In an effort to capture results from genome-wide association studies that aren't available in the published literature, a team of UK- and US-based researchers have developed and launched a new resource dubbed WikiGWA, a Wikipedia-style platform that allows scientists to share published and unpublished GWAS results.
Because it captures unpublished data, WikiGWA should help "alleviate publication bias and provide an invaluable resource for researchers interested in first-hand findings," the developers write in the European Journal of Human Genetics.
Besides capturing unpublished findings, WikiGWA provides a forum for researchers to share their results directly with the community rather than leaving that task up to journal publishers, Jie Huang, one of WikiGWA's developers and a co-author on the paper, says.
Huang, a researcher in the Wellcome Trust Sanger Institute's human genetics department, also points out that WikiGWA includes a tool for visualizing association results — software developed at the University of Michigan called LocusZoom — which is a feature not offered by similar GWAS resources.
Researchers can submit results to WikiGWA that are based on models of inheritance, factors of interest, and more for the same GWA study. The platform could also be used to share results from candidate gene studies, although its primary focus will be GWAS results, the developers say.
In the paper, they acknowledge that WikiGWA is one of several resources for sharing GWAS study data, but note that most of these only collect outcomes from published studies and have a rather limited scope in terms of the kinds of data they collect.
WikiGWA's crowdsourcing philosophy is similar to Wikipedia's, but rather than relying entirely on the wiki platform as resources like SNPedia have done, its creators chose to develop the user interface and backend databases themselves, "giving us more flexibility to design a platform that meets users' needs," the paper says.
Those needs include support for sharing and storing large quantities of data as well as linking tools such as LocusZoom software for exploring GWAS results, Huang adds.
Besides, "we [felt] much more comfortable [using our own] code," he adds. "We [wanted] something simple [that] we [could] understand, control, and customize."
Furthermore, "we [wanted] to build a database that … can hold millions of records," Huang says. According to the paper, Wiki-GWA can hold phenotype-genotype -associations for tens of millions of records.
At present, WikiGWA contains nearly 300,000 SNP associations, primarily in cardiovascular and metabolic conditions, and several dozen researchers have signed up to use the resource.