The biological functions of proteins often depend on dynamic structural ensembles. In this work, we develop a flow-based generative modeling approach for learning and sampling the conformational landscapes of proteins. We repurpose highly accurate single-state predictors such as AlphaFold and ESMFold and fine-tune them under a custom flow matching framework to obtain sequence-conditoned generative models of protein structure called AlphaFlow and ESMFlow. When trained and evaluated on the PDB, our method provides a superior combination of precision and diversity compared to AlphaFold with MSA subsampling. When further trained on ensembles from all-atom MD, our method accurately captures conformational flexibility, positional distributions, and higher-order ensemble observables for unseen proteins. Moreover, our method can diversify a static PDB structure with faster wall-clock convergence to certain equilibrium properties than replicate MD trajectories, demonstrating its potential as a proxy for expensive physics-based simulations. Code is available at https://github.com/bjing2016/alphaflow.
Improving AlphaFlow for Efficient Protein Ensembles Generation
The Chinese University of Hong Kong（香港中文大学）；Shanghai Jiao Tong University School of Medicine（上海交通大学医学院）；Xi’an Jiaotong University（西安交通大学）；Shanghai Institute of Materia Medica, Chinese Academy of Sciences（中国科学院上海药物研究所）
University of Cambridge（剑桥大学）；The National Institute of Agricultural Botany（英国国家农业植物学研究所）；University College London（伦敦大学学院）；Alan Turing Institute（艾伦·图灵研究所）；Chalmers University（查尔姆斯理工大学）
The goal of Protein Structure Prediction (PSP) problem is to predict a protein's 3D structure (confirmation) from its amino acid sequence. The problem has been a 'holy grail' of science since the Noble prize-winning work of Anfinsen demonstrated that protein conformation was determined by sequence. A recent and important step towards this goal was the development of AlphaFold2, currently the best PSP method. AlphaFold2 is probably the highest profile application of AI to science. Both AlphaFold2 and RoseTTAFold (another impressive PSP method) have been published and placed in the public domain…
University of Central Florida（中佛罗里达大学）；University of Texas at San Antonio（德克萨斯大学圣安东尼奥分校）；Air Force Research Laboratory（空军研究实验室）；Argonne National Laboratory（阿贡国家实验室）；SRI International（SRI国际）；Defense Advanced Research Projects Agency（国防高级研究计划局）
Protein folding neural networks (PFNNs) such as AlphaFold predict remarkably accurate structures of proteins compared to other approaches. However, the robustness of such networks has heretofore not been explored. This is particularly relevant given the broad social implications of such technologies and the fact that biologically small perturbations in the protein sequence do not generally lead to drastic changes in the protein structure. In this paper, we demonstrate that AlphaFold does not exhibit such robustness despite its high accuracy. This raises the challenge of detecting and quantifyi…
From Prediction to Simulation: AlphaFold 3 as a Differentiable Framework for Structural Biology
Department of Computer Engineering, Ma.C., Islamic Azad University（计算机工程系，伊斯兰教Azad大学）；Department of Biological Sciences and Technologies, Faculty of Basic Sciences, Islamic Azad University（生物科学与技术系，基础科学学院，伊斯兰教Azad大学）
Protein structures and functions are determined by a contiguous arrangement of amino acid sequences. Designing novel protein sequences and structures with desired geometry and functions is a complex task with large state spaces. Here we develop a novel protein design pipeline consisting of two deep learning algorithms, ProteinSolver and AlphaFold2. ProteinSolver is a deep graph neural network that generates amino acid sequences such that the forces between interacting amino acids are favorable and compatible with the fold while AlphaFold2 is a deep learning algorithm that predicts the protein …
AlphaFold is a neural-network-based tool for the prediction of 3D structures of protein. In CASP14, a blind structure prediction challenge, it performed significantly better than other competitors, which makes it the best available structure prediction tool. One of the outputs of AlphaFold is the probability profile of residue-residue distances. This makes it possible to score any conformation of the studied protein to express its compliance with the AlphaFold model. Here we show how this score can be used to drive protein folding simulation by metadynamics and parallel tempering metadynamics.…
