Many protein design applications, such as binder or enzyme design, require scaffolding a structural motif with high precision. Generative modelling paradigms based on denoising diffusion processes emerged as a leading candidate to address this motif scaffolding problem and have shown early experimental success in some cases. In the diffusion paradigm, motif scaffolding is treated as a conditional generation task, and several conditional generation protocols were proposed or imported from the Computer Vision literature. However, most of these protocols are motivated heuristically, e.g. via analogies to Langevin dynamics, and lack a unifying framework, obscuring connections between the different approaches. In this work, we unify conditional training and conditional sampling procedures under one common framework based on the mathematically well-understood Doob's h-transform. This new perspective allows us to draw connections between existing methods and propose a new variation on existing conditional training protocols. We illustrate the effectiveness of this new protocol in both, image outpainting and motif scaffolding and find that it outperforms standard methods.
MotifBench: A standardized protein design benchmark for motif-scaffolding problems
Shanghai Jiao Tong University(上海交通大学);Tsinghua University(清华大学);Oxford University(牛津大学);NVIDIA(英伟达);Microsoft Research(微软研究院);Massachusetts Institute of Technology(麻省理工学院);University of Washington(华盛顿大学);Stanford University(斯坦福大学)
Improved motif-scaffolding with SE(3) flow matching
Computer Science and Artificial Intelligence Laboratory(计算机科学与人工智能实验室);Massachusetts Institute of Technology(麻省理工学院);University of Oxford(牛津大学);University of Cambridge(剑桥大学);Microsoft Research AI4Science(微软研究院AI4Science部门)
Towards Multiscale Graph-based Protein Learning with Geometric Secondary Structural Motifs
Department of Mathematics and Scientific Computing and Imaging (SCI) Institute University of Utah(数学与科学计算与成像研究所(SCI)大学);Department of Mathematics, UCLA(数学系,加州大学洛杉矶分校);Department of Mathematics, UC Davis(数学系,加州大学戴维斯分校)
Protein-SE(3): Benchmarking SE(3)-based Generative Models for Protein Structure Design
School of Computer Science and Technology, East China Normal University(东华师范大学计算机科学与技术学院);Shanghai Institute of AI for Education, East China Normal University(东华师范大学上海人工智能教育研究所);AI Lab, Research Center for Industries of the Future, Westlake University(西湖大学未来产业研究中心人工智能实验室)
A Fusion-Driven Approach of Attention-Based CNN-BiLSTM for Protein Family Classification -- ProFamNet
School of Computer Science and IT, Institute of Management Sciences(管理科学学院计算机科学与信息技术学院);National University of Computer and Emerging Sciences(国家计算机与新兴科学大学);Ghulam Ishaq Institute of Emerging Sciences(古拉姆·伊沙克新兴科学学院)
3D Molecule Generation from Rigid Motifs via SE(3) Flows
School of Computation, Information and Technology, Technical University of Munich, Germany(计算、信息与技术学院,慕尼黑技术大学,德国);Munich Data Science Institute (MDSI), Technical University of Munich, Germany(慕尼黑数据科学研究所(MDSI),慕尼黑技术大学,德国);Munich Center for Machine Learning (MCML), Munich, Germany(慕尼黑机器学习中心(MCML),慕尼黑,德国)