“All of Us” Opens Precision Medicine Data Platform for Beta Testing

The All of Us precision medicine research program has opened its data platform for beta testing, allowing investigators to offer feedback on its preliminary resources.

The All of Us Researcher Workbench will ultimately help support studies on many different aspects of health and disease, leading to more personalized treatments and prevention strategies going forward.

“Now, researchers can begin using our initial dataset and tools in studies and tell us what’s working and what we can improve. This moment is an important step in our effort to accelerate new discoveries,” Josh Denny, MD, MS, chief executive officer of the All of Us Research Program, wrote in the announcement.

The early version of the Researcher Workbench includes data shared from nearly 225,000 of the first participants in the project, 75 percent of whom are from communities historically underrepresented in research and more than 45 percent of diverse races and ethnicities.

EHR information; data from six initial surveys on demographics, lifestyle factors, and overall health; and baseline physical measures are also included in the initial platform. Researchers can use tools that can help them set up collaborative workspaces and build customized cohorts.

Research teams will need experience with R or Python programming languages to conduct analyses on the platform, as the program does not yet support integrations with other statistical software. For future versions of the Researcher Workbench, the All of Us team is working to expand analysis tools.

The platform will grow more comprehensive over time as the project adds new data and tools, but the current version has some key limitations.

“Because participants take part in the program at different paces and we are still enrolling, we don’t have variables for all participants; in particular, survey completion rates vary, and the collection and harmonization of electronic health record data remain a work in progress. We have done some preliminary testing on biological plausibility of the data; other curation efforts are still underway,” Denny noted.

The All of Us group has also taken precautions to protect patient privacy, Denny said.

“While we already strip out names and other identifiers from participant data at the outset, we’ve made additional adjustments in the curation process,” he said.

“These include shifting dates and hiding or grouping the records of small clusters of participants to further reduce the risks of reidentification. These modifications may pose challenges for epidemiological studies or research on specific subcategories of people.”

The Researcher Workbench also requires researchers to analyze data within the secure cloud-based All of Us platform. Researchers can’t download individual-level program data on local computers.

The aim is to further refine the data platform and allow for better research in the future.

“As with any beta testing period, there will be technical bugs, and we’ll work through them. We rely on researchers’ feedback to help us identify usability issues, iterate with us to improve the data and tools, and plan future enhancements,” Denny said.

Researchers with NIH eRA Commons accounts may apply for access to the platform if their institutions have signed a data use agreement with the program. Right now, any US-based, academic, non-profit, or healthcare organization can enter into the data use agreement.

After the initial stage of the beta phase, the All of Us team plans to add other means of identity verification beyond eRA Commons and open the platform more broadly.

“Bioinformatics and health services researchers will likely find the most value in our initial dataset, particularly for studies that evaluate the frequency of certain diseases or conditions. Researchers with a focus on health disparities and underrepresented populations will also find the current dataset useful, given its size and diversity,” Denny stated.

Denny pointed out that the Researcher Workbench has adopted a data passport model, which makes the information widely accessible. After researchers register with the program, agree to the rules, and complete training on the responsible conduct of research, the All of Us team will grant them permission to explore the data for a wide range of studies, rather than determining access on a project-by-project basis.

The team will also regularly update its protocol to add new types of data over time, with ongoing data curation throughout the process.

“We extend a warm welcome to researchers interested in exploring the Researcher Workbench. We want to design this platform with you, to make it the best resource it can be. Your comments and suggestions are central to that effort. Thank you in advance for being generous with feedback so we can improve,” Denny concluded.