SE(3) diffusion model with application to protein backbone generation

Abstract

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 R 3 , 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. 1

DiffTopo: Fold exploration using coarse grained protein topology representations

“…The designability of the backbone is evaluated through self-consistency assessment. Drawing inspiration from the research of FrameDiff (Yim et al, 2023b) and ProtDiff , we quantify self-consistency using Cα RMSD (scRMSD, lower values are better) and predicted local distance difference test (pLDDT, higher values are better). To assess the novelty of the backbone, we pick generated backbones with high designability (ScRMSD <3 Åand pLDDT >90) and search them in the entire CATH database and report their highest TMscore, referred to as Max PDB TMscore.…”

Structure‐Based Drug Discovery with Deep Learning**

“…Diffusion models [218] -a family of generative models inspired by non-equilibrium thermodynamics -are also gaining increasing popularity in deep learning thanks to their generative capabilities, and have found pioneering applications in the molecular sciences. [101,127,212,219] These approaches have reached stateof-the-art in several deep learning applications and are expected to propel SBDD in the future.…”

DiffModeler: Large Macromolecular Structure Modeling in Low-Resolution Cryo-EM Maps Using Diffusion Model

“…Notably, the diffusion model has demonstrated considerable success in various areas of image processing, such as image generation [23][24][25][26] , segmentation 27,28 , and translation 29,30 . Additionally, it has found applications in bioinformatics, including protein docking 31 and protein design 32,33 . Building upon these successful applications, DiffModeler integrates the diffusion model to enhance the extraction of structural information, facilitating accurate structure modeling for cryo-EM maps at intermediate resolutions.…”