Diffusion models have emerged as powerful tools for molecular generation, particularly in the context of 3D molecular structures. Inspired by non-equilibrium statistical physics, these models can generate 3D molecular structures with specific properties or requirements crucial to drug discovery. Diffusion models were particularly successful at learning 3D molecular geometries' complex probability distributions and their corresponding chemical and physical properties through forward and reverse diffusion processes. This review focuses on the technical implementation of diffusion models tailored for 3D molecular generation. It compares the performance, evaluation methods, and implementation details of various diffusion models used for molecular generation tasks. We cover strategies for atom and bond representation, architectures of reverse diffusion denoising networks, and challenges associated with generating stable 3D molecular structures. This review also explores the applications of diffusion models in $\textit{de novo}$ drug design and related areas of computational chemistry, such as structure-based drug design, including target-specific molecular generation, molecular docking, and molecular dynamics of protein-ligand complexes. We also cover conditional generation on physical properties, conformation generation, and fragment-based drug design. By summarizing the state-of-the-art diffusion models for 3D molecular generation, this review sheds light on their role in advancing drug discovery as well as their current limitations.
Pharmacophore-Conditioned Diffusion Model for Ligand-Based De Novo Drug Design
Department of Chemistry Carnegie Mellon University（化学系卡内基梅隆大学）；Department of Machine Learning Carnegie Mellon University（机器学习系卡内基梅隆大学）；Departments of Chemistry and Biomedical Engineering Carnegie Mellon University（化学与生物医学工程系卡内基梅隆大学）
Target Specific De Novo Design of Drug Candidate Molecules with Graph Transformer-based Generative Adversarial Networks
Biological Data Science Lab, Dept. of Computer Engineering, Hacettepe University（哈切塔佩大学计算机工程系生物数据科学实验室）；Dept. of Computer Engineering, Hacettepe University（哈切塔佩大学计算机工程系）；Dept. of Chemistry, Middle East Technical University（中东技术大学化学系）；Dept. of Physics, Middle East Technical University（中东技术大学物理系）；Dept. Of Artificial Intelligence Engineering, Bahcesehir University（巴切塞希尔大学人工智能工程系）；Cancer Systems Biology Lab, Graduate School of Informatics, Middle East Technical University（中东技术大学信息研究生院癌症系统生物学实验室）；Dept. of Pharmaceutical Chemistry, Faculty of Pharmacy, Gazi University（加齐大学药学系药剂化学系）；Laboratory of Molecular Modeling, Evias Pharmaceutical R&D Ltd.（Evias制药研发有限公司分子建模实验室）；Dept. of Electrical and Electronics Engineering, İskenderun Technical University（伊斯肯德伦技术大学电子与电气工程系）；Institute for Computational Biomedicine, Heidelberg University（海德堡大学计算生物医学研究所）；Dept. of Bioinformatics, Graduate School of Health Sciences, Hacettepe University（哈切塔佩大学健康科学研究生院生物信息学系）
The ultimate goal of various fields is to directly generate molecules with desired properties, such as finding water-soluble molecules in drug development and finding molecules suitable for organic light-emitting diode (OLED) or photosensitizers in the field of development of new organic materials. In this respect, this study proposes a molecular graph generative model based on the autoencoder for de novo design. The performance of molecular graph conditional variational autoencoder (MGCVAE) for generating molecules having specific desired properties is investigated by comparing it to molecula…
We propose a molecular generative model based on the conditional variational autoencoder for de novo molecular design. It is specialized to control multiple molecular properties simultaneously by imposing them on a latent space. As a proof of concept, we demonstrate that it can be used to generate drug-like molecules with five target properties. We were also able to adjust a single property without changing the others and to manipulate it beyond the range of the dataset.
OMTRA: A Multi-Task Generative Model for Structure-Based Drug Design
Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh（计算生物学系，医学院，匹兹堡大学）；Ray and Stephanie Lane Computational Biology Department, School of Computer Science, Carnegie Mellon University（Ray和Stephanie Lane计算生物学系，计算机科学学院，卡内基梅隆大学）
De Novo AI Research（De Novo AI研究院）；University of Pennsylvania（宾夕法尼亚大学）
Deep learning has significantly advanced and accelerated de novo molecular generation. Generative networks, namely Variational Autoencoders (VAEs) can not only randomly generate new molecules, but also alter molecular structures to optimize specific chemical properties which are pivotal for drug-discovery. While VAEs have been proposed and researched in the past for pharmaceutical applications, they possess deficiencies which limit their ability to both optimize properties and decode syntactically valid molecules. We present a recurrent, conditional $β$-VAE which disentangles the latent space …
UniMoMo: Unified Generative Modeling of 3D Molecules for De Novo Binder Design
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）；Renmin University of China（中国人民大学）；Bytedance（字节跳动）
Central South University（中南大学）；University of Helsinki（赫尔辛基大学）
The rational design of novel molecules with desired bioactivity is a critical but challenging task in drug discovery, especially when treating a novel target family or understudied targets. Here, we propose PGMG, a pharmacophore-guided deep learning approach for bioactivate molecule generation. Through the guidance of pharmacophore, PGMG provides a flexible strategy to generate bioactive molecules with structural diversity in various scenarios using a trained variational autoencoder. We show that PGMG can generate molecules matching given pharmacophore models while maintaining a high level of …
DecompOpt: Controllable and Decomposed Diffusion Models for Structure-based Molecular Optimization
School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Center for Research on Intelligent Perception and Computing (CRIPAC)（智能感知与计算研究中心）；State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS)（多模态人工智能系统国家重点实验室）；Institute of Automation, Chinese Academy of Sciences (CASIA)（中国科学院自动化研究所）；ByteDance Research（字节跳动研究部）；Halıcıoğlu Data Science Institute, University of California San Diego（加州大学圣迭戈分校哈利西奥格鲁数据科学学院）
Shandong Xiehe University（山东协和学院）；Graduate School of Advanced Science and Engineering, Hiroshima University（广岛大学高等科学与工程学院）；Graduate School of Informatics, Nagoya University（名古屋大学信息学研究科）
IBM Research（IBM研究院）；MIT-IBM Watson AI Lab（MIT-IBM沃森人工智能实验室）；IBM Research Europe（IBM欧洲研究院）
The novel nature of SARS-CoV-2 calls for the development of efficient de novo drug design approaches. In this study, we propose an end-to-end framework, named CogMol (Controlled Generation of Molecules), for designing new drug-like small molecules targeting novel viral proteins with high affinity and off-target selectivity. CogMol combines adaptive pre-training of a molecular SMILES Variational Autoencoder (VAE) and an efficient multi-attribute controlled sampling scheme that uses guidance from attribute predictors trained on latent features. To generate novel and optimal drug-like molecules f…
Human-level molecular optimization driven by mol-gene evolution
Hainan Institute of Zhejiang University（浙江大学海南研究院）；Institute of Marine Biology and Pharmacology, Ocean College, Zhejiang University（浙江大学海洋学院海洋生物与药理研究所）；College of Pharmaceutical Sciences and Cancer Center, Zhejiang University（浙江大学药学院与癌症中心）
Structure-based Drug Design with Equivariant Diffusion Models
École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）；University of Cambridge（剑桥大学）；Cornell University（康奈尔大学）；Prescient Design, Genentech（普雷赛特设计公司，基因泰克公司）；Chinese Academy of Mathematics and System Science（中国科学院数学与系统科学研究院）；Heart and Lung Research Institute, University of Cambridge（剑桥大学心肺研究所）；University of Rome “La Sapienza”（罗马大学）；Microsoft Research AI4Science（微软研究院AI4Science）；University of Amsterdam（阿姆斯特丹大学）；University of Oxford（牛津大学）；AITHYRA Institute（艾西拉研究所）
Auxiliary Discrminator Sequence Generative Adversarial Networks (ADSeqGAN) for Few Sample Molecule Generation
School of Pharmacy, University of Pittsburgh, Pittsburgh, Pennsylvania 15261, United States；School of Software \& Microelectronics, Peking University, Beijing, 100871, China
ETH Zurich（苏黎世联邦理工学院）；ETH Singapore SEC Ltd（苏黎世联邦理工新加坡SEC有限公司）
Structure-based drug design uses three-dimensional geometric information of macromolecules, such as proteins or nucleic acids, to identify suitable ligands. Geometric deep learning, an emerging concept of neural-network-based machine learning, has been applied to macromolecular structures. This review provides an overview of the recent applications of geometric deep learning in bioorganic and medicinal chemistry, highlighting its potential for structure-based drug discovery and design. Emphasis is placed on molecular property prediction, ligand binding site and pose prediction, and structure-b…
Automatic design with machine learning and molecular simulations has shown a remarkable ability to generate new and promising drug candidates. Current models, however, still have problems in simulation concurrency and molecular diversity. Most methods generate one molecule at a time and do not allow multiple simulators to run simultaneously. Additionally, better molecular diversity could boost the success rate in the subsequent drug discovery process. We propose a new population-based approach using grammatical evolution named ChemGE. In our method, a large population of molecules are updated …
Diffusion Models at the Drug Discovery Frontier: A Review on Generating Small Molecules versus Therapeutic Peptides
Xinjiang Key Laboratory of Biological Resources and Genetic Engineering（新疆生物资源与基因工程重点实验室）；College of Life Science and Technology, Xinjiang University（新疆大学生命科学与技术学院）；School of Mathematics and System Sciences, Xinjiang University（新疆大学数学与系统科学学院）
In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active towards a given biological target, we propose to fine-tune the model with small …
Permutation invariant multi-output Gaussian Processes for drug combination prediction in cancer
MRC Biostatistics Unit, University of Cambridge（剑桥大学医学研究委员会生物统计学单元）；Oslo Centre for Biostatistics and Epidemiology (OCBE), University of Oslo（奥斯陆大学奥斯陆生物统计学与流行病学中心）；Broad Institute of MIT & Harvard（麻省理工学院与哈佛大学博德研究所）；CRUK Cambridge Centre Ovarian Programme（英国癌症研究剑桥中心卵巢项目）；Cambridge Institute of Therapeutic Immunology and Infectious Disease (CITIID)（剑桥治疗免疫学与传染病研究所）
MolSculpt: Sculpting 3D Molecular Geometries from Chemical Syntax
AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, China（人工智能科学（AI4S）优选计划，北京大学深圳研究生院，中国）；Faculty of Materials Science, Shenzhen MSU-BIT University, Shenzhen, China（材料科学学院，深圳MSU-BIT大学，中国）
Photo-acid generators (PAGs) are compounds that release acids ($H^+$ ions) when exposed to light. These compounds are critical components of the photolithography processes that are used in the manufacture of semiconductor logic and memory chips. The exponential increase in the demand for semiconductors has highlighted the need for discovering novel photo-acid generators. While de novo molecule design using deep generative models has been widely employed for drug discovery and material design, its application to the creation of novel photo-acid generators poses several unique challenges, such a…
