FrameDiPT: SE(3) Diffusion Model for Protein Structure Inpainting
Authors:
- Cheng ZhangCZCheng Zhang
- Adam LeachALAdam Leach
- Thomas MakkinkTMThomas Makkink
- Miguel ArbesúMAMiguel Arbesú
- Ibtissem KadriIKIbtissem Kadri
- Liron MizrahiLMLiron Mizrahi
- Sabrine KrichenSKSabrine Krichen
- Maren LangMLMaren Lang
- Andrey TovchigrechkoATAndrey Tovchigrechko
- Nicolas Lopez CarranzaNCNicolas Lopez Carranza
- Uğur ŞahinUŞUğur Şahin
Preprint
Abstract
Protein structure prediction field has been revolutionised by deep learning with protein folding models such as AlphaFold 2 and ESMFold. These models enable rapid in silico prediction and have been integrated into de novo protein design and protein-protein interaction (PPI) prediction. However, biologically relevant features dependent on conformational distributions cannot be estimated with these models. Diffusion models, a novel class of generative models, have been developed to learn conformational distributions and applied to de novo protein design. Limited work has been done on protein structure inpainting, where a masked section is recovered by simultaneously conditioning on its sequence and the rest of the structure. In this work, we propose FrameD iff i n P ain T ing (FrameDiPT), a generalised model for protein inpainting. This is important for T-cells given the hyper-variability of the complementarity determining region (CDR) loops. We evaluated the model on CDR loop design for T-cell receptors and achieved comparable prediction accuracy to ProteinGenerator and RFdiffusion with limited training data and learnable parameters. Different from deterministic structure prediction models, FrameDiPT captures the conformational distribution at different regions and binding states, highlighting a key advantage of generative models. The model and inference code have been released 1 .Keywords
This publication has 23 references indexed in Scilit:
- Antigen-Specific Antibody Design and Optimization with Diffusion-Based Generative Models for Protein StructuresbioRxiv, 2022
- Robust deep learning based protein sequence design using ProteinMPNNbioRxiv, 2022
- Protein complex prediction with AlphaFold-MultimerbioRxiv, 2021
- Accurate prediction of protein structures and interactions using a three-track neural networkScience, 2021
- Highly accurate protein structure prediction with AlphaFoldNature, 2021
- Structural understanding of T cell receptor triggeringCellular & Molecular Immunology, 2020
- STCRDab: the structural T-cell receptor databaseNucleic Acids Research, 2017
- A Connection Between Score Matching and Denoising AutoencodersNeural Computation, 2011
- Conformational changes and flexibility in T-cell receptor recognition of peptide–MHC complexesBiochemical Journal, 2008
- Full Activation of the T Cell Receptor Requires Both Clustering and Conformational Changes at CD3Immunity, 2007