Predicting the docking between proteins and ligands is a crucial and challenging task for drug discovery. However, traditional docking methods mainly rely on scoring functions, and deep learning-based docking approaches usually neglect the 3D spatial information of proteins and ligands, as well as the graph-level features of ligands, which limits their performance. To address these limitations, we propose an equivariant transformer neural network for protein-ligand docking pose prediction. Our approach involves the fusion of ligand graph-level features by feature processing, followed by the learning of ligand and protein representations using our proposed TAMformer module. Additionally, we employ an iterative optimization approach based on the predicted distance matrix to generate refined ligand poses. The experimental results on real datasets show that our model can achieve state-of-the-art performance.
Docking is an important tool in computational drug discovery that aims to predict the binding pose of a ligand to a target protein through a combination of pose scoring and optimization. A scoring function that is differentiable with respect to atom positions can be used for both scoring and gradient-based optimization of poses for docking. Using a differentiable grid-based atomic representation as input, we demonstrate that a scoring function learned by training a convolutional neural network (CNN) to identify binding poses can also be applied to pose optimization. We also show that an iterat…
Hong Kong Baptist University（香港浸会大学）；Shanghai AI Laboratory（上海人工智能实验室）；Huazhong University of Science and Technology（华中科技大学）；Shanghai Jiao Tong University（上海交通大学）；Microsoft Research AI4Science（微软研究院AI科学部）
An accurate nucleic acid-small molecule docking framework via geometric deep learning with large-scale pretraining
College of Pharmaceutical Sciences, Zhejiang University（浙江大学药学院）；Faculty of Health Sciences, University of Macau（澳门大学健康科学学院）；Zhejiang Provincial Key Laboratory for Intelligent Drug Discovery and Development, Jinhua Institute of Zhejiang University（浙江省智能药物发现与开发重点实验室，浙江大学金华研究院）；Shanghai Innovation Institute（上海创新研究院）
Beijing Key Laboratory of Big Data Management and Analysis Methods（大数据管理与分析方法北京市重点实验室）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；School of Computer Science and Technology, Huazhong University of Science and Technology（华中科技大学计算机科学与技术学院）；Microsoft Research AI4Science（微软研究院AI4Science）；School of Information Science and Technology, University of Science and Technology of China（中国科学技术大学信息科学技术学院）
KAIST（韩国科学技术院）；HITS incorporation（HITS 公司）；KI for Artificial Intelligence, KAIST（韩国科学技术院人工智能研究院）
Recently, deep neural network (DNN)-based drug-target interaction (DTI) models were highlighted for their high accuracy with affordable computational costs. Yet, the models' insufficient generalization remains a challenging problem in the practice of in-silico drug discovery. We propose two key strategies to enhance generalization in the DTI model. The first is to predict the atom-atom pairwise interactions via physics-informed equations parameterized with neural networks and provides the total binding affinity of a protein-ligand complex as their sum. We further improved the model generalizat…
Most widely used ligand docking methods assume a rigid protein structure. This leads to problems when the structure of the target protein deforms upon ligand binding. In particular, the ligand's true binding pose is often scored very unfavorably due to apparent clashes between ligand and protein atoms, which lead to extremely high values of the calculated van der Waals energy term. Traditionally, this problem has been addressed by explicitly searching for receptor conformations to account for the flexibility of the receptor in ligand binding. Here we present a deep learning model trained to ta…
College of Chemistry and Molecular Engineering, Biomedical Pioneering Innovation Center, Peking University（北京大学化学与分子工程学院，生物医学前沿创新中心）；Huawei（华为）；Beijing Changping Lab（北京昌平实验室）
Virtual screening, including molecular docking, plays an essential role in drug discovery. Many traditional and machine-learning based methods are available to fulfil the docking task. The traditional docking methods are normally extensively time-consuming, and their performance in blind docking remains to be improved. Although the runtime of docking based on machine learning is significantly decreased, their accuracy is still limited. In this study, we take the advantage of both traditional and machine-learning based methods, and present a method Deep Site and Docking Pose (DSDP) to improve t…
