ML/AI Tools to Advance Genomic Translational Research (MAGen)

The need for increased research to develop tools to translate genomic research findings into clinical applications has been highlighted by NHGRI-convened workshops. ML/AI techniques have successfully identified novel patterns in non-genomic clinical datasets, receiving FDA approval for ML/AI-enabled medical devices in fields such as radiology, cardiology, and various other disciplines of medicine. Vast amounts of multimodal data are available through advances in FAIR sharing standards and policies, facilitating the potential of using ML/AI approaches in translational genomics. There is a need to utilize advances in technologies to translate genomic research findings into tools for potential future clinical applications that is we need tools that bridge basic and implementation research referred to in this initiative as, genomics translational research tools developed with disease expert involvement from the design through development and testing. Maximizing the potential of ML/AI requires thoughtful attention to the ethical, legal, and social implications (ELSI) of its use in genomic medicine and healthcare more broadly. The long-term vision for successful tools would be to test them in clinical implementation settings.

NHGRI aims to establish a research Consortium, ML/AI Tools to Advance Genomic Translational Research (MAGen), to collaboratively explore the feasibility of Machine Learning (ML) and Artificial Intelligence (AI) tools that can enhance the accuracy and precision of predicting how individuals with pathogenic genetic variants manifest disease. The ML/AI tools will leverage existing multimodal genomic and non-genomic data and will be cross validated in genomic translational research settings to ensure the robustness and generalizability of the tools for translational purposes. Importantly, the Consortium will explore the ethical, legal, and social implications (ELSI) of integrating ML/AI tools into genomic medicine through the establishment of an ELSI Framework for their development, and through implementation of ELSI research projects. 

The Consortium will compose of MAGen Development Sites with multi-disciplinary teams to develop the ML/AI tools and the MAGen Coordinating Center which will be responsible for technical, scientific, and administrative coordination activities to support the Consortium research activities including development of Consortium wide plans, coordinated ELSI-informed ML/AI tool development and cross validation, and dissemination of the developed resources.

• From the September 18th, 2023, NHGRI Advisory Council Meeting
• Concept Clearance Presentation
• Concept Document (PDF)

MAGen NOFO Webinar
June 26, 2024

• Frequently Asked Questions