November 3, 2017

Sage Bionetworks is part of a consortium of universities and research groups which has received a five-year, $25 million grant from the National Center for Advancing Translational Science

Sage Bionetworks is part of a consortium of universities and research groups which has received a five-year, $25 million grant from the National Center for Advancing Translational Science

Bringing a Team Science Approach to Research Informatics

November 3, 2017 – Healthcare Informatics recently published a news item about the creation of a new National Center for Data to Health (CD2H). A consortium of universities and research groups has received a five-year, $25 million grant from the National Center for Advancing Translational Science to coalesce and coordinate informatics activities across the Clinical and Translational Science Award (CTSA) Program, a network of more than 50 medical research institutions.

In a follow-up interview, Melissa Haendel, Ph.D., an associate professor in the Department of Medical Informatics and Clinical Epidemiology at Oregon Health & Science University, one of the project leaders, discussed the new center’s approach and goals in more detail.

One area of focus involves data sharing and interoperability, which means advancing work on ontologies, phenotypes and data standards plays a role.

Another goal is to foster a spirit of collaborative innovation and a “social coding” environment for the informatics teams working in the CTSAs. “We want to understand what tools and algorithms are being developed at each site and to see where there are overlaps, gaps and opportunities to benchmark those,” Haendel said. “We also want to see how we can create a coding community across sites.”

The consortium members have always been committed to sharing successes and knowledge transfer across CTSAs, she said, but it has proved challenging. “We are all at different institutions, with different environments, clinical infrastructure and policies. There are a lot of innovations within each CTSA that are not necessarily easy to discover or transfer to another site, so this is an effort to bust those open a bit and learn from each other a little more than we have.”

Haendel noted that CD2H would build upon some of the successes in sharing across CTSAs but in a more innovative and open-ended way.

“I come more from the intersection of basic research informatics and clinical informatics and do a lot of open science,” she said. “We are trying to bring some of those perspectives to the consortium. We can create a more welcoming environment so that it is not just people in the CTSAs that can participate. Anybody in the community, including commercial interests, pharma, other universities and medical centers that might not have a CTSA can participate, too”.

Another area of focus is making it easier to find people with specific expertise and how attribution is given for contributions, she said. “If someone creates an algorithm or shares a data set, we should have mechanisms to track those contributions and value those in the context of the informatics work we do,” Haendel said. “We should also be able to identify experts to help us with specific problems. We have enormous expertise in clinical and basic research informatics across the CTSAs but if you need to do something very specific, we don’t have the easiest way of searching for that person.”

Traditional methods have involved looking for someone who has published a paper, but it takes six to nine months to publish a paper. The content related to informatics is not well represented in the literature anyway, she added. “People are writing code, making clinical instruments and doing analytics, not writing papers. We want to improve the ability to find collaborators and experts for particular areas in computational science.”

The shift to open data, open software, and open resources, is changing the way people work in biomedical informatics, she said. One of Haendel’s colleagues on the project is John Wilbanks, chief commons officer at Sage Bionetworks, who has spent his career advocating for open data innovation systems.

“One of the reasons it takes so long for a drug to get to market, or any finding to lead to a change in clinical care is because of this lengthy cycle we go through of hording our data and hiding our algorithms in order to publish a paper so we can be first and get tenure and promotion. The patients deserve better than that,” Haendel said. “If we can address the attribution and contribution problem, you can be first just by posting it on GitHub or the web.”

As it builds up the team science approach and governance, the CD2H will focus on two thematic areas: rare disease and human health across the lifespan.

CD2H also plans to create an inventory of its training materials to help train the next generation of informaticians to participate in a more open way, she said. Read more.