World's first genetic database for autism boosts research
Does autism have a genetic basis? This is one of the most pressing questions in autism research, but few genes have been implicated in the disorder, and those that have been linked typically play minor roles.
"There are so many views about the genetic etiology of complex disorders — and the jury is still out," said Sharmila Banerjee Basu, president and chief scientific officer of MindSpec, an informatics organization devoted to autism research. Banerjee Basu believes that scientists will only be able to answer this vexing question if they have the right tools at their fingertips. That's why MindSpec is developing SFARI Gene in partnership with the Simons Foundation.
Launched in 2008, SFARI Gene is the world's first genetic database for autism. Continuously updated by a team of curators, the database allows researchers to search for genes that have been linked to autism in the peer-reviewed scientific literature. Unlike similar projects such as AlzGene, a genetic database for Alzheimer's, SFARI Gene does not just house information about common genetic variants — it also includes information about genes identified from genetic association studies, rare single-gene mutations and genes implicated in genetic diseases.
SFARI Gene is a repository of "genes that have been studied in the context of autism in any way," said Alan Packer, associate director for research at the Simons Foundation. "We want all the information there."
Banerjee Basu, who was a staff scientist at the National Human Genome Research Institute until 2005, says it is crucial for the database to contain such diverse information because autism is so complex. "It is more likely that a common variant, in conjunction with some other single-gene or multi-gene effect like copy number variation, will participate in a complex way to underlie autism," she said. The hope is that such a repository will help researchers synthesize various aspects of the disease and understand how genes and networks work together.
With so much information, of course, comes the need for simple ways to sift through and analyze it. SFARI Gene categorizes each gene according to how it is linked to autism — whether it is a highly penetrant variant, confers only a small risk or plays a role in other genetic syndromes. (Some genes belong to more than one category.) The database also indicates the type of variation observed — whether a particular gene is affected by single-nucleotide polymorphisms, deletions or insertions, for instance. And it reflects the strength of each genetic link by listing the number of studies that show positive versus negative associations with autism. "It allows you to review the genes as they have appeared in the literature in a fair and weighted way," says Eric Morrow, director of the Developmental Disorders Genetics Research Program in Psychiatry at the Brown University Medical School.
A database wouldn't be complete without references. "One of the things we are trying to do is to really keep up with the current literature," Banerjee Basu said. Each gene entry includes a primary PubMed reference — the first published study that linked the gene to autism — and a list of additional references. MindSpec performs daily PubMed searches, and its experts review new references before adding them to the database. SFARI Gene users can also add new references themselves; the organization adds the references after checking them. The software includes references to studies that elucidate important details about the gene, even if the studies do not relate specifically to autism. In Morrow's experience, SFARI Gene provides "an accurate reflection of the literature."
A host of new features are in the works for SFARI Gene. In the fall of 2009, SFARI Gene will include genetic information gleaned from animal models. Animal model work "is really diverse and wide-ranging," said Packer, so a centralized, searchable and updatable database on animal model findings will "be a tremendous contribution."
Also in the works is a way to characterize complex interactions between proteins encoded by autism-linked genes. A particular gene entry may soon include links to entries for related genes and even visualizations of interaction networks. "As we learn more and more in molecular biology," Banerjee Basu explained, it’s becoming clear that "nothing really works in isolation. A cellular state is driven by combined interaction of many different components."
Packer also hopes to develop a "scoring" system for each gene to better indicate the strength of its relationship to autism based on the published literature. It would be extremely helpful to find a way to rank or score genes so that researchers can put various genetic contributions into perspective more easily, he adds.
Eventually, Packer envisions that SFARI Gene will not only be useful to researchers who are looking for specific information about individual genes; he hopes that it will also generate novel research ideas. Scientists could, for instance, use SFARI Gene to identify pathways or networks of genes that have never before been studied in the context of autism. "We do want to make this more than a static repository of data," Packer said. "We'd like it to be a way to generate new hypotheses that could be tested down the road." Morrow, for instance, has used SFARI Gene to check whether genes that play a role in other types of intellectual disabilities have ever been implicated in autism.
SFARI Gene is designed to be adaptable because the needs of the database may change as researchers learn more about the genetics and molecular logic of autism. "Our goal is to give the most current view of research data on this complex disorder," Banerjee Basu said, "and let science take its course."