National Database for Autism Research (NDAR): Accelerating Scientific Discovery through Collaborative Bioinformatics

Background:

As autism spectrum disorder (ASD) research increasingly uses computation, informatics, and information technologies, the need for shared conventions and standards has become increasingly obvious and urgent. The use of such common approaches (e.g., including common data formats, vocabularies, ontologies, etc.) is necessary in order for digital data, tools, and resources to work together. When data, tools, and resources can work together, tremendous value is added to the entire research enterprise. And, when common approaches are developed not only for a research community, but by that research community, the chances are good that those conventions and standards will be broadly adopted, and that value will be added to that community’s research efforts.

Objectives:

The National Database for Autism Research (NDAR) is the National Institutes of Health’s (NIH) response to help organize the research community through a bioinformatics platform, helping the community accelerate scientific discovery by: 1) promoting data sharing, 2) helping organize community based solutions that address community based needs and 3) enabling an investigator to effectively communicate detailed research results.

Methods:

NDAR is a collaborative effort with the ASD research community, for which NDAR is designed to serve, and is the result of formal feedback from many research investigators. NDAR is built on the Biomedical Informatics Research Network (BIRN) platform which allows federation with digital tools and data in other research communities, and promotes collaborative research. NDAR is now accepting submission of over 10,000 discrete variables such as common clinical data (e.g. ADI-R, ADOS, etc.), imaging data in a variety of formats, and genomics data including raw, processed, and results data. Data are accepted based upon defined submission schedules relevant to the research ensuring that data are shared at the appropriate times. In 2008, NDAR received initial data submissions from the NIH Autism Centers of Excellence (ACE) grantees. Other researchers are encouraged to participate in 2009.

Results:

NDAR provides an infrastructure to store, search across, and analyze various types of data while also providing longitudinal storage of a research participant’s anonymized information generated by one or more research studies. This allows a researcher to associate a single research participant’s anonymized genetic, imaging, clinical assessment and other information even if the data were collected at different locations or through different studies. Based upon community feedback, NDAR is defining a community-based common data dictionary to be extended and enhanced by the research field. Data submission statistics will be shared at IMFAR.

Conclusions:

NDAR facilitates the formation of a world-wide network of autism researchers, regardless of funding source, by harmonizing data, research tools, and institutions so that autism researchers can collaborate in new and productive ways. By doing so, NDAR gives researchers access to more data than they can collect on their own and provides robust tools to analyze the information, making it easier and faster for researchers to gather, evaluate, and share autism research information from a variety of sources, regardless of where that data may reside.