We introduce AceFF, a pre-trained machine learning interatomic potential (MLIP) optimized for small molecule drug discovery. While MLIPs have emerged as efficient alternatives to Density Functional Theory (DFT), generalizability across diverse chemical spaces remains difficult. AceFF addresses this by refining the TensorNet2 architecture and training it on a comprehensive dataset of drug-like compounds. This approach yields a force field that balances high-throughput inference speed with DFT-level accuracy. We introduce two models sharing the same architecture: \mbox{AceFF-2}, trained on our own dataset of small molecules and supporting the essential medicinal chemistry elements (H, B, C, N, O, F, Si, P, S, Cl, Br, I); and \mbox{AceFF-2-OMOL}, trained on a subset of OMol25 containing the AceFF-2 elements plus the common biomolecular ion elements Na and K. The models are explicitly trained to handle charged states. Validation against rigorous benchmarks, including complex torsional energy scans, molecular dynamics trajectories, batched minimizations, and tests of force and energy accuracy, demonstrates that AceFF is state-of-the-art for organic molecules in the accuracy and speed regime important for drug discovery. The AceFF-2 and AceFF-2-OMol model weights and inference code are available at https://huggingface.co/Acellera/AceFF-2.0 and https://huggingface.co/Acellera/AceFF-2-OMOL.
MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design
George Mason University(乔治梅森大学);Georgia Institute of Technology(佐治亚理工学院);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Mila - Université de Montréal(米拉 - 蒙特利尔大学)
Towards Evolutionary-based Automated Machine Learning for Small Molecule Pharmacokinetic Prediction
Baker Heart and Diabetes Institute(贝克心脏与糖尿病研究所);School of Chemistry & Molecular Biosciences, The University of Queensland(昆士兰大学化学与分子生物科学学院);The University of Melbourne(墨尔本大学)
Molecule Joint Auto-Encoding: Trajectory Pretraining with 2D and 3D Diffusion
Huawei Technologies Co., Ltd.(华为技术有限公司);Chinese Academy of Sciences(中国科学院);Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China(电子科技大学深圳高等研究院);Université de Montréal(蒙特利尔大学)
FSscore: A Machine Learning-based Synthetic Feasibility Score Leveraging Human Expertise
EPFL(洛桑联邦理工学院);Laboratory of Artificial Chemical Intelligence (LIAC)(人工化学智能实验室(LIAC));Laboratory of Protein Design and Immunoengineering (LPDI)(蛋白质设计与免疫工程实验室(LPDI))
A photonic chip-based machine learning approach for the prediction of molecular properties
Nanyang Technological University(南洋理工大学);National University of Singapore(新加坡国立大学);University of California, Merced(加州大学默塞德分校);Tongji University(同济大学);Agency for Science, Technology and Research(新加坡科技研究局);Advanced Micro Foundry(先进微铸造厂);Tencent America(腾讯美国)
Nutmeg and SPICE: Models and Data for Biomolecular Machine Learning
Stanford University(斯坦福大学);Molecular Sciences Software Institute(分子科学软件研究所);Virginia Polytechnic Institute and State University(弗吉尼亚理工学院暨州立大学);Sloan Kettering Institute(斯隆·凯特琳研究所);Memorial Sloan Kettering Cancer Center(纪念斯隆·凯特琳癌症中心)
