Genomewide Association and Meta-Analysis of Autism Spectrum Disorder in the Multi-Ethnic Charge Cohort

Thursday, May 11, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
C. L. Simpson1, R. J. Schmidt2, K. Kim3, R. Hansen4 and I. Hertz-Picciotto5, (1)University of Tennessee Health Science Center, Memphis, TN, (2)Public Health Sciences, University of California Davis, Davis, CA, (3)Department of Public Health Sciences, University of California, Davis, Davis, CA, (4)UCD MIND Institute, Sacramento, CA, (5)University of California at Davis, Davis, CA

The Childhood Autism Risks from Genetics and the Environment (CHARGE) study is population-based cohort of children with autism or developmental delay and typically developing children, recruited from a statewide database of persons receiving services from regional centers in northern California, from clinical and self-referrals and referrals from other studies at the MIND Institute. Genomewide association studies (GWAS) are a standard genetic epidemiological tool for the assessment of genetic contributions to risk of disease and have produced evidence for genetic variants in a range of psychiatric and neurodevelopmental disorders. A number of GWAS have been performed in ASD and identified many associations, however there have been few successful replications, perhaps in part because of high polygenicity and variable effect sizes.


We performed GWAS in the CHARGE cohort using the Affymetrix European-specific array, which was developed in conjunction with UCSF and Kaiser Permanente.


Genotypes were called in Genotyping Console and standard quality control measures were applied. Principal components analysis (PCA) was used to compare self-reported ancestry with HapMap anchors and extreme outliers removed. Each ethnic ancestry population was then subject to separate PCA to generate eigenvalues and used to control population stratification in the association analysis. Data were imputed to the Haplotype Reference Consortium reference panel. All quality control and analyses for each population were performed in R and PLINK, and meta-analysis across populations performed using METAL.


Subjects were separated into six ethnically distinct; non-Hispanic whites, Hispanic whites, Hispanic other, African American, Asian and Multi-ethnic. The African American and Asian groups were not analyzed due to very low subject numbers.

Five genomewide significant signals were detected on chromosome 2, with a minimum p value of 1.6x10-8. All signals were located in introns of the ceramide kinase-like gene CERKL. This gene contains multiple transcripts and non-coding RNA’s and encodes a protein with ceramide kinase-like domains but does not phosphorylate ceramide and is currently of unknown function. It is widely expressed, but different transcripts are expressed in different tissues and at different time points of development.


Meta-analysis of multi-ethnic cohorts is a useful tool for dissection of complex traits such as autism and here identifies genome-wide significant signals on chromosome 2 in the CERKL gene. This gene is of unknown function and so its relevance to ASD cannot be assessed. Other genes in the region include integrin alpha 4 (ITGA4), a cell surface adhesion and signaling protein associated with the autoimmune disorders inflammatory bowel disease and multiple sclerosis and the neuronal differentiation gene NEUROD1, known to be associated with Type 1 diabetes. Additional analyses and deeper investigation into these results will be presented.

See more of: Genetics
See more of: Genetics