权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Next-generation alchemical free energy methods and quantum/machine-learning models for drug discovery

用于药物发现的下一代炼金自由能方法和量子/机器学习模型

基本信息

批准号：
10736499
负责人：
Darrin M York
金额：
$ 33.69万
依托单位：
RUTGERS, THE STATE UNIV OF N.J.
依托单位国家：
美国
项目类别：
财政年份：
2015
资助国家：
美国
起止时间：
2015-08-01 至 2027-08-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10736499
关键词：
2019-nCoV Academia Acceleration Address Affinity Bathing Benchmarking Binding Binding Proteins Books Chemicals Collaborations Complex Computer software Data Data Set Drug Modelings Electronics Electrostatics Family Fluorescence Polarization Free Energy Goals Government Graph Individual Industry Ions JAK2 gene Lagrangian multiplier Libraries Ligand Binding Ligands Machine Learning Maps Measurement Methods Migration Inhibitory Factor Modeling Molecular Molecular Chaperones Pathway interactions Phase Proteins Quantum Mechanics Reproducibility Research Personnel Role SARS-CoV-2 variant Sampling Series Solvents Structure System Technology Testing Therapeutic Thermodynamics Training Universities Validation Water Work base cost deep learning density design drug discovery drug-like compound experience improved inhibitor insight intermolecular interaction machine learning model mechanical energy member molecular mechanics new technology next generation protonation quantum receptor response scaffold simulation tautomer tool validation studies

项目摘要

Next-generation alchemical free energy methods and quantum/machine-learning models for drug discovery. PI: Darrin M. York, Rutgers University, Piscataway, NJ 08854-8087 USA. Alchemical free energy (AFE) simulations are indispensable in various aspects of drug discovery by enabling the prediction of ligand binding afﬁnity and selectivity. A critical barrier to progress is the current limitation in pre- cision and accuracy of AFE simulations that restricts their predictive capability. The current proposal addresses these barriers with new AFE methods and models that will be integrated into the GPU-accelerated AMBER soft- ware suite used worldwide (over 30K users) in academia, government labs and industry. Speciﬁcally, we propose to: 1. Create advanced technology for robust high-precision AFE simulations; 2. Develop accurate quantum mechanical/deep-learning potential (QDπ) force ﬁelds for drug discovery and 3. Validate precision and accu- racy of AFE methods and QDπ model. In Aim 1, we will develop new technologies for robust and reproducible calculation of ligand-protein binding free energies of compound libraries. The methods work together to enable highly precise, converged AFE simulations across thermodynamic graph networks. In Aim 2, we will develop a highly accurate and computationally efﬁcient general quantum deep-potential interaction (QDπ) force ﬁeld model for drug discovery. The QDπ model will be formulated as a machine learning potential correction (∆-MLP) to the quantum mechanical/molecular mechanical (QM/MM) energy using fast, approximate 3rd-order density-functional tight-binding QM model and well-established AMBER MM force ﬁelds and compatible water and ions models. The ∆-MLP will leverage our recently developed range-corrected deep-learning potential (DPRc) for accurate intra- and intermolecular interactions. In Aim 3, we will conduct in depth validation studies of the AFE methods from Aim 1 and QDπ model of Aim 2 on a systematic set of benchmark systems, including macrophage migration inhibitory factor (MIF), JAK2 JH2 domain, SARS-Cov2 Mpro, and sigma 1 and 2 receptors.

新一代炼金术自由能方法和量子/机器学习模型药物发现。派：达林·M·约克，罗格斯大学，美国新泽西州皮斯卡塔韦，邮编：08854-8087。炼金术自由能(AFE)模拟在药物发现的各个方面都是不可或缺的，因为它使 ﬁ刚性和选择性的配基结合预测。取得进展的一个关键障碍是目前在预付款方面的限制 AFE模拟的精确度和准确性限制了它们的预测能力。目前的提案针对的是这些障碍与新的AFE方法和模型将被集成到GPU加速的琥珀软件中- Ware套件在全球学术界、政府实验室和工业中使用(超过3万用户)。特别是ﬁ，我们建议目标：1.为稳健的高精度AFE模拟创造先进技术；2.开发精确的量子机械/深度学习潜力(QDπ)强制ﬁ领域用于药物发现和3.验证精确度和准确性原子力学法和QDπ模型的合理性。在目标1中，我们将开发新的技术，以实现健壮和可重复使用化合物文库配体-蛋白质结合自由能的计算。这些方法协同工作以实现跨热力学图形网络的高精度、聚合AFE模拟。在目标2中，我们将制定一个高精度和高计算精度的ﬁ广义量子深势相互作用(QDπ)力ﬁ场模型用于药物研发。QDπ模型将作为机器学习势校正(∆-mlp)来表示采用快速近似三阶密度泛函的量子力学/分子力学(QM/MM)能量紧束缚QM模型和公认的琥珀MM力ﬁ场和相容的水和离子模型。这个 ∆-mlp将利用我们最近开发的距离校正深度学习潜力(Dprc)来实现准确的内部学习。以及分子间相互作用。在目标3中，我们将从以下方面对AFE方法进行深入验证研究基于包括巨噬细胞迁移在内的系统基准系统的AIM 1和AIM 2的QDπ模型抑制因子(MIF)、JAK2 JH2结构域、SARS-Cov2 MPRO以及Sigma 1和2受体。