AbDiffuser: full-atom generation of in-vitro functioning antibodies
Karolis Martinkus, Jan Ludwiczak, Wei-Ching Liang, Julien Lafrance-Vanasse, Isidro Hötzel, Arvind Rajpal, Yan Wu, Kyunghyun Cho, Richard Bonneau, Vladimir Gligorijevic, Andreas Loukas
摘要
We introduce AbDiffuser, an equivariant and physics-informed diffusion model for the joint generation of antibody 3D structures and sequences. AbDiffuser is built on top of a new representation of protein structure, relies on a novel architecture for aligned proteins, and utilizes strong diffusion priors to improve the denoising process. Our approach improves protein diffusion by taking advantage of domain knowledge and physics-based constraints; handles sequence-length changes; and reduces memory complexity by an order of magnitude, enabling backbone and side chain generation. We validate AbDiffuser in silico and in vitro. Numerical experiments showcase the ability of AbDiffuser to generate antibodies that closely track the sequence and structural properties of a reference set. Laboratory experiments confirm that all 16 HER2 antibodies discovered were expressed at high levels and that 57.1% of the selected designs were tight binders.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper30
- Protein Conformation Generation via Force-Guided SE(3) Diffusion ModelsYan Wang, Lihao Wang, Yuning Shen, Yiqun Wang 等ICML 2024 · 被引用 65 次
- Protein Discovery with Discrete Walk-Jump SamplingNathan C. Frey, Daniel Berenberg, Karina Zadorozhny, Joseph Kleinhenz 等ICLR 2024 · 被引用 54 次
- Full-Atom Peptide Design with Geometric Latent DiffusionXiangzhe Kong, Yinjun Jia, Wenbing Huang, Yang LiuNeurIPS 2024 · 被引用 48 次
- Antigen-Specific Antibody Design via Direct Energy-based Preference OptimizationXiangxin Zhou, Dongyu Xue, Ruizhe Chen, Zaixiang Zheng 等NeurIPS 2024 · 被引用 48 次
- Harmonic Self-Conditioned Flow Matching for joint Multi-Ligand Docking and Binding Site DesignHannes Stärk, Bowen Jing, Regina Barzilay, Tommi S. JaakkolaICML 2024 · 被引用 43 次
它引用的顶会 Paper22
- Denoising Diffusion Probabilistic ModelsJonathan Ho, Ajay Jain, Pieter AbbeelNeurIPS 2020 · 被引用 35,902 次
- Improved Denoising Diffusion Probabilistic ModelsAlexander Quinn Nichol, Prafulla DhariwalICML 2021 · 被引用 5,234 次
- MLP-Mixer: An all-MLP Architecture for VisionIlya O. Tolstikhin, Neil Houlsby, Alexander Kolesnikov, Lucas Beyer 等NeurIPS 2021 · 被引用 3,862 次
- Structured Denoising Diffusion Models in Discrete State-SpacesJacob Austin, Daniel D. Johnson, Jonathan Ho, Daniel Tarlow 等NeurIPS 2021 · 被引用 2,256 次
- E(n) Equivariant Graph Neural NetworksVictor Garcia Satorras, Emiel Hoogeboom, Max WellingICML 2021 · 被引用 1,432 次
相关 Paper
- Antibody Design Using a Score-based Diffusion Model Guided by Evolutionary, Physical and Geometric ConstraintsTian Zhu, Milong Ren, Haicang ZhangICML 2024 · 被引用 13 次
- Antigen-Specific Antibody Design and Optimization with Diffusion-Based Generative Models for Protein StructuresShitong Luo, Yufeng Su, Xingang Peng, Sheng Wang 等NeurIPS 2022 · 被引用 331 次
- Repurposing AlphaFold3-like Protein Folding Models for Antibody Sequence and Structure Co-designNianzu Yang, Songlin Jiang, Jian Ma, Huaijin Wu 等NeurIPS 2025 · 被引用 3 次
- Dynamic Geometric Equivariant Network for Full-Atom Antibody DesignWeihong Huang, Feng Yang, Qiang Zhang, Juan LiuAAAI 2026
- Towards All-Atom Foundation Models for Biomolecular Binding Affinity PredictionLiang Shi, Zuobai Zhang, Huiyu Cai, Santiago Miret 等ICLR 2026
