New York-based Polyphron has raised $20 million in seed financing to develop a physical and simulated human tissue platform intended to support drug development and biological research. The company is building infrastructure that combines AI-driven tissue manufacturing, stem-cell biology, autonomous experimentation and computational tissue models.
Polyphron is addressing a growing challenge in AI-enabled drug development: while AI systems can increasingly generate drug candidates, biological hypotheses and experimental designs, testing those predictions against human biology remains comparatively slow and costly. The company aims to create a verification layer through which AI-generated predictions can be evaluated using donor-diverse living human tissue.
Its Tissue Testing Environments comprise two connected components. The Tissue Foundry is an automated physical system that grows human micro-tissues from stem cells using developmental signals involved in natural tissue formation. The Tissue World Model provides a simulation layer trained on longitudinal tissue trajectories to predict responses to genetic, chemical, biological and environmental changes across donors.
Polyphron’s first tissue model focuses on cardiac tissue and is currently in production across multiple donor lines. The company is expanding its donor panel to capture greater biological variation and plans to introduce liver tissue this year, applying the same manufacturing and measurement infrastructure to areas including drug metabolism and liver toxicity.
“AI will not cure all diseases without a verification layer. The cost of producing another new therapeutic hypothesis or intervention in many modalities could soon approach zero but this will not lead to cures until something can answer, ‘What will this intervention actually do to a heterogeneous human biological system over time?’” said co-founder and CEO Matthew Osman. “The missing piece is a scalable environment where models can test predictions against living human biology – and our bet is that tissue is the right abstraction layer for this scalable environment.”
Polyphron ultimately intends its combined physical and simulated systems to predict therapeutic responses across biological diversity before clinical trials, including potential toxicity and differences in patient response. The financing will support development of this infrastructure as the company expands its tissue models and AI capabilities.
Click here for the original news story.
