OpenProtein.AI Expands Access to AI Tools for Protein Design

OpenProtein.AI, founded by MIT alumni Tristan Bepler and Tim Lu, has introduced a no-code platform that allows scientists to design and analyze proteins using artificial intelligence. The company offers open-source models and computational tools for protein engineering, according to MIT News.

The platform provides access to foundation models that help predict protein structure and function. It includes PoET, short for Protein Evolutionary Transformer, which generates related protein sequences and learns from new experimental data without retraining. Researchers can upload their own data, train models, and analyze results through a web interface or API.

OpenProtein.AI makes its tools available to pharmaceutical and biotech companies, while academic researchers can use the platform for free. The company’s technology is already being used by Boehringer Ingelheim to support protein engineering for treatments targeting cancer and inflammatory diseases.

A newer version of its model, PoET-2, delivers higher performance using fewer computational resources. The founders plan to continue developing models that capture dynamic protein behavior and interactions, broadening the use of AI in biological research.