All-atom molecular simulations offer detailed insights into macromolecular phenomena, but their substantial computational cost hinders the exploration of complex biological processes. We introduce Advanced Machine-learning Atomic Representation Omni-force-field (AMARO), a new neural network potential (NNP) that combines an O(3)-equivariant message-passing neural network architecture, TensorNet, with a coarse-graining map that excludes hydrogen atoms. AMARO demonstrates the feasibility of training coarser NNP, without prior energy terms, to run stable protein dynamics with scalability and generalization capabilities.
Coarse-graining (CG) accelerates molecular simulations of protein dynamics by simulating sets of atoms as singular beads. Backmapping is the opposite operation of bringing lost atomistic details back from the CG representation. While machine learning (ML) has produced accurate and efficient CG simulations of proteins, fast and reliable backmapping remains a challenge. Rule-based methods produce poor all-atom geometries, needing computationally costly refinement through additional simulations. Recently proposed ML approaches outperform traditional baselines but are not transferable between prot…
Variational auto-encoder frameworks have demonstrated success in reducing complex nonlinear dynamics in molecular simulation to a single non-linear embedding. In this work, we illustrate how this non-linear latent embedding can be used as a collective variable for enhanced sampling, and present a simple modification that allows us to rapidly perform sampling in multiple related systems. We first demonstrate our method is able to describe the effects of force field changes in capped alanine dipeptide after learning a model using AMBER99. We further provide a simple extension to variational dyna…
MD-LLM-2: A Transferable Language Model of Molecular Dynamics with Physical Conditioning and Explicit Path Probabilities
Yusuf Hamied Department of Chemistry, University of Cambridge（剑桥大学优素福·哈米德化学系）；Department of Biotechnology, Agricultural University of Athens（雅典农业大学生物技术系）；Archimedes Unit, Athena Research Center（阿瑞斯研究中心阿尔基米德单元）
Massive molecular simulations of drug-target proteins have been used as a tool to understand disease mechanism and develop therapeutics. This work focuses on learning a generative neural network on a structural ensemble of a drug-target protein, e.g. SARS-CoV-2 Spike protein, obtained from computationally expensive molecular simulations. Model tasks involve characterizing the distinct structural fluctuations of the protein bound to various drug molecules, as well as efficient generation of protein conformations that can serve as an complement of a molecular simulation engine. Specifically, we …
Design of a Protein Potential Energy Landscape by Parameter Optimization
我们提出了一个自动化的协议，通过优化其参数来设计蛋白质能量函数的能量景观。参数进行优化，使不仅全球最低能量构象成为天然的，但也不同于天然结构的构象具有更高的能量比那些接近天然的。我们成功地应用我们的协议的UNRES势能的参数优化，使用betanova，1fsd，36个残基的子结构域的鸡绒毛蛋白头（PDB ID 1vii），和10-55个残基的片段的葡萄球菌蛋白A（PDB ID 1bdd）的训练集。新的协议的参数优化表现出更好的性能比以前的方法，其中只有最低能量的天然的和非天然的构象之间的差异进行了调整，而不考虑各种程度的天然相似的构象。我们还对未包含在训练集中的其他蛋白质进行了折刀测试，并获得了有希望的结果。结果表明，我们使用的四种蛋白质的训练集获得的参数在一定程度上可以转移到其他蛋白质。
Endeavoring toward a transferable, predictive coarse-grained explicit-chain model for biomolecular condensates underlain by liquid-liquid phase separation (LLPS), we conducted multiple-chain simulations of the N-terminal intrinsically disordered region (IDR) of DEAD-box helicase Ddx4, as a test case, to assess the roles of electrostatic, hydrophobic, cation-$π$, and aromatic interactions in amino acid sequence-dependent LLPS. We evaluated 3 residue-residue interaction schemes with a shared electrostatic potential. Neither a common hydrophobicity scheme nor one augmented with arginine/lysine-ar…
Atomically detailed simulations of RNA folding have proven very challenging in view of the difficulties of developing realistic force fields and the intrinsic computational complexity of sampling rare conformational transitions. To tackle both these issues, we extend to RNA an enhanced path sampling method previously successfully applied to proteins. In this scheme, the information about the RNA's native structure is harnessed by a soft history-dependent biasing force, which is added to the atomistic force field, thus promoting the generation of productive folding trajectories. Here, we report…
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）
Machine learning potentials are an important tool for molecular simulation, but their development is held back by a shortage of high quality datasets to train them on. We describe the SPICE dataset, a new quantum chemistry dataset for training potentials relevant to simulating drug-like small molecules interacting with proteins. It contains over 1.1 million conformations for a diverse set of small molecules, dimers, dipeptides, and solvated amino acids. It includes 15 elements, charged and uncharged molecules, and a wide range of covalent and non-covalent interactions. It provides both forces …
