AI-Powered Lab Automates Protein Engineering, Showing Promise for Expediting Scientific Discovery
-
A self-driving lab with AI and robots successfully engineered new proteins without human input, showing promise for automating scientific discovery.
-
The system relies on a machine learning model that proposes new protein sequences to test based on previous results.
-
In an experiment, the autonomous lab engineered enzymes to be stable at much higher temperatures.
-
Challenges remain in making the hardware and software more sophisticated to tackle more complex protein engineering problems.
-
The goal is to automate monotonous scientific tasks, not replace human scientists, so researchers can focus on higher-level work.