Argonne National Laboratory（阿贡国家实验室）；Brookhaven National Laboratory（布鲁克海文国家实验室）
We propose a benchmark to study surrogate model accuracy for protein-ligand docking. We share a dataset consisting of 200 million 3D complex structures and 2D structure scores across a consistent set of 13 million "in-stock" molecules over 15 receptors, or binding sites, across the SARS-CoV-2 proteome. Our work shows surrogate docking models have six orders of magnitude more throughput than standard docking protocols on the same supercomputer node types. We demonstrate the power of high-speed surrogate models by running each target against 1 billion molecules in under a day (50k predictions pe…
Molecular docking, given a ligand molecule and a ligand binding site (called ``pocket'') on a protein, predicting the binding mode of the protein-ligand complex, is a widely used technique in drug design. Many deep learning models have been developed for molecular docking, while most existing deep learning models perform docking on the whole protein, rather than on a given pocket as the traditional molecular docking approaches, which does not match common needs. What's more, they claim to perform better than traditional molecular docking, but the approach of comparison is not fair, since tradi…
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Tsinghua University（清华大学）；University of Chinese Academy of Sciences（中国科学院大学）；Peking University（北京大学）；University of Washington（华盛顿大学）；Georgia Institute of Technology（佐治亚理工学院）；Helixon Research（Helixon研究院）
Delta Score: Improving the Binding Assessment of Structure-Based Drug Design Methods
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）；Peking University（北京大学）
The field of machine learning for drug discovery is witnessing an explosion of novel methods. These methods are often benchmarked on simple physicochemical properties such as solubility or general druglikeness, which can be readily computed. However, these properties are poor representatives of objective functions in drug design, mainly because they do not depend on the candidate's interaction with the target. By contrast, molecular docking is a widely successful method in drug discovery to estimate binding affinities. However, docking simulations require a significant amount of domain knowled…
Broad Institute of MIT and Harvard（麻省理工学院和哈佛大学博德研究所）；Massachusetts Institute of Technology（麻省理工学院）
Computational antibody design seeks to automatically create an antibody that binds to an antigen. The binding affinity is governed by the 3D binding interface where antibody residues (paratope) closely interact with antigen residues (epitope). Thus, predicting 3D paratope-epitope complex (docking) is the key to finding the best paratope. In this paper, we propose a new model called Hierarchical Equivariant Refinement Network (HERN) for paratope docking and design. During docking, HERN employs a hierarchical message passing network to predict atomic forces and use them to refine a binding compl…
We present a simple, modular graph-based convolutional neural network that takes structural information from protein-ligand complexes as input to generate models for activity and binding mode prediction. Complex structures are generated by a standard docking procedure and fed into a dual-graph architecture that includes separate sub-networks for the ligand bonded topology and the ligand-protein contact map. This network division allows contributions from ligand identity to be distinguished from effects of protein-ligand interactions on classification. We show, in agreement with recent literatu…
Here we give a detailed description of MOLS 2.0.The source code and binary of MOLS 2.0 are distributed free (under GNU Lesser General Public License) for the scientific community. They are freely available for download at https://sourceforge.net/projects/mols2-0/files/.
Contrastive Multi-Task Learning with Solvent-Aware Augmentation for Drug Discovery
Department of Health Technology and Informatics, The Hong Kong Polytechnic University（健康科技与信息学系，香港理工大学）；Department of Nuclear Medicine and PET, Hong Kong Sanatorium and Hospital（核医学与PET部，香港疗养院及医院）；School of Data Science, Lingnan University（数据科学学院，岭南大学）
HIV integrase is a 32 kDa protein produced from the C-terminal portion of the Pol gene product, and is an attractive target for new anti-HIV drugs. Integrase is an enzyme produced by a retrovirus (such as HIV) that enables its genetic material to be integrated into the DNA of the infected cell. Raltegravir and Elvitegravir are two important drugs against integrase.
