Autism Genetic Database (AGD): Using Bioinformatics to Study the Genetics of Autism

Background: Autism is a neurodevelopmental disorder with onset in early childhood. It belongs to a group of conditions known as autism spectrum disorders (ASD). The complex and heterogeneous clinical manifestation of this genetic disorder requires a multidisciplinary approach to simultaneously investigate the impact of multiple genetic factors. Recently, we have developed the first comprehensive autism repository, Autism Genetic Database (AGD), which comprises a list of reported autism susceptibility genes and copy number variations (CNVs), as well as all known human noncoding RNAs and fragile sites. The ultimate goal of this bioinformatics project is to provide a public repository of autism genetic information for the research community. In the past few years, several studies have generated CNV data on healthy individuals to be used as a baseline for evaluating pathogenic relevance of disease related CNVs. Furthermore, CNVs have been linked to various types of developmental and psychiatric disorders including schizophrenia, bipolar disorder, ADHD, and mental retardation. Clinical and genetic overlap among different neurodevelopmental disorders has indicated the importance of evaluating their common underlying mechanisms. Objectives: The objectives of this study were (1) to update our newly developed AGD database and to expand its scope by adding new features; and (2) to identify testable hypotheses by applying statistical analyses using AGD data. Methods: Susceptibility genes with at least one suggestive autism association were included. CNVs with at least one reported disease association were included and classified based on the association with ASD or other neurodevelopmental disorders. Statistical data analyses were performed using updated information stored in AGD. Results: Since the initial development of AGD, there have been new reports of genes and CNVs associated with autism. The list of autism susceptibility genes and CNVs were updated in our database according to literature searches. CNVs in healthy individuals as well as other neurodevelopmental disorders (i.e., ADHD, schizophrenia, and bipolar disorder) were added and labeled in the AGD database. Statistical data analyses were conducted using information stored in AGD to evaluate the distribution and potential correlation of different data-points in autism. To further elucidate autism specific profiles, applied analyses were also performed based on the distribution of reported CNVs in other neurodevelopmental disorders and healthy subjects. Conclusions: Our study demonstrates that bioinformatic tools will enable rigorous evaluation of the growing list of autism susceptibility factors. Our comprehensive autism database provides an opportunity to perform initial in silico analyses identifying previously undetected potential association/interaction profiles of these multiple genetic factors. The AGD database not only introduces a new and useful bioinformatics tool to the research community, but also enhances basic molecular genetic research projects by providing models and analytically driven hypotheses to be evaluated in the laboratory.