Skip to yearly menu bar Skip to main content

Workshop: Machine Learning in Structural Biology Workshop

Scalable Multimer Structure Prediction using Diffusion Models

Peter Pao-Huang · Bowen Jing · Dr. Bonnie Berger


Accurate protein complex structure modeling is a necessary step in understanding the behavior of biological pathways and cellular systems. While some works have attempted to address this challenge, there is still a need for scaling existing methods to larger protein complexes. To address this need, we propose a novel diffusion generative model (DGM) that predicts large multimeric protein structures by learning to rigidly dock its chains together. Additionally, we construct a new dataset specifically for large protein complexes used to train and evaluate our DGM. We substantially improve prediction runtime and completion rates while maintaining competitive accuracy with current methods.

Chat is not available.