AlphaFold and Protein Language Models
In a talk hosted by the University of Wisconsin-Madison’s Computational Biology, Ecology, and Evolution (ComBEE) community, Dr. Hannah Wayment-Steele discusses AlphaFold, a model developed by DeepMind that predicts protein structures from amino acid sequences, along with other similar protein language models ESM and EvoFormer. Wayment-Steele describes the model’s architecture, applications in structural biology, and current pitfalls such as with isoforms.
Bio: Dr. Hannah Wayment-Steele is an assistant professor at UW-Madison with a PhD in chemistry from Stanford. Her lab focuses on biomolecular dynamics and mechanisms that underlie biological function and shape evolution, bringing together the fields of data science, deep learning, and NMR spectroscopy.