|
Today's cancer researchers, whether they are combing through animal genomes or collecting medical histories from human patients, are in the midst of an information explosion. New techniques in biomedical research create huge amounts of data, greatly increasing the chance of scientific breakthroughs but making those advances difficult to achieve. The cancer Biomedical Informatics Grid™ project is developing a computing infrastructure to help scientists accelerate progress in cancer prevention and treatment by synthesizing, standardizing and analyzing all that data.
The caBIG™ pilot project, funded by the National Institutes of Health and coordinated by the National Cancer Institute, bridges people and institutions as well as technologies and tools.
"We've taken on all the major organization and social challenges of getting a fairly large community of geographically separated people and institutions to work together," says Chief Operating Officer Peter Covitz from the NCI Center for Bioinformatics.
Image Courtesy caBIG™ |
Since the beginning of the three-year caBIG™ pilot program in February 2004, that community has swelled to about 500 researchers from institutions across the United States. Developers have focused on several areas of concern for cancer research, including standardization of data formats and tools, ontologies and semantics. Scientists' interactions with the healthcare community also create a unique set of security and privacy challenges for a grid infrastructure.
"Maintaining privacy is a societal challenge, not just a technological one," explains Covitz. "Cancer is a disease of genes, and sharing genetic data is controversial because the right person with the right analysis technology could conceivably identify someone based on what would seem like a fairly arcane set of numbers."
While some caBIG™ community members have been working on the policy challenges of sharing medical data between institutions worldwide, others are focused on creating and adapting security technologies to solve the problem of sharing medical data on the caGrid, caBIG™'s grid computing infrastructure.
The caGrid test bed has been operating since 2005, and now includes 10 nodes, each offering a different data or analytical service. While many of the applications that will be available through the grid are in use today, Covitz's goal for the caGrid is to make it easy—even automatic—for researchers to discover those tools, data and services.
In addition to planning for the first production release of caGrid, Covitz and other caBIG™ leaders are preparing for a transition from pilot project to production project in March 2007.
"Ultimately caBIG™ is about speeding the improvement of cancer diagnosis, treatment and prevention," says Covitz. "The benefits to cancer researchers will emerge gradually as more and more different kinds of software work together. We really want scientists to spend more time doing science, answering cutting-edge questions with computational tools right at their fingertips."
Learn more at the caBIG™ Web site.
—Katie Yurkewicz
e-mail this article
|
