At a glance
Portal provides free, open access to most comprehensive atlas of cell, tissues, and organ systems of healthy human bodies available in a single resource.
Online visualization and analysis tools allow global research community to examine 9,000 datasets from 500 donors collected in an NIH-funded project.
Healthy baseline data will empower human and AI research into health and disease.
The future of federally funded research at Harvard Medical School — supported by taxpayers and done in service to humanity — remains uncertain. Learn more.
Researchers and data scientists from dozens of universities, hospitals, and other institutions across the United States have built an online portal that provides free, open access to one of the most comprehensive atlases of the cells, tissues, and organ systems of healthy human bodies.
The portal, funded by the federal government and described Nov. 2 in Nature Communications, promises to empower human researchers and artificial intelligence efforts to understand human health and disease.
The portal is the online home of the data collected by the Human BioMolecular Atlas Program (HuBMAP), an eight-year effort funded by the National Institutes of Health Common Fund involving more than 600 consortium members at over 60 institutions. The atlas was compiled from more than 9,000 datasets from more than 500 healthy donors across dozens of organs and tissues — a combination of scope and scale not currently available in any other single resource, the researchers said.
The HuBMAP Data Portal includes web-based tools that allow the global research community to rapidly analyze visual, spatial, and biomolecular data from the collection. Users can view data by organ system, search gene expression patterns and molecular signatures across tissues, and explore single-cell data across diverse populations of cells.
“The HuBMAP program has generated an enormously rich resource for deepening our understanding of human health and disease,” said senior author Nils Gehlenborg, professor of biomedical informatics in the Blavatnik Institute at Harvard Medical School, principal investigator of the Humans in Data Integration, Visualization, and Exploration (HIDIVE) Lab at HMS, and one of the leaders of the portal project.
“The portal gives users access to the data and also provides tools they can use to explore and analyze the complex relationships between the trillions of cells that make up each healthy human body in a way that’s never been possible before,” Gehlenborg said.
A healthy baseline to help understand disease
The HuBMAP data complements a growing cohort of biomedical atlases that collect information about the cellular and biochemical status of people with various diseases. Having a healthy sample to compare is necessary to paint a full picture of human health and disease, the researchers said.
“To understand disease, you first need to know what healthy looks like,” said first author Morgan Turner, director of scientific operations in the HIDIVE Lab.
The portal brings together more than 25 types of molecular and imaging data from dozens of healthy human organs and tissues, Turner said. Because the data follow community-driven, FAIR data and metadata standards, they can be compared across labs, organs, and donors.
“Those standards are also what power the portal’s visualization and analysis tools and make the data ready for AI,” Turner said.
Work on the portal was led by researchers at HMS, the Pittsburgh Supercomputing Center, the University of Pittsburgh, Carnegie Mellon University, the University of Pennsylvania, and Stanford University. The team’s role included making sure that the HuBMAP samples, which had been captured with diverse tools and technologies, could be compared in a way that was scientifically sound.
To do this, the portal team worked with other consortium members to develop community-driven data and metadata standards for contributors to follow when sharing data. The team then built standardized processing pipelines so that molecular and imaging data from many labs and technologies are uniformly processed and can be compared.
Data, visualization, and analysis
The team also designed the portal so that researchers can get a sense of whether the atlas contains information that would help them answer their questions without first downloading the massive datasets or using heavy compute resources.
For more advanced analyses, users can download data directly and use the portal’s openly available metadata and programming interfaces. Access to protected data, such as sequencing data that could identify donors, requires an approval process. Pitt and the PSC are leading the efforts to build and maintain computing infrastructure to power the portal and to ensure ongoing access to this resource.
Authorship, funding, disclosures
Additional authors include Thomas C. Smits, Tiffany S. Liaw, Brendan Honick, Bill Shirey, Lisa Choy, Nikolay Akhmetov, Shaokun An, David Betancur, Dominic Bordelon, Karl Burke, Ivan Cao-Berg, John Conroy, Chris Csonka, Penny Cuda, Sean Donahue, Stephen Fisher, Derek Furst, Ed Hanna, Josef Hardi, Tabassum Kakar, Mark S. Keller, Devin Lange, Xiang Li, Yan Ma, Alison McWilliams, Austen Money, Richard Morgan, Eric Moerth, Juan Muerto, Mark A. Musen, Emily Nic, Martin J. O’Connor, Gesina Phillips, Alexander J. Ropelewski, Ryan Sablosky, Sravani Saripalli, Max Sibilla, Derek Simmel, Alan Simmons, Xu Tang, Joel Welling, Zhou Yuan, Martin Hemberg, HuBMAP Consortium, Matthew Ruffalo, Jonathan Silverstein, and Philip Blood.
A complete list of consortium authors can be found in the paper.
This research is supported by the NIH Common Fund through the Office of Strategic Coordination/Office of the NIH Director (awards OT2OD033758, OT2OD033759, OT2OD033761, and U54HD104392).
Gehlenborg is a co-founder and equity owner of Datavisyn. Ropelewski has equity interests in Illumina and 10x Genomics.
