This protein-engineering breakthrough generates over 10M data points and turbocharges AI in just three days - Phys.org

Researchers have developed a high-throughput method called Sequence Display that can generate over 10 million protein activity data points in a single experiment. According to a Phys.org report, this provides sufficient data to train AI models for protein engineering applications. The method significantly accelerates data collection, compressing what previously took much longer into approximately three days.

The technique enables rapid generation of large-scale datasets on protein functions. These datasets can be used to train artificial intelligence systems to predict and design proteins with specific properties. The researchers indicate this could lead to advancements in developing new enzymes, therapeutics, and materials.

The breakthrough represents a substantial increase in efficiency for biological data generation. By providing massive datasets quickly, it potentially transforms the pace of discovery in protein engineering. The method's development comes amid growing interest in applying AI to biological design challenges.