26097
Whole Genome Sequencing and Rare Variant Discovery in the Aspire Autism Spectrum Disorder Cohort
ASPIRE (Autism SPectrum Interdisciplinary REsearch) cohort is comprised of Individuals that have been diagnosed with ASD (DSM-IV; ADOS-G/ADI-R) and underwent a detailed standardized phenotyping protocol, including morphometrics, through Provincial Medical Genetics Program in BC.
Objectives:
Our goal was to characterize rare genetic variation using whole genome sequencing in a subset of the ASPIRE cohort and to examine genetic findings in the context of patients’ deep phenotype data. To achieve this, we assembled a robust bioinformatics pipeline for identification and prioritization of potentially ASD associated variants that incorporates publically available resources and tools as well as in-house developed ones.
Methods:
We obtained whole genome sequences (Illumina paired-end 100 base pair reads, average depth 30x) for each subject. Variants were called using the Genome Analysis Toolkit (GATK) against the human reference genome, and filtered for quality and rarity in population data. To prioritize variants further, we relied on a combination of existing and in-house bioinformatics tools incorporating both gene-level metrics (genic intolerance to mutation, functional effect prediction) and variant level metrics (population frequency, predicted damage). Our efforts included recurating and harmonizing variants reported in the ASD literature, totalling close to 5000 variants from over 2500 ASD individuals. We have made the resulting database available as MARVdb (www.chibi.ubc.ca/marvdb). To collaboratively analyze variants in the context of phenotypes, we used ASPIREdb (aspiredb.chibi.ubc.ca), an interactive web application developed by our lab, which allows researchers to search, organize, analyze and visualize variants and phenotypes associated with a set of human subjects.
Results:
In total 97 high-priority candidate variants were identified, affecting 66 subjects. All of these variants were heterozygous and all but one were autosomal. Of these, 31 were predicted loss-of-function mutations, 3 affecting genes previously associated with ASD and 3 affecting genes associated with other neurodevelopmental disorders. A total of 66 candidate MS mutations were prioritized as potentially pathogenic, including 12 affecting literature-associated ASD genes, and 8 affecting genes associated with other neurodevelopmental disorders. High priority variants were subjected to trio resequencing to assess inheritance.
Conclusions:
Using this approach we have identified, among others, a novel de novo splice site variant in SCN2A, predicted to result in a loss of function, in an individual with severe autism and intellectual disability. This finding adds to the evidence that mutations in this gene can be associated with autism without comorbid seizure disorders.