While it is clear that autism has a strong inherited genetic component, many large scale genetic studies that have relied only on a general ASD diagnosis or information collected only on affected individuals have had limited success in identifying risk alleles. An alternative is to examine behavioral biomarkers, especially for language ability, to assist with gene mapping, and to extend the data collection beyond just the individual diagnosed with autism to include family members that also exhibit a language delay. This project will use this approach by utilizing an existing database that contains clinical and genetic information for people with autism and all their family members in order to identify biomarkers for risk of autism. Potential behavioral biomarkers will be validated and candidate genes will be examined for DNA variations in sequence or copy number. Overall, the combined analysis of behavioral and genetic biomarkers will enable development of more accurate models for the prediction of risk for autism spectrum disorders.