Genetics plays a major role in autism, but the relationships are complex and not well understood. Individuals with autism frequently also have other disorders such as fragile X syndrome, Rett syndrome and tuberous sclerosis. Mutations in specific genes — such as neuroligins, which help relay signals between neurons — have also been linked to autism.

Overall, more than 20 genes and chromosomal regions are known to cause autism when defective but together, these account for only around 10 percent of cases. Huda Zoghbi and her colleagues at Baylor College of Medicine have taken on the weighty task of developing a strategy to uncover the genes involved in the remaining majority of autism cases.

The researchers reason that autism-related molecular pathways, such as those involved in the functioning of neuronal networks, are a good starting point in their search for potential autism genes. They have assembled a collection of proteins known to be implicated in these pathways and then screened a library of all human brain proteins to find other proteins that interact with those in the collection. After ruling out a number of genes involved in disorders whose symptoms are distinct from those of autism, Zoghbi and colleagues are left with about 500 genes for further study.

Zoghbi and collaborator Richard Gibbs propose to sequence these 500 genes in DNA samples from individuals with autism enrolled in the Simons Simplex Collection. They plan to screen the samples for chromosomal deletions or duplications, and to analyze the DNA from the individuals’ parents.

Using these genetic techniques, Zoghbi and colleagues aim to pinpoint genes in which sequence variations or defects are linked to autism and determine whether these events are heritable or occur spontaneously. The researchers hope that the identification of new autism genes will pave the way for studies of protein function that may help better understand the mechanics of autism at the cellular level.