Researchers at Iowa State University have developed a bioreactor system that uses artificial intelligence to monitor and optimize enzyme production. The project, backed by Novonesis and funded by Schmidt Sciences, ran from 2023 to 2026.
The team built a reinforcement learning agent capable of controlling the bioreactor in real time. To support this, they created a digital twin of the system to test and refine the AI-based control approach. They also built a bioreactor array to collect live data on variables such as enzyme activity and cell concentration, which the AI agent used to learn and improve its decisions.
Monitoring is a persistent challenge in enzyme manufacturing. Most bioreactors can only track basic conditions like temperature, pH, and oxygen levels using standard sensors. Enzyme activity, one of the most important quality indicators, typically requires manual sampling and lab testing. That process is slow and labor intensive.
The new system changes that. It allows manufacturers to continuously and automatically measure enzyme activity and cell density alongside standard parameters. Researchers say this kind of real-time insight could speed up product development and improve consistency, helping new bioprocesses reach the market faster.
An unexpected outcome emerged along the way. The team found that the low-cost reactors built for the project were well suited for teaching. Those units are now being adapted into educational kits to train students in bioprocessing techniques.
Looking ahead, the researchers are working with a private sector partner to develop a more advanced, highly autonomous bioreactor. That next-generation system will include built-in sensors for continuous enzyme activity monitoring, designed for commercial use. The goal is to give biomanufacturers a scalable tool that supports faster, more reliable production as demand for biomanufactured products grows.
The post New sensor tech tracks enzyme activity in real time appeared first on World Bio Market Insights.
