The HuBMAP Framework for Advancing Data FAIRness

Area of Science:

• Biomedical research
• Data science
• Bioinformatics

Background:

• The FAIR Guiding Principles (Findable, Accessible, Interoperable, Reusable) aim to enhance data sharing and reuse in science.
• Implementing FAIR principles in scientific workflows is difficult without standardized infrastructure.
• The NIH Human Bio-Molecular Atlas Program (HuBMAP) manages extensive, diverse datasets across multiple institutions and assay types.

Purpose of the Study:

• To develop and implement a standardized, FAIR-compliant data ecosystem within the HuBMAP consortium.
• To create community-endorsed metadata reporting standards to ensure data quality and interoperability.
• To establish a model workflow for generating and disseminating FAIR data for broad scientific use.

Main Methods:

• Developed detailed, harmonized metadata reporting standards and schemas across diverse assays.
• Implemented these standards throughout the research lifecycle, from data collection to packaging.
• Utilized technology to ensure adherence to standards and compliance with regulations like HIPAA.

Main Results:

• Successfully operationalized FAIR data principles through a metadata-centered workflow.
• Generated over 10,000 FAIR datasets from more than 40 institutions, covering over 50 assay types.
• Established a Data Portal and Human Reference Atlas for open dissemination of FAIR data.

Conclusions:

• HuBMAP's metadata reporting standards and workflow effectively achieve data FAIRness.
• The developed procedures provide a replicable model for other scientific communities seeking to maximize data value.
• The HuBMAP workflow is available as open-source technology, facilitating adoption by other consortia like SenNet.