Getting more from low-resolution methods: the combination of SAXS and atomistic molecular simulation

蛋白质的生