Adapting protein language models for structure-conditioned design
Tuesday September 17th, 4-5pm EST | Jeff Ruffolo, PhD (Profluent Bio)
Abstract: Generative models for protein design trained on experimentally determined structures have proven useful for a variety of design tasks. However, such methods are limited by the quantity and diversity of structures used for training, which represent a small, biased fraction of protein space. Here, we describe proseLM, a method for protein sequence design based on adaptation of protein language models to incorporate structural and functional context. We show that proseLM benefits from the scaling trends of underlying language models, and that the addition of non-protein context – nucleic acids, ligands, and ions – improves recovery of native residues during design by 4-5% across model scales. These improvements are most pronounced for residues that directly interface with non-protein context, which are faithfully recovered at rates >70% by the most capable proseLM models. We experimentally validated proseLM by optimizing the editing efficiency of genome editors in human cells, achieving a 50% increase in base editing activity, and by redesigning therapeutic antibodies, resulting in a PD-1 binder with 2.2 nM affinity.
Preprint: https://www.biorxiv.org/content/10.1101/2024.08.03.606485v1
Jeff Ruffolo is a Machine Learning Scientist at Profluent Bio, where he develops machine learning methods for functional protein design. He obtained his PhD in biophysics at Johns Hopkins University, where he worked in the lab of Jeffrey Gray. During this time, he developed deep learning tools for antibody structure prediction, language modeling, and representation learning. At Profluent, he has contributed to the OpenCRISPR initiative and led the development of next-generation protein language models with atomistic control for function protein generation.