Future Under Review for caBIG and NCI's Informatics Strategy

By Allison Proffitt  

March 20, 2012 | One year after the board of scientific advisors for the working group on caBIG called for a one-year moratorium on new projects, contracts, and subcontracts by caBIG, and a thorough audit of all aspects of the caBIG budget and expenditures, the Center for Biomedical Informatics and Information Technology (CBIIT) at the National Cancer Institute has a new interim director and is building a new game plan.   

As interim director, George A. Komatsoulis stressed to Bio-IT World that “caBIG is continuing to operate” and, “the NCI intends to remain engaged in a substantial way in the biomedical informatics space.”   

The board of scientific advisors’ (BSA) report in March 2011 identified several areas of strength for caBIG, primarily a continued belief in the value of caBIG’s initial mission, which included standards of data exchange and interoperability. But the BSA report also raised concerns that the organization’s long term vision was driven by technology advances rather than cancer advances; about the lack of caBIG applications for clinicians and basic researchers; and lack of independent oversight. (PDF slide deck of the report)  

“There were a number of areas—interoperability, semantics, data standards, imaging, tissue banking tools—that were considered to be successful and those projects have in fact continued to move forward [since the 2011 BSA report], said Komatsoulis. However, much of caBIG’s portfolio is under review.   

The scientific advisory group is transitioning from looking only at caBIG, to the broader NCI informatics strategy, Komatsoulis said, and the advisory group will be influential in helping, “drive the investments that we’re going to wind up making in caBIG.” The group’s expanded focus will also include electronic health records, new developments in genomics, and other areas. “A very very broad remit,” Komatsoulis noted. “What we are working with them on now is we are asking them to review and provide feedback on proposed investments and the continuation of some of our investments in the caBIG program.”   

Project Adoption  

As part of the reevaluation, Komatsoulis said some caBIG projects or tools will be moved outside of the confines of NCI for open source development and maintenance.   

“Where we intend to focus our efforts over the next several months is on the types of capabilities that were exclusively endorsed by the BSA report: interoperability frameworks and technology data standards,” he said. “The other thing that we intend to do over the next several months is move many of our existing caBIG tools into true open source development projects. They’ve always been released under open source licenses, but in this case it’s our intention to move projects to true open source development projects.”  

The scientific advisory group will be influential in deciding which tools are, “appropriate for continuing NCI-led development and caBIG-led development versus the projects that are ready to go into a full open source environment.”   

Komatsoulis couldn’t yet say which tools will be released. ”We’re still developing the set of criteria that we’re going to use to help us define what becomes an open source development project,” he said. “I think it’s important that as tools hit certain levels of maturity, they get turned over to the open source community to maintain rather than have it be maintained by the government.”  

The first tool to be released is caLIMS v2, a laboratory information management system for which a group of users has “created a small consortium to maintain this product,” Komatsoulis said. But not every project will be “adopted”.   

“Some of these [tools] may go out before there is an indentified community that has decided that they want to adopt it, and some of them may go out when there’s a clear adoptive community,” he said. “Everything’s basically on the table. If the tool or service isn’t providing sufficient value to the community, it’s going to be eliminated.”   

Komatsoulis stressed, though, that these tools were built with community input and wouldn’t be eliminated without consulting the community. “We’re working on ways to access this usage and impact. We’re going to need the support of end users to let us know what is and isn’t working,” he said. “Just as we wouldn’t start one without community input, we wouldn’t stop one without community input.”  

Komatsoulis suggested the still-active caBIG workspaces as the best way to participate in the discussion and give feedback, but also noted that users could communicate with the scientific advisory group members.   

Director Search  

CBIIT is also conducting a director search. Former director Kenneth H. Buetow resigned from NCI in December 2011 (and has since accepted a post at Arizona State University; see, Buetow Moves On from caBIG). The scientific advisory group is making recommendations for a permanent replacement.   

But the Center is not merely biding its time during the search. “Certainly we are not sitting still while we wait for a new director,” Komatsoulis said. “To a certain extent we’re trying to focus on decisions that retain the flexibility for the new director and aren’t going to cause him or her to have to come in and rewind things, but we have a relatively clear set of directions from the BSA report.”   

In addition to the director search and project evaluation, Komatsoulis and CBIIT have been developing a new strategic plan, he said. “We have also begun planning a symposium to be held this year to reassess community needs and develop recommendations for a next-generation biomedical informatics initiative that’s going to build on the investment, the success, and the lessons learned from the initial caBIG activities.”