AI Model Decodes Cell Signaling Fingerprints Across Diverse Cell Types in Embryonic Development

IRIS Identifies Universal Cell Signaling Patterns

Researchers at Whitehead Institute led by Pulin Li and graduate student Nicholas Hutchins have developed IRIS, an AI model that decodes cell signaling fingerprints across diverse cell types during embryonic cell development1

1

2

2

. Published in Nature Methods on September 8, this neural network-based model marks a significant advance in understanding how embryonic cells receive chemical signals that determine their fate. The breakthrough discovery reveals that each signaling pathway leaves behind a unique fingerprint—a distinctive pattern of gene activity—that remains consistent across different cell types for the same signaling pathway1

1

.

This consistency enables scientists to reconstruct signaling histories across many cell types using pathway-specific fingerprints, eliminating the need to map each pathway separately in every cell type. "Think of voice recognition systems like Siri, which are trained mainly in English, but then use that training to help them recognize other languages," explains Pulin Li, assistant professor of biology at MIT. "This is called transfer learning, and this is why IRIS can work across many different cell types"2

2

.

Training IRIS on Human Embryonic Stem Cells

Li and Hutchins trained IRIS on a large experimental dataset measuring how thousands of human embryonic stem cells responded to dozens of combinations of six major signaling pathways at multiple stages of development1

1

. This comprehensive atlas maps how signaling combinations influence cell behavior. The AI model examines a cell's overall gene activity patterns and estimates which signaling pathways were likely active at specific times during development2

2

.

The team tested IRIS using single cells from mouse embryos during gastrulation, a critical stage when cells rapidly branch into different fates. IRIS accurately predicted when and where specific signaling pathways would activate in cells destined to become heart, gut, muscle, and spinal cord tissue1

1

. This variation in signaling molecules guides cells to form the right structures in the right places within developing embryos.

Implications for Stem Cell Engineering and Regenerative Medicine

The ability to decode cell signaling fingerprints opens possibilities for comprehensively mapping the signaling histories of every cell inside a mouse or human embryo at an unprecedented scale2

2

. This AI-driven approach represents a major advance over traditional methods requiring researchers to experimentally test every pathway in every possible cell type—an arduous and painstaking process.

The findings could accelerate stem cell engineering for regenerative medicine and improve organoid development. Once researchers learn the pattern of signals that drives a stem cell to become a specific cell type, they can recreate those signals to control the fate of stem cells in lab or medical settings1

1

. When researchers employed IRIS to identify signals needed to create a cell type critical for lung development, the model predicted that activating a specific signaling pathway would encourage lung-specific development. Experiments in mouse embryos confirmed the prediction2

2

.

This practical roadmap for guiding stem cells into specific, functional cell types could improve the creation of organoids—miniature 3D models that mimic real organs—for studying disease mechanisms and testing new drugs. Scientists can now understand the fundamentals of cell-to-cell communication while gaining tools to retrace the sequence of instructions cells receive, offering insights into how tissues develop and how these processes go awry in disease.