Convergence Accelerator Phase I (RAISE): MPrint-OKN

NSF融合加速器支持以团队为基础的多学科努力，以应对国家重要性的挑战，并在不久的将来展示可交付成果的潜力。 这个融合加速器第一阶段项目的更广泛的影响和潜在的社会效益是创建一个知识网络和设计平台（MPrint-OKN），用于加速，高度针对性地创建从药物到智能材料的先进产品。MPrint-OKN计划创建预测工具，以吸引渴望将其数据贡献给集体的用户群，以便获得强大的工具，帮助他们最大限度地发挥数据的价值。这种互惠互利的交流将导致越来越多的至关重要的科学数据的收集，彻底改变设计和制造更高效，成本更低，更环保的最终产品。通过将计算机、数据和配方科学领域的学术、政府和工业研究人员与化学、材料和产品工程师融合在一起，MPrint-OKN平台将成为一个中心，其主要目标有两个：1）创建工具，使合作伙伴能够发现分子系统，提供更好的性能，更低的环境影响，和/或相对于现有技术更好的经济性，以及2）通过减少发现周期时间来加速从实验室到市场的研究转变的速度。该团队已经与国家实验室和政府机构（如阿贡、爱达荷州、洛斯阿拉莫斯和NIST）以及公司（包括陶氏、默克、斯伦贝谢和苏伊士）建立了合作伙伴关系。该项目将培养一个充满活力的、协作的、多学科的社区，并在重要的经济部门创造机器学习和数据科学的创新方法，这将有助于美国保持其在数据科学和材料科学方面的全球领导地位。 MPrint-OKN的研究目标是创建一个有价值的协作系统，开发和分发最先进的分子模型，机器学习，数据科学和数据可视化工具，可用于需要分子系统进行产品开发的许多学科。我们寻求通过将先进的工具交给更多的研究人员来降低发现和开发下一代分子应用的成本和时间。MPrint-OKN的智力价值的一个方面是通过机器学习将量子力学分子表示与实验数据交叉而得出的。这些新机器学习工具的创建将使合作伙伴能够预测分子如何在复杂系统中相互作用，并对分子结构和产品性能之间的关系进行新的洞察。作为对科学工具开发的补充，MPrint-OKN项目将帮助培育来自经济各个领域的思想领袖生态系统，这些领域依赖分子来推动创新。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响力审查标准进行评估，被认为值得支持。

The NSF Convergence Accelerator supports team-based, multidisciplinary efforts that address challenges of national importance and show potential for deliverables in the near future. The broader impact and potential societal benefit of this Convergence Accelerator Phase I project is to create a knowledge network and design platform (the MPrint-OKN) for the accelerated, highly targeted creation of advanced products ranging from medicines to smart materials. The MPrint-OKN plans to create predictive tools that will attract a user-base eager to contribute their data to the collective whole in order to gain access to powerful tools that help them maximum the value of their data. This mutually beneficial exchange will result in an ever-growing collection of critically important scientific data, revolutionizing the design and manufacture of more efficient, less costly, and environmentally friendly end-products. By converging academic, government, and industrial researchers in computer, data, and formulation science with chemical, materials, and product engineers, the MPrint-OKN platform will act as a hub with two main goals: 1) create tools enabling partners to discover molecular systems that provide improved performance, lower environmental impact, and/or better economics relative to state of the art and 2) accelerate the speed of research transition from laboratory to marketplace by decreasing the discovery cycle time. The team already has partnerships in place with national laboratories and government agencies (such as Argonne, Idaho, Los Alamos, and NIST) as well as corporations (including Dow, Merck, Schlumberger, and Suez). This project will foster a vibrant, collaborative, multi-disciplinary community, and create innovative approaches to machine learning and data science in important economic sectors which will help the U.S. maintain its global leadership in data sciences and material sciences. The research objective of MPrint-OKN is to create a valuable collaborative system that develops and distributes the most advanced molecular models, machine learning, data science, and data visualization tools available to the many disciplines requiring molecular systems for product development. We seek to reduce both the cost and time of discovering and developing next-generation molecular-based applications by placing advanced tools in the hands of more researchers. One aspect of the intellectual merit of the MPrint-OKN is derived by intersecting quantum-mechanical molecular representations with experimental data through machine learning. This creation of these new machine learning tools will enable partners to predict how molecules interact in complex systems and enable new insight into the relationships between molecular structure and product performance. Complementing the development of scientific tools, the MPrint-OKN project will help foster the ecosystem of thought leaders from all sectors of our economy that rely on molecules to drive their innovation.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.