BIGDATA: IA: Collaborative Research: In Situ Data Analytics for Next Generation Molecular Dynamics Workflows

BIGDATA：IA：协作研究：下一代分子动力学工作流程的原位数据分析

• 批准号: 1740990
• 负责人: Harel Weinstein
• 金额: $ 49.71万
• 依托单位: Joan and Sanford I. Weill Medical College of Cornell University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-10-01 至 2023-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1740990&HistoricalAwards=false
• 关键词: BIGDATA; IA; Collaborative; Research; Situ

Molecular dynamics simulations studying the classical time evolution of a molecular system at atomic resolution are widely recognized in the fields of chemistry, material sciences, molecular biology and drug design; these simulations are one of the most common simulations on supercomputers. Next-generation supercomputers will have dramatically higher performance than do current systems, generating more data that needs to be analyzed (i.e., in terms of number and length of molecular dynamics trajectories). The coordination of data generation and analysis cannot rely on manual, centralized approaches as it does now. This interdisciplinary project integrates research from various areas across programs such as computer science, structural molecular biosciences, and high performance computing to transform the centralized nature of the molecular dynamics analysis into a distributed approach that is predominantly performed in situ. Specifically, this effort combines machine learning and data analytics approaches, workflow management methods, and high performance computing techniques to analyze molecular dynamics data as it is generated, save to disk only what is really needed for future analysis, and annotate molecular dynamics trajectories to drive the next steps in increasingly complex simulations' workflows. The investigators tackle the data challenge of data analysis of molecular dynamics simulations on the next-generation supercomputers by (1) creating new in situ methods to trace molecular events such as conformational changes, phase transitions, or binding events in molecular dynamics simulations at runtime by locally reducing knowledge on high-dimensional molecular organization into a set of relevant structural molecular properties; (2) designing new data representations and extend unsupervised machine learning techniques to accurately and efficiently build an explicit global organization of structural and temporal molecular properties; (3) integrating simulation and analytics into complex workflows for runtime detection of changes in structural and temporal molecular properties; and (4) developing new curriculum material, online courses, and online training material targeting data analytics. The project's harnessed knowledge of molecular structures' transformations at runtime can be used to steer simulations to more promising areas of the simulation space, identify the data that should be written to congested parallel file systems, and index generated data for retrieval and post-simulation analysis. Supported by this knowledge, molecular dynamics workflows such as replica exchange simulations, Markov state models, and the string method with swarms of trajectories can be executed ?from the outside? (i.e., without reengineering the molecular dynamics code).

在原子分辨率下研究分子系统经典时间演化的分子动力学模拟在化学、材料科学、分子生物学和药物设计等领域得到广泛认可；这些模拟是超级计算机上最常见的模拟之一。下一代超级计算机的性能将大大高于现有系统，产生更多需要分析的数据（即分子动力学轨迹的数量和长度）。数据生成和分析的协调不能像现在这样依赖于手动的、集中的方法。这个跨学科项目整合了来自不同领域的研究，如计算机科学、结构分子生物科学和高性能计算，将分子动力学分析的集中性质转变为主要在原位执行的分布式方法。具体来说，这项工作结合了机器学习和数据分析方法、工作流管理方法和高性能计算技术，在生成分子动力学数据时分析分子动力学数据，仅将未来分析真正需要的数据保存到磁盘，并注释分子动力学轨迹，以推动日益复杂的模拟工作流程的下一步。研究人员通过以下方式解决了下一代超级计算机上分子动力学模拟数据分析的数据挑战：(1)创建了新的原位方法，通过局部将高维分子组织的知识简化为一组相关的结构分子性质，来跟踪分子动力学模拟中的构象变化、相变或结合事件等分子事件；(2)设计新的数据表示和扩展无监督机器学习技术，以准确有效地建立结构和时间分子性质的明确全局组织；(3)将模拟和分析集成到复杂的工作流程中，用于运行时检测结构和时间分子性质的变化；(4)开发针对数据分析的新课程材料、在线课程和在线培训材料。该项目利用了运行时分子结构转换的知识，可用于将模拟引导到模拟空间中更有前途的领域，识别应该写入拥挤的并行文件系统的数据，并为检索和模拟后分析索引生成的数据。在这些知识的支持下，分子动力学工作流程，如复制交换模拟、马尔可夫状态模型和具有轨迹群的字符串方法可以执行。从外面？（即，无需重新设计分子动力学代码）。

项目成果

Graphic Encoding of Macromolecules for Efficient High-Throughput Analysis

• DOI: 10.1145/3233547.3233607
• 发表时间: 2018-08
• 期刊: Proceedings of the 2018 ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics
• 作者: Trilce Estrada;Jeremy Benson;Hector Carrillo-Cabada;Asghar M. Razavi;M. Cuendet;H. Weinstein;E. Deelman;M. Taufer

A Machine Learning Approach for the Discovery of Ligand-Specific Functional Mechanisms of GPCRs

• DOI: 10.3390/molecules24112097
• 发表时间: 2019-06-01
• 期刊: MOLECULES
• 影响因子: 4.6
• 作者: Plante, Ambrose;Shore, Derek M.;Weinstein, Harel
• 通讯作者: Weinstein, Harel