NCI Moonshot: NCI-DOE Collaboration

NCI 登月计划：NCI 与 DOE 合作

• 批准号: 10551025
• 负责人: LEONARD FREEDMAN
• 金额: $ 500万
• 依托单位: LEIDOS BIOMEDICAL RESEARCH, INC.
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-08-31 至 2023-09-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10551025
• 关键词: Achievement; Address; Algorithms; Area; Biology; Cancer Biology; Cellular Membrane; Collaborations; Communities; Computational Science; Computer Models; Contractor; Data; Data Science; Department of Energy; Development; Diagnosis; Disease; Drug Design; Goals; Government; High Performance Computing; Infrastructure; Joints; Knowledge; Laboratories; Lead; Low Dose Radiation; Machine Learning; Malignant Neoplasms; Medicine; Methodology; Mission; Modeling; National Cancer Institute; Oncology; Outcome; Patients; Pilot Projects; Population; Positioning Attribute; Pre-Clinical Model; Precision Medicine Initiative; Radiobiology; Research; Resources; Role; Shapes; Structure; Systems Biology; Technology; Therapeutic; Treatment outcome; Vendor; anticancer research; data modeling; data repository; deep learning; design; drug discovery; educational atmosphere; experience; frontier; high end computer; in silico; individual patient; knowledge base; large scale data; meetings; next generation; novel diagnostics; open source; precision oncology; predictive modeling; public-private partnership; repository; screening; software development; tool

The Department of Energy (DOE)-National Cancer Institute (NCI) Collaboration was formed to jointly accelerate federal missions in precision oncology and computing through an alignment of needs and has been driven by three key national initiatives: the National Strategic Computing Initiative, the Precision Medicine Initiative, and the Beau Biden Cancer Moonshot. NCI has a critical need for increased computational capacity and sophisticated computational models to identify promising new treatments; deepen understanding of cancer biology; understand the impact of new diagnostics, treatments, and patient factors in cancer outcomes at the individual patient level; and to integrate pre-clinical model data for cancer research, diagnosis and treatment. DOE has a need for partnerships with user communities to broaden the functionality of next-generation high-performance computers and to advance the DOE mission in low dose radiation and systems biology for energy applications. The DOE-NCI Collaboration brings together the High Performance Computing (HPC) expertise and resources of the DOE with the NCI cancer biology and oncology knowledge base, infrastructure and data repositories to support accelerating capable exascale computing technologies and advance the frontiers of precision oncology, computational and data science, and advanced computing applied to cancer. Initial collaborations between the NCI and DOE were fueled by the National Strategic Computing Initiative (NSCI) 2015 executive order, which promotes a whole-of-government approach to bringing the unique national computing capabilities of lead agencies to transform broad deployment agency missions while meeting their own mission objectives. The DOE, a lead agency for NSCI, partnered with NCI, a broad deployment agency for NSCI, to develop exascale ready tools, algorithms, and capabilities to enhance precision medicine for cancer; this further aligns the DOE-NCI Collaboration initiated under NSCI with the Precision Medicine Initiative and the Beau Biden Cancer Moonshot. Collaborative projects already underway between the DOE-NCI include the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) pilot projects, Exascale CANcer Distributed Learning Environment (CANDLE), and the Accelerating Therapeutics for Opportunities in Medicine (ATOM) Consortium, a public-private partnership under the Cancer Moonshot aimed at changing drug discovery and design paradigms through the application of computational/in silico technologies. The challenges within each JDACS4C pilot were used to shape the priorities for CANDLE, a DOE-supported Exascale Computing Project involving multiple HPC vendors, which addresses a shared need across the pilots to develop predictive models using large-scale data. Exploiting exascale technologies and capabilities anticipated for deep and machine learning, CANDLE will deliver an open source, collaboratively developed software platform providing deep learning methodologies to the community that will be used to advance precision oncology. In addition, it will establish a new paradigm for cancer research for years to come by making effective use of the ever-growing volumes and diversity of cancer-related data to build predictive models, provide better understanding of the biology underlying disease and, ultimately, provide guidance and support decisions on anticipated outcomes of treatment for individual patients.

能源部 (DOE) - 国家癌症研究所 (NCI) 合作的成立是为了通过协调需求，共同加速联邦在精准肿瘤学和计算方面的任务，并受到三项关键国家举措的推动：国家战略计算计划、精准肿瘤学计划和精准计算计划。医学倡议和博·拜登癌症登月计划。 NCI 迫切需要提高计算能力和复杂的计算模型来识别有前途的新疗法；加深对癌症生物学的理解；了解新的诊断、治疗和患者因素对个体患者癌症结果的影响；整合癌症研究、诊断和治疗的临床前模型数据。 美国能源部需要与用户社区建立合作伙伴关系，以扩大下一代高性能计算机的功能，并推进能源部在低剂量辐射和能源应用系统生物学方面的使命。 DOE-NCI 合作将 DOE 的高性能计算 (HPC) 专业知识和资源与 NCI 癌症生物学和肿瘤学知识库、基础设施和数据存储库结合在一起，以支持加速强大的百亿亿次计算技术，并推进精准肿瘤学、计算技术的前沿发展。数据科学以及应用于癌症的先进计算。 NCI 和 DOE 之间的初步合作是由 2015 年国家战略计算计划 (NSCI) 行政命令推动的，该命令提倡采取整体政府方法，利用牵头机构独特的国家计算能力来改变广泛的部署机构任务，同时满足他们的要求。自己的使命目标。 NSCI 的牵头机构 DOE 与 NSCI 的广泛部署机构 NCI 合作，开发百亿亿级就绪工具、算法和能力，以增强癌症精准医疗；这进一步使 NSCI 下发起的 DOE-NCI 合作与精准医学计划和博·拜登癌症登月计划保持一致。 DOE-NCI 之间已经开展的合作项目包括癌症高级计算解决方案联合设计 (JDACS4C) 试点项目、Exascale CANcer 分布式学习环境 (CANDLE) 以及加速医学治疗机会 (ATOM) 联盟（一个公共-癌症登月计划下的私人合作伙伴关系旨在通过应用计算/计算机技术来改变药物发现和设计范式。 每个 JDACS4C 试点中的挑战都被用来确定 CANDLE 的优先级，CANDLE 是一个由美国能源部支持、涉及多个 HPC 供应商的百亿亿次计算项目，它满足了试点之间使用大规模数据开发预测模型的共同需求。 CANDLE 将利用深度学习和机器学习预期的百亿亿次技术和能力，提供一个开源、协作开发的软件平台，为社区提供深度学习方法，用于推进精准肿瘤学。此外，它将通过有效利用不断增长的数量和多样性的癌症相关数据来建立预测模型，更好地了解潜在疾病的生物学，并最终，为未来几年的癌症研究建立新的范式。为个体患者的预期治疗结果提供指导和支持决策。