The CNV Module of AutDB: A New Resource for the Clinical Genetics of Autism Spectrum Disorders

Background: The advent of increasingly sensitive genome-wide scanning arrays has led to the discovery of numerous submicroscopic chromosomal deletions and duplications, more commonly known as copy number variants (CNVs), throughout the human genome.  CNVs are considered to be one of the leading genetic causes of neuropsychiatric disorders, with an estimated 10-20% of autism spectrum disorder (ASD) cases resulting from the presence of one or more CNVs in an affected individual.  A number of recurring CNVs with some degree in prevalence in autistic populations have been identified; however, the pathogenic relevance of the vast majority of CNVs identified in autistic cases remains unclear.  The necessity of determining and prioritizing potentially pathogenic CNVs in individuals with ASD is of great importance, as chromosomal microarray (CMA) screening of autistic individuals is increasingly being used to identify CNVs that confer at least some degree of risk of disease.  This information will be essential in the interpretation of genetic screening results in newborns and the subsequent identification of potentially at-risk individuals before the onset of disease.   

Objectives: The CNV module of the autism genetic database AutDB (http://mindspec.org/autdb.html) has been designed to function as an online resource for the ASD research community and consists of detailed annotations of published scientific reports in which one or more CNVs have been identified in an ASD population cohort.  In order to determine the global risk conferred by a given CNV in individuals with ASD, we performed a critical assessment of CNVs that were observed in ASD cohorts across multiple published reports in the CNV module database using a range of mathematical, statistical, and bioinformatics analyses. 

Methods: We prioritized CNV loci from these published reports based upon a number of criteria, including the number of studies in which a CNV was observed at that locus, the frequency of a CNV at that locus within ASD and control populations, the gene content of the CNV, and the mechanism of CNV inheritance. 

Results: While a number of previously characterized recurring CNV loci were among the highest ranking loci, a number of less well-characterized loci were also identified that warrant further investigation. 

Conclusions: We anticipate that the CNV module of AutDB will be a valuable resource for the clinical genetics of autism spectrum disorders.