Chemical Engineering Seminar

Artificial intelligence and machine learning are reshaping protein science by revealing patterns that connect protein sequences to their structures and functions at unprecedented scale. These predictive models allow researchers to navigate vast protein sequence spaces and identify molecules with useful properties for applications in medicine, energy, and biotechnology. In this talk, I will describe my group's work developing data-driven models that learn sequence–function relationships directly from large experimental datasets and use them to design improved proteins that go beyond previously observed examples. I will also highlight our efforts to build fully autonomous "self-driving laboratories" that combine AI-based decision making with robotic experimentation, enabling proteins to be designed, tested, and optimized without human intervention. Together, these approaches illustrate a new paradigm for protein engineering in which artificial intelligence and experimentation are tightly integrated to accelerate scientific discovery.