SE(3) diffusion model with application to protein backbone generation
This addresses protein engineering for biomedicine and chemistry by providing a principled method for generating novel protein backbones, though it builds on prior diffusion models in this domain.
The paper tackles the challenge of designing novel protein structures by developing FrameDiff, a diffusion model on SE(3) that generates protein backbones without relying on pretrained prediction networks, achieving designable monomers up to 500 amino acids and generalization beyond known structures.
The design of novel protein structures remains a challenge in protein engineering for applications across biomedicine and chemistry. In this line of work, a diffusion model over rigid bodies in 3D (referred to as frames) has shown success in generating novel, functional protein backbones that have not been observed in nature. However, there exists no principled methodological framework for diffusion on SE(3), the space of orientation preserving rigid motions in R3, that operates on frames and confers the group invariance. We address these shortcomings by developing theoretical foundations of SE(3) invariant diffusion models on multiple frames followed by a novel framework, FrameDiff, for learning the SE(3) equivariant score over multiple frames. We apply FrameDiff on monomer backbone generation and find it can generate designable monomers up to 500 amino acids without relying on a pretrained protein structure prediction network that has been integral to previous methods. We find our samples are capable of generalizing beyond any known protein structure.