A rigorous formalism for estimating noncovalent binding free energies and thermodynamic expectations from calculations in which receptor configurations are sampled independently from the ligand is derived. Due to this separation, receptor configurations only need to be sampled once, facilitating the use of binding free energy calculations in virtual screening. Demonstrative calculations on a host-guest system yield good agreement with previous free energy calculations and isothermal titration calorimetry measurements. Implicit ligand theory provides guidance on how to improve existing molecula…
Rigid Protein-Protein Docking via Equivariant Elliptic-Paraboloid Interface Prediction
Tsinghua University（清华大学）；Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；Beijing Key Laboratory of Big Data Management and Analysis Methods（北京大数据管理与分析方法重点实验室）
While generative models have recently become ubiquitous in many scientific areas, less attention has been paid to their evaluation. For molecular generative models, the state-of-the-art examines their output in isolation or in relation to its input. However, their biological and functional properties, such as ligand-target interaction is not being addressed. In this study, a novel biologically-inspired benchmark for the evaluation of molecular generative models is proposed. Specifically, three diverse reference datasets are designed and a set of metrics are introduced which are directly releva…
Motivation: The coronavirus disease 2019 (COVID-19) caused by a new type of coronavirus has been emerging from China and led to thousands of death globally since December 2019. Despite many groups have engaged in studying the newly emerged virus and searching for the treatment of COVID-19, the understanding of the COVID-19 target-ligand interactions represents a key chal-lenge. Herein, we introduce COVID-19 Docking Server, a web server that predicts the binding modes between COVID-19 targets and the ligands including small molecules, peptides and anti-bodies. Results: Structures of proteins in…
Northeastern University（东北大学）；Brigham and Women’s Hospital（布莱根妇女医院）；Harvard Medical School（哈佛医学院）；University of California, San Diego（加州大学圣地亚哥分校）；Central European University（中欧大学）
Identifying novel drug-target interactions (DTI) is a critical and rate limiting step in drug discovery. While deep learning models have been proposed to accelerate the identification process, we show that state-of-the-art models fail to generalize to novel (i.e., never-before-seen) structures. We first unveil the mechanisms responsible for this shortcoming, demonstrating how models rely on shortcuts that leverage the topology of the protein-ligand bipartite network, rather than learning the node features. Then, we introduce AI-Bind, a pipeline that combines network-based sampling strategies w…
Stanford University（斯坦福大学）；Stanford University School of Medicine（斯坦福大学医学院）
Empirical scoring functions based on either molecular force fields or cheminformatics descriptors are widely used, in conjunction with molecular docking, during the early stages of drug discovery to predict potency and binding affinity of a drug-like molecule to a given target. These models require expert-level knowledge of physical chemistry and biology to be encoded as hand-tuned parameters or features rather than allowing the underlying model to select features in a data-driven procedure. Here, we develop a general 3-dimensional spatial convolution operation for learning atomic-level chemic…
Simulating drug binding and unbinding is a challenge, as the rugged energy landscapes that separate bound and unbound states require extensive sampling that consumes significant computational resources. Here, we describe the use of interactive molecular dynamics in virtual reality (iMD-VR) as an accurate low-cost strategy for flexible protein-ligand docking. We outline an experimental protocol which enables expert iMD-VR users to guide ligands into and out of the binding pockets of trypsin, neuraminidase, and HIV-1 protease, and recreate their respective crystallographic protein-ligand binding…
A Physically-Informed Subgraph Isomorphism Approach to Molecular Docking Using Quantum Annealers
Politecnico di Milano（米兰理工大学）；Department of Physical and Chemical Sciences, University of L’Aquila（拉奎拉大学物理与化学科学系）；Istituto Italiano di Tecnologia（意大利理工学院）；E4 Computer Engineering（E4计算机工程公司）；CINECA；Dompé farmaceutici S.p.A.（Dompé制药股份公司）
Egypt-Japan University of Science and Technology (E-JUST)（埃及日本科技大学）；Alexandria University（亚历山大大学）
Computational docking is the core process of computer-aided drug design; it aims at predicting the best orientation and conformation of a small drug molecule when bound to a target large protein receptor. The docking quality is typically measured by a scoring function: a mathematical predictive model that produces a score representing the binding free energy and hence the stability of the resulting complex molecule. We analyze the performance of both learning techniques on the scoring power, the ranking power, docking power, and screening power using the PDBbind 2013 database. For the scoring …
A disease that caused by dengue virus (DENV) has become the major health problem of the world. Nowadays, no effective treatment is available to overcome the disease due to the level of dengue virus pathogeneses. A novel treatment method such as antiviral drug is highly necessary for coping with the dengue disease. Envelope protein is one of the non-structural proteins of DENV, which engaged in the viral fusion process. The fusion process is mediated by the conformational change in the protein structure from dimer to trimer state. The previous research showed the existing cavity on the dimer st…