OpenProteinSet: Training data for structural biology at scale
Harvard University（哈佛大学）；Harvard Medical School（哈佛医学院）；Columbia University（哥伦比亚大学）；Prescient Design, Genentech（普雷赛特设计公司，基因泰克）；New York University（纽约大学）；Flatiron Institute（弗拉蒂伦研究所）
AlphaFold predicts protein structures from the amino acid sequence at or near experimental resolution, solving the 50-year-old protein folding challenge, leading to progress by transforming large-scale genomics data into protein structures. AlphaFold will also greatly change the scientific research model from low-throughput to high-throughput manner. The AlphaFold framework is a mixture of two types of workloads: MSA construction based on CPUs and model inference on GPUs. The first CPU stage dominates the overall runtime, taking hours for a single protein due to the large database sizes and I/…
AF2-Mutation: Adversarial Sequence Mutations against AlphaFold2 on Protein Tertiary Structure Prediction
Shanghai AI Laboratory（上海人工智能实验室）；Research Institute of Intelligent Complex Systems, Fudan University（复旦大学智能复杂系统研究院）；Institute of Metabolism and integrative biology, Fudan University（复旦大学代谢与整合生物学研究所）；Zelixir Biotech（泽里瑟生物科技公司）；WAVES, Ghent University（根特大学WAVES机构）
In Dec 2020, the results of AlphaFold2 were presented at CASP14, sparking a revolution in the field of protein structure predictions. For the first time, a purely computational method could challenge experimental accuracy for structure prediction of single protein domains. The code of AlphaFold2 was released in the summer of 2021, and since then, it has been shown that it can be used to accurately predict the structure of most (ordered) proteins and many protein-protein interactions. It has also sparked an explosion of development in the field, improving AI-based methods to predict protein com…
Accurate protein structure prediction can significantly accelerate the development of life science. The accuracy of AlphaFold2, a frontier end-to-end structure prediction system, is already close to that of the experimental determination techniques. Due to the complex model architecture and large memory consumption, it requires lots of computational resources and time to implement the training and inference of AlphaFold2 from scratch. The cost of running the original AlphaFold2 is expensive for most individuals and institutions. Therefore, reducing this cost could accelerate the development of…
Generating Highly Designable Proteins with Geometric Algebra Flow Matching
Heidelberg Institute for Theoretical Studies（海德堡理论研究所）；Heidelberg University（海德堡大学）；Max Planck Institute for Polymer Research（马克斯·普朗克聚合物研究所）；Stockholm University（斯德哥尔摩大学）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）
The computer artificial intelligence system AlphaFold has recently predicted previously unknown three-dimensional structures of thousands of proteins. Focusing on the subset with high-confidence scores, we algorithmically analyze these predictions for cases where the protein backbone exhibits rare topological complexity, i.e. knotting. Amongst others, we discovered a $7_1$-knot, the most topologically complex knot ever found in a protein, as well several 6-crossing composite knots comprised of two methyltransferase or carbonic anhydrase domains, each containing a simple trefoil knot. These dee…
DiffPack: A Torsional Diffusion Model for Autoregressive Protein Side-Chain Packing
Mila - Québec AI Institute（米拉-魁北克人工智能研究所）；Université de Montréal（蒙特利尔大学）；Intel（英特尔公司）；HEC Montréal（蒙特利尔高等商学院）；CIFAR AI Research Chair（加拿大高级研究所人工智能研究主席）
The Chinese University of Hong Kong (CUHK)（香港中文大学）；King Abdullah University of Science and Technology (KAUST)（阿卜杜拉国王科技大学）；The CUHK Shenzhen Research Institute（香港中文大学深圳研究院）
Protein-RNA interactions are of vital importance to a variety of cellular activities. Both experimental and computational techniques have been developed to study the interactions. Due to the limitation of the previous database, especially the lack of protein structure data, most of the existing computational methods rely heavily on the sequence data, with only a small portion of the methods utilizing the structural information. Recently, AlphaFold has revolutionized the entire protein and biology field. Foreseeably, the protein-RNA interaction prediction will also be promoted significantly in …
Out-of-equilibrium selection pressure enhances inference from protein sequence data
Institute of Bioengineering, School of Life Sciences, École Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院生命科学学院生物工程学院）；SIB Swiss Institute of Bioinformatics（瑞士生物信息学研究所）
Deep learning has contributed to major advances in the prediction of protein structure from sequence, a fundamental problem in structural bioinformatics. With predictions now approaching the accuracy of crystallographic resolution in some cases, and with accelerators like GPUs and TPUs making inference using large models rapid, fast genome-level structure prediction becomes an obvious aim. Leadership-class computing resources can be used to perform genome-scale protein structure prediction using state-of-the-art deep learning models, providing a wealth of new data for systems biology applicati…
Yusuf Hamied Department of Chemistry University of Cambridge（尤瑟夫·哈米德化学系剑桥大学）；Department of Computer Science and Technology University of Cambridge（计算机科学与技术系剑桥大学）；Institute for Drug Discovery Leipzig University（药物发现研究所莱比锡大学）；Department of Chemistry Ludwig-Maximilans-University Munich, Germany（化学系路易斯·马克西米利安大学慕尼黑，德国）
The University of Hong Kong（香港大学）；Peking University（北京大学）；Microsoft Research Asia（微软亚洲研究院）；University of Waterloo（滑铁卢大学）
Protein language models have excelled in a variety of tasks, ranging from structure prediction to protein engineering. However, proteins are highly diverse in functions and structures, and current state-of-the-art models including the latest version of AlphaFold rely on Multiple Sequence Alignments (MSA) to feed in the evolutionary knowledge. Despite their success, heavy computational overheads, as well as the de novo and orphan proteins remain great challenges in protein representation learning. In this work, we show that MSAaugmented models inherently belong to retrievalaugmented methods. Mo…
University of Texas at San Antonio（得克萨斯大学圣安东尼奥分校）；Argonne National Laboratory（阿贡国家实验室）；University of Central Florida（中佛罗里达大学）；Air Force Research Laboratory（空军研究实验室）；SRI International（斯坦福国际研究院）
Deep neural networks such as AlphaFold and RoseTTAFold predict remarkably accurate structures of proteins compared to other algorithmic approaches. It is known that biologically small perturbations in the protein sequence do not lead to drastic changes in the protein structure. In this paper, we demonstrate that RoseTTAFold does not exhibit such a robustness despite its high accuracy, and biologically small perturbations for some input sequences result in radically different predicted protein structures. This raises the challenge of detecting when these predicted protein structures cannot be t…
Insilico Medicine Shanghai Ltd（英矽智能上海有限公司）；Insilico Medicine Kong Kong Ltd（英矽智能香港有限公司）
The AlphaFold computer program predicted protein structures for the whole human genome, which has been considered as a remarkable breakthrough both in artificial intelligence (AI) application and structural biology. Despite the varying confidence level, these predicted structures still could significantly contribute to structure-based drug design of novel targets, especially the ones with no or limited structural information. In this work, we successfully applied AlphaFold in our end-to-end AI-powered drug discovery engines constituted of a biocomputational platform PandaOmics and a generative…
West China Hospital, Sichuan University（四川大学华西医院）；University of Gothenburg（哥德堡大学）；The Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital)（中国科学院大学附属肿瘤医院（浙江省肿瘤医院））；Institute of Basic Medicine and Cancer (IBMC), Chinese Academy of Sciences（中国科学院基础医学与肿瘤研究所（IBMC））
De novo molecular design has facilitated the exploration of large chemical space to accelerate drug discovery. Structure-based de novo method can overcome the data scarcity of active ligands by incorporating drug-target interaction into deep generative architectures. However, these strategies are bottlenecked by the small fraction of experimentally determined protein or complex structures. In addition, the cost of molecular generation is computationally expensive due to 3D representations of both molecule and protein. Here, we demonstrate a widely used and fast protein sequence-based reinforce…
