Autonomous discovery of cell growth kinetics using ai agents
Autonomous Structural Discovery Of Bioprocess Dynamics And Kinetics Via LLM-Driven Agents
Artificial Intelligence
Summary
Understanding how cells grow and react in manufacturing processes is hard because the rules they follow are complicated or unknown. The authors created a system where an artificial intelligence agent autonomously figures out these hidden rules by running experiments and analyzing data. This method breaks down complex interactions into simpler parts and triggers new tests to clarify uncertain areas. They tested this on a chemical reactor and accurately identified the true underlying processes without needing prior knowledge.
What this means in practice
- •For biomanufacturing engineers: Automatically identify and refine kinetic models for continuous cell culture processes without pre-existing kinetic data.
- •For chemical process control teams: Dynamically design experiments to resolve uncertainties in reaction kinetics during reactor operation.
Authors
Erdal Aydin, Yaman Arkun
Abstract
High-fidelity digital twins in continuous biomanufacturing are often bottlenecked by incomplete kinetic models, while standard data-driven algorithms (e.g., SINDy, PINNs) lack interpretability or impose rigid physical assumptions. This paper introduces a hybrid closed-loop architecture for autonomous kinetic state-space discovery driven by a Large Language Model (LLM) agent. The framework employs a sequential "isolation-peeling" methodology across orthogonal operational regimes. This algorithmic peeling mathematically decouples entangled multiplicative (inhibitory) and additive (catalytic) biological rate law. Grounded in a Two-Level Retrieval-Augmented Generation (RAG) library, the framework utilizes Moving Horizon Estimation (MHE) and logarithmic decomposition to decouple complex cell growth kinetics into orthogonal multiplicative (inhibitory) and additive (regulatory) residual channels. When statistical degeneracy occurs an Active Design of Experiments (DoE) controller dynamically triggers targeted perturbations to decorrelate competing model hypotheses. Evaluated on a continuous stirred-tank reactor exhibiting unmodeled Haldane inhibition, Luong product toxicity, and secondary metabolite autocatalysis, the framework autonomously expanded the state space and recovered ground-truth parameters with under 6% error without a priori kinetic knowledge.