Machine Learning for Screening Large Organic Molecules
有机半导体是廉价、可扩展和可持续的电子器件、发光二极管和光电子器件的有前途的材料。对于有机光电池来说,在广阔的化合物空间中寻找具有合适性质的化合物是一个挑战。例如,电离能应该适合太阳光的光谱,并且能级必须允许有效的电荷传输。在这里,开发了一种机器学习模型,用于快速准确地估计给定分子结构的HOMO和LUMO能量。它建立在SchNet模型(Schütt et al.(2018))的基础上,并增加了“Set 2Set”读出模块(Vinyals et al.(2016))。Set 2Set模块比求和和平均聚合具有更强的表达能力,更适合于所考虑的复杂量。大多数以前的模型都是在相当小的分子上训练和评估的。因此,第二个贡献是通过添加来自其他来源的更大分子并建立一致的训练/验证/测试分割来扩展机器学习方法的范围。作为第三个贡献,我们做了一个多任务分析,以解决来自不同理论层次的不同来源的问题。所有这三个贡献结合起来,使模型的准确性接近化学准确性。
Data-Error Scaling Laws in Machine Learning on Combinatorial Mutation-prone Sets: Proteins and Small Molecules
University of Basel(巴塞尔大学);ETH Zurich(苏黎世联邦理工学院);Swiss Nanoscience Institute(瑞士纳米科学研究所);University of Toronto(多伦多大学);Vector Institute(向量研究所);TU Berlin(柏林技术大学)
SPICE, A Dataset of Drug-like Molecules and Peptides for Training Machine Learning Potentials
Stanford University(斯坦福大学);University of California, Irvine(加州大学欧文分校);Open Molecular Software Foundation(开放分子软件基金会);Acellera Labs(Acellera实验室);University of Notre Dame(圣母大学);Newcastle University(纽卡斯尔大学);Memorial Sloan Kettering Cancer Center(纪念斯隆-凯特琳癌症中心);Virginia Polytechnic Institute and State University(弗吉尼亚理工学院暨州立大学);Universitat Pompeu Fabra(庞培法布拉大学);ICREA(加泰罗尼亚研究与高级研究院);Weill Cornell Graduate School of Medical Sciences(威尔康奈尔医学院研究生院)
机器学习潜力是一个重要的工具,分子simulation. We描述的SPICE数据集,一个新的量子化学数据集,用于训练相关的模拟药物样小分子与蛋白质相互作用的潜力,高质量的数据集的短缺,训练他们的发展受到阻碍。它包含超过110万种不同的小分子、二聚体、二肽和溶剂化氨基酸的构象。它包括15个元素,带电和不带电的分子,以及广泛的共价和非共价相互作用。它提供了在ω B 97 M-D3(BJ)/def 2-TZVPPD理论水平上计算的力和能量,以及其他有用的量,例如多极矩和键级。我们在上面训练了一组机器学习潜力,并证明它们可以在广泛的化学空间区域内实现化学准确性。它可以作为一个有价值的资源,用于创建可转移的,准备使用潜在的功能,用于分子模拟。
Machine Learning for Scent: Learning Generalizable Perceptual Representations of Small Molecules
Google Research(谷歌研究院);Arizona State University(亚利桑那州立大学);University of Toronto(多伦多大学);Vector Institute for Artificial Intelligence(向量人工智能研究所);Canadian Institute for Advanced Research(加拿大高等研究院)
Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
Fritz Haber Institute of the Max Planck Society(马克斯·普朗克学会弗里茨·哈伯研究所);Technische Universität Berlin(柏林工业大学);Citrine Informatics;University of Konstanz(康斯坦茨大学)
Revving up 13C NMR shielding predictions across chemical space: Benchmarks for atoms-in-molecules kernel machine learning with new data for 134 kilo molecules
Machine learning for predicting control landscape maps of quantum molecular dynamics: Laser-induced three-dimensional alignment of asymmetric top molecules
Elucidating atmospheric brown carbon -- Supplanting chemical intuition with exhaustive enumeration and machine learning
为了解开C12 H12 O 7异构体的结构,在大气室实验中被鉴定为姜醇的光吸收光氧化产物,我们应用了基于图形的分子生成器和机器学习工作流程。为了以无偏差的方式实现这一点,生成了C12 H12 O 7的整个化学子空间的分子图,假设异构体包含两个C6环;这导致了2.6亿个分子图和1.2亿个稳定结构。使用量子化学激发能和振子强度作为训练数据,我们使用核岭回归和模拟的UV/Vis吸收光谱预测这些量。然后在不同方法的误差范围内确定了分子引起实验光谱的概率。光谱可能与实验光谱匹配的分子根据结构特征进行聚类,得到> 500,000个分子的簇。虽然我们确定了几个特征,这些特征与导致实验光谱的高概率相关,但无法给出必要特征的明确组成。因此,吸收光谱不足以唯一鉴别一种特定的异构体结构。如果从实验数据中知道更多的结构特征,结构的数量可以减少到几万个候选者。我们提供了一个程序,以检测当足够的碎片数据已被包括在内,以减少可能的分子的数量。如果在无偏差分子生成阶段已经应用结构数据,则获得获得有效候选物的最有效策略。然而,系统计数是必要的,以避免错误识别分子,同时它保证没有其他分子也符合所讨论的光谱。