Prediction of protein-ligand complexes for flexible proteins remains still a challenging problem in computational structural biology and drug design. Here we present two novel deep neural network approaches with significant improvement in efficiency and accuracy of binding mode prediction on a large and diverse set of protein systems compared to standard docking. Whereas the first graph convolutional network is used for re-ranking poses the second approach aims to generate and rank poses independent of standard docking approaches. This novel approach relies on the prediction of distance matric…
We have applied a computational strategy, based on the synergy of virtual screening, docking and molecular dynamics techniques, aimed at identifying possible lead compounds for the non-covalent inhibition of the main protease 3CL-pro of the SARS-Cov2 Coronavirus. Based on the recently resolved 6LU7 PDB structure, ligands were generated using a multimodal structure-based design and then optimally docked to the 6LU7 monomer. Docking calculations show that ligand-binding is strikingly similar in SARS-CoV and SARS-CoV2 main proteases, irrespectively of the protonation state of the catalytic CYS-HI…
Molecular docking is a central method in the computer-based screening of compound libraries as a part of the rational approach to drug design. Although the method has proved its competence in predicting binding modes correctly, its inherent complexity puts high demands on computational resources. Moreover the chemical space to be screened is prohibitively large. Therefore the application of filtering prior to docking is a promising concept. We implemented a pre-docking filter based on the tangent distance algorithm originally conceived for optical character recognition. The challenging transfe…
Predicting the binding structure of a small molecule ligand to a protein -- a task known as molecular docking -- is critical to drug design. Recent deep learning methods that treat docking as a regression problem have decreased runtime compared to traditional search-based methods but have yet to offer substantial improvements in accuracy. We instead frame molecular docking as a generative modeling problem and develop DiffDock, a diffusion generative model over the non-Euclidean manifold of ligand poses. To do so, we map this manifold to the product space of the degrees of freedom (translationa…
SkeleDock is a scaffold docking algorithm which uses the structure of a protein-ligand complex as a template to model the binding mode of a chemically similar system. This algorithm was evaluated in the D3R Grand Challenge 4 pose prediction challenge, where it achieved competitive performance. Furthermore, we show that, if crystallized fragments of the target ligand are available, SkeleDock can outperform rDock docking software at predicting the binding mode. This article also addresses the capacity of this algorithm to model macrocycles and deal with scaffold hopping. SkeleDock can be accesse…
We study the binding of plant hormone IAA on its receptor TIR1 introducing a novel computational method that we call tomographic docking and that accounts for interactions occurring along the depth of the binding pocket. Our results suggest that selectivity is related to constraints that potential ligands encounter on their way from the surface of the protein to their final position at the pocket bottom. Tomographic docking helps develop specific hypotheses about ligand binding, distinguishing binders from non-binders, and suggests that binding is a three-step mechanism, consisting of engageme…
Side chain flexibility is an important factor in ligand binding. In order to determine the extent to which side chain flexibility is involved in ligand binding, a knowledge-based approach was taken. A database composed of examples of protein structures in the presence or absence of a given ligand is used to analyze which side chains undergo side chain conformational changes. Such an analysis has determined that up to 40% of binding site do not present side chain conformational changes. A total of three residues undergoing side chain conformational changes encompass approximately 85% of the bin…
Shenzhen University（深圳大学）；Tencent AI Lab（腾讯人工智能实验室）
Binding affinity prediction of three-dimensional (3D) protein ligand complexes is critical for drug repositioning and virtual drug screening. Existing approaches transform a 3D protein-ligand complex to a two-dimensional (2D) graph, and then use graph neural networks (GNNs) to predict its binding affinity. However, the node and edge features of the 2D graph are extracted based on invariant local coordinate systems of the 3D complex. As a result, the method can not fully learn the global information of the complex, such as, the physical symmetry and the topological information of bonds. To addr…
Computational Biology Center, IBM Thomas J. Watson Research（IBM托马斯·J·沃森研究中心计算生物学中心）
Proteins in complex with small molecule ligands represent the core of structure-based drug discovery. However, three-dimensional representations are absent from most deep-learning-based generative models. We here present a graph-based generative modeling technology that encodes explicit 3D protein-ligand contacts within a relational graph architecture. The models combine a conditional variational autoencoder that allows for activity-specific molecule generation with putative contact generation that provides predictions of molecular interactions within the target binding pocket. We show that mo…
