rosenfeld J.A., Vu T.H., Baker C., Williams C., Stalker H., Hamid R., Hannig V. A copy number variation morbidity map of developmental delay. Nat. Genet. 2011;43:838–846. - PMC - PubMed
    1. Jakobsson M., Scholz S.W., Scheet P., Gibbs J.R., VanLiere J.M., Fung H.-C., Szpiech Z.A., Degnan J.H., Wang K., Guerreiro R. Genotype, haplotype and copy-number variation in worldwide human populations. Nature. 2008;451:998–1003. - PubMed

Publication types