On May 27, the All of Us Research Program launched the beta version of its cloud-based research platform, the Researcher Workbench. It marks the first time that individual-level data from the massive National Institutes of Health precision medicine initiative has become broadly available for study.
The workbench was created by a team at the All of Us Data and Research Center (DRC), led by Vanderbilt University Medical Center, working with the Broad Institute of MIT and Harvard and Verily Life Sciences (a subsidiary of Alphabet, Inc.).
Outlined in a 2015 State of the Union address by President Barack Obama and launched by the NIH in May 2018, All of Us is an historic effort to gather data from one million or more people to accelerate research and improve health. Originally called the Precision Medicine Initiative, All of Us is unprecedented in its scale and scope. In 2016, Congress authorized $1.5 billion over 10 years for the program.
The project aims to provide a new window into the intersection of human biology, lifestyle and the environment. More than 270,000 study participants have completed initial stages of the program, having agreed to share their electronic health records, provided physical measurements, completed the first three participant surveys and donated at least one biospecimen.
The DRC is focused on taking in rich and diverse participant data and making it available for study. This involves cleaning and standardizing the data, building tools and cloud computing capacity to empower scientific analysis, and keeping the data secure and de-identifying it as it’s shared with researchers.
Approved researchers will use the research platform to test hypotheses and derive publishable results.
“The twin challenges we faced when this program was envisioned were to make data easily available to researchers while preserving participant autonomy and data security,” said Dan Roden, MD, Senior Vice President for Personalized Medicine at VUMC and co-principal investigator for the DRC.
“Accordingly, we’ve developed a unique cloud-based platform that is bringing researchers to the data and analysis tools, while ensuring security and appropriate data use. While this approach is now being considered by other precision medicine-oriented biorepositories, All of Us is the first to develop a solution that implements this concept.”
Tools available with the beta launch include: Workspaces, where research teams house projects; Cohort Builder, giving researchers autonomy to create, review, and annotate cohorts in the All of Us dataset; Dataset Builder, where researchers can build and preview a dataset for one or more cohorts by selecting the desired concepts, sets, and values; Notebooks, allowing users with R or Python experience to perform high-powered queries and analysis within the All of Us dataset; and Phenotype Library, offering pre-built cohorts based on published computational phenotype algorithms.
The cloud-based research platform will grow from here, adding new tools and new data, including genomic data and data collected from wearable technology.
“Currently, 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,” said Melissa Basford, MBA, associate director, Vanderbilt Institute for Clinical and Translational Research. “Researchers with a focus on health disparities and underrepresented populations will also find the current dataset useful, given the diversity of participants in All of Us.”
Access is open to researchers affiliated with institutions that have signed a data use agreement with All of Us. Currently, only U.S.-based academic, nonprofit, or health care organizations can enter into these agreements. To gain access, researchers for now must also have an account on the NIH’s online research grants administration hub, eRA Commons.
Also available is the Data Browser, providing interactive views of publicly available summary data from All of Us.
The principal investigator for the DRC is Paul Harris, PhD, professor of Biomedical Informatics at VUMC. More than 100 people from VUMC and the Meharry-Vanderbilt Alliance have contributed to All of Us, among them Robert Carroll, PhD, Robert Cronin, MD, MS, Brad Malin, PhD, Andrea Ramirez, MD, and Consuelo Wilkins, MD, MSCI.
The DRC’s original principal investigator, Joshua Denny, MD, MS, now works at the NIH as chief executive officer of All of Us.
