New Chair, Heidi Rehm, on the Future Focus of the Global Alliance for Genomics and Health
By Allison Proffitt
October 31, 2023 | In January, Heidi Rehm will take over the leadership of the Global Alliance for Genomics and Health, GA4GH. Rehm will replace out-going Chair Ewan Birney, who has held the Chair role since 2016. Rehm has served as a GA4GH Vice-Chair since 2018; she was elected by the GA4GH Board of Directors in September.
Rehm sat down with Stan Gloss to talk about the mission of GA4GH and areas of future focus for the latest episode of the Trends from the Trenches podcast.
The Global Alliance for Genomics and Health was conceived to enable data sharing to advance genomics and health both for research and medicine, Rehm explained. “Really we’re focused on the building of standards and policies that can enable global data sharing so that we can advance genetic medicine, genomic research,” she said. The Global Alliance is organized into work streams, areas of work such as the cloud work stream, the data security work stream, data use and researcher identity work stream, regulatory and ethics work stream, and many more. “We’ve really worked in a lot of different areas!” she said,
In every work stream, the Global Alliance is addressing the granular questions of how data are shared in a secure way in an interoperable format so that analyses and datasets can be combined to further research. For example, in the regulatory and ethics work stream, a team has focused on consent. “A lot of work has gone into the development of consent templates that others can then use in their own research that then allows that downstream sharing to be able to happen,” Rehm said. “That’s a policy framework that is required to enable use of data, as opposed to an actual standard for how the data gets shared.”
The Global Alliance is committed to practical advances, Rehm said, not just standards creation, and convenes its community to discuss the practical needs of standards implementation, troubleshooting the process along the way. “It’s not just, ‘Can we develop standards and stick them out there and assume that everyone is going to figure out how to use them?’ We really have to convene people together and have them actively work together on implementing those standards and revising them when they’re not working,” Rehm said. “That learning and cross-fertilization of approaches is really critical to allowing everybody to advance together as opposed to living in silos.”
These sorts of foundational standards and projects have taken the group’s attention since its founding in 2013. Another area of focus for the organization in recent history has been sustainability of the organization. “In our early phase, we really just had to figure out how to work together, how to convene, what were our needs, and begin developing the infrastructure and the funding,” Rehm said. While Global Alliance work streams and driver projects are fueled by volunteers, Rehm pointed out that staff is essential to organizational survival. “Consortia across countries, across everything don’t survive if you don’t have dedicated funding and a committed staff that can carve out their time. We have many full-time employees that are the critical underpinning of the Global Alliance.”
Under Rehm’s leadership, the next decade for the Global Alliance will turn much of its attention to the implementation of the standards the organization has been developing in its first decade. “We’ll continue to build more standards; more are needed. There’s still lots of work in that space to be done,” Rehm said. “But we’ll really start convening to define the implementations, figure out where our gaps are, and execute on the work we’re doing to actually solve the real-world problems in genomics and health, research and medicine.”
She listed three particular use cases she’s eager to address.
There are many groups around the globe generating very large datasets who are willing to share those data, Rehm said. For example, she mentioned the large datasets generated by the UK Biobank and the NIH’s All of Us research program in the US. Researchers from all over the world would love to work with data from both datasets, but accessing and combining the data is difficult. “Within countries, we have our own systems [of researcher authentication]—in the US, it’s the use of dbGaP and eRA Commons and that system—but that is not present around the world. How do we authenticate a researcher that’s coming from a non-US site to use US data?” Rehm asks. The Global Alliance’s Passports specification addresses this issue. “But to do the actual analyses we need to be able to send analytical approaches to a dataset, retrieve back the results, combine them and aggregate them with other results,” she said. “The standards to send out, do those analyses, bring back… Those are some of the areas we are really working on.”
As another example, she cites an implementation project using Data Connect, a standard for discovery and search of biomedical data. Building off Matchmaker Exchange, a GA4GH driver project, the new variant-level matching project queries large datasets for a specific variant and returns phenotype to the researcher. “This is critical in rare disease analyses where we’re looking for incredibly rare variants that might have just a handful of individuals around the world with important information to learn from just those few individuals,” Rehm said. “So how can we connect databases to return both the existence of a genotype but the associated metadata—like phenotype—with that.”
Finally, as Chief Genomics Officer at Massachusetts General Hospital and Co-Director of the Program in Medical and Population Genetics at the Broad Institute of MIT and Harvard, Rehm works closely with clinical labs. Connecting data streams between the clinic and research is another priority.
“We are recognizing the need to engage more deliberately with our clinical communities, and that includes clinical laboratories, healthcare systems, the software systems that support healthcare like the electronic health record, and providers. All of those individuals are critical, both because there’s a lot of data being generated in the clinical community and that needs to be part of the research ecosystem, [and] we also want to better translate the results of the research into the clinical arena. That bi-directional tight-knit engagement is critical to the work that we do in genomics and health. We have to better support that clinical side of things.”