AI for Accelerating Scientific Discovery in Human Disease
AI for Accelerating Scientific Discovery in Human Disease
Embedded Files
We believe AI can dramatically accelerate scientific discovery toward solving the most urgent challenges in human biology - challenges that currently harm countless lives.
Guided by this vision, we seize the opportunities where AI can have the greatest impact, so far working on digital twins for human biology, AI for optimal experiment design, and closed-loop AI-biology systems for automated high-throughput discovery.
Previous Innovation
Applied AI
On the applied AI side we developed SUMMIT: the world’s first self-updating, multiscale platform for biological simulations, designed to unify thousands of causal connections from scientific literature in minutes rather than weeks.
Unlike existing alternatives that are either limited to a single biological scale or require manual updates, our platform is the first to combine both capabilities, allowing users to simulate therapies and visualize complex network effects across a vast biological landscape. By integrating the latest scientific findings at the push of a button, the product enables researchers to identify novel combination therapies and explore biological interactions with unprecedented speed and scope.
Fundamental AI
On the fundamental AI side, we overcame limitations of traditional metabolic simulations by developing Neural Differential Equations for metabolic simulations, with our approach achieving a 100x computational speedup and a 90% accuracy improvement over existing methods.
Members
Andrei is an AI scientist working on AI for accelerating scientific discovery in biomedicine. He received formal training from leading labs in AI and computational biology, having a track-record of innovation in machine learning research (FrODO, axPPO, NDEs).
Now he works as a PhD researcher and leads the Griffin Labs research organization where he coordinated the successful execution of SUMMIT and NDEs for simulating metabolomics.
Elias is an AI engineer and longevity advocate. He made numerous open-source contributions on AI for science and agentic workflows, all while completing his AI BSc with summa cum laude in only two years.
He played a crucial role in turning SUMMIT from an idea to a scalable user-centric platform - as the leading software developer, he implemented the agentic workflow and user interface, optimizing latency and scalability.
Udesh is a PhD researcher and host of the One Deeper podcast, now working on fundamental machine learning research for tackling novel problems in biomedicine.
He played a crucial role in turning NDEs for simulating metabolomics from an intuition into a rigorous scientific study in which we showcased the promise of physics-informed neural networks for simulations in biomedicine.
Contact
Contact
Reach us on linkedin.com/company/griffin-labs-ai
Page updated
Google Sites
Report abuse