An integrated 'workbench' environment for Quantum Crystallography

在上个世纪，X射线结晶学一直是确定晶体材料分子结构的金标准分析技术：它确定了化学物质中原子的类型及其相互之间的位置。量子结晶学(QCR)根据电子的分辨率计算结构和性质。这种从原子水平提高分辨率的步骤类似于能够观察到微观有机体和能够为细胞成像之间的区别。QCR是一种令人兴奋的方法，将传统的衍射技术与量子力学计算相结合，大大改进了化学/晶体结构，并能够计算大量相关的详细性质-本质上它本身就是一门新的科学。QCR因此将成为下个世纪理解材料结构和性质的基本基础方法，并将应用于化学、物理、材料和生物科学。这项技术的许多方面已经由许多小组从广泛的学科背景(从计算物理、化学结晶学到生物系统模拟)开发出来。在过去的三十年里，这些贡献产生了大量独立的软件，其中一些现在不受支持，只有在编写它们时才能与可用的工具和计算能力保持一致。因此，尽管QCR的潜力巨大，但QCR的未来目前掌握在少数专家手中，仅在某些小组或子学科中实践。该项目将这些程序无缝地结合在一起，并将它们与更成熟的结晶学软件集成在一起，从而提供更低的使用门槛和前端工具，使得只需适度的培训就能实现直观的工作。这将使更广泛的用户群能够利用QCR的力量。一系列社区参与研讨会和讨论会议都认识到了这一点的必要性。因此，它现在被国际结晶学联盟和欧洲结晶学协会的晶体委员会强烈推荐。我们将效仿结构生物学和高性能计算社区(例如https://cci.lbl.gov/cctbx_docs/index.html和http://www.phenix-online.org/))的榜样，为不同的代码开发一个社区驱动的平台和插件框架。使用在社区内开发的已建立的标准，将有可能在当前和未来的一系列软件中支持针对许多不同用例的集成工作流。此外，该项目还将与国际结晶学委员会一道，推动扩展已建立的标准，以专门支持QCR，这也将在数据管理、出版和数据/模型再利用等其他领域产生效益。南安普顿和达勒姆的研究软件工程师将在现有框架(Cctbx)上构建和开发平台，组装已知的和新的QCR组件，并将现有的各种软件组合成一个用户友好、可扩展和可维护的工作台。该团队包括在QCR和更广泛的普通结晶学社区中具有相当地位的成员，通过这一团队将使开发人员和用户群体能够继续深入接触。因此，该平台将通过社区共识、使用和反馈来发展。我们预计，由于贡献者和用户社区的增长，该项目将具有可持续性。具体地说，我们计划遵循协作计算项目(https://www.ccp.ac.uk/)，)的方法，即一个维护和管理的系统为开发人员的参与和协调提供重点，一个社区主力工具和一个理想的培训平台。

For the last century X-ray crystallography has been the gold standard analytical technique to determine the molecular structure of a crystalline material: it resolves the types of atoms in a chemical and their positions relative to each other.Quantum Crystallography (QCr) calculates structure and properties at the resolution of the electron. This step up in resolution from the atomic level is akin to the difference between being able to observe a microscopic organism and being able to image a cell. QCr is an exciting approach combining traditional diffraction techniques with quantum mechanical calculations, resulting in much improved chemical/crystal structures and the ability to calculate a great number of associated properties in significant detail - it is in essence a new science in its own right.QCr is therefore set to become the fundamental underpinning approach to understanding materials structure and properties for the next century and will be applied across chemistry, physics, materials and biological sciences.Many facets of this technique have been developed by numerous groups from a broad background of disciplines (ranging from computational physics, through chemical crystallography to biological systems simulation). These contributions have produced a multitude of stand-alone software throughout the last three decades, some of which are now unsupported and only available in line with the tools and computing power available when they were written. Therefore, despite its enormous potential, the future of QCr is currently in the hands of a few experts and practiced only in certain groups or sub-disciplines.This project will bring these programs together seamlessly and integrate them with more established crystallographic software, thus providing a lower barrier to use and the front-end tools enabling intuitive working with only modest levels of training necessary. This will enable a much wider user base to exploit the power of QCr. The need for this has been recognised in a range of community engagement workshops and discussion meetings. Accordingly, it is now highly recommended by the crystallographic commissions of The International Union of Crystallography and The European Crystallographic Association.We will follow the example of the Structural Biology and High-Performance Computing communities (e.g. https://cci.lbl.gov/cctbx_docs/index.html and http://www.phenix-online.org/) in developing a community-driven platform with a plug-in framework for different codes. Using established standards developed within the community it will be possible to support integrated workflows for many different use cases across a range of current and future software. Furthermore, this project, in conjunction with the international crystallographic commissions, will drive forward the extension of established standards to specifically support QCr, which will also generate benefits in other areas such as data management, publishing and data/model reuse. Research Software Engineers at Southampton & Durham will build on existing frameworks (the cctbx) and develop the platform, assemble known and new QCr components and combine existing diverse software into a user-friendly, extensible and maintainable workbench. The team includes members with considerable standing in both QCr and the wider general crystallography communities and through this will enable continued, deep engagement with both developer and user groups. The platform will therefore evolve through community consensus, use and feedback. We envisage sustainability of this project as a result of the growth of a community of contributors and users. Specifically, we plan to follow the approach of a Collaborative Computing Project (https://www.ccp.ac.uk/), whereby a maintained and curated system provides the focus for the engagement and coordination of developers, a community workhorse tool and an ideal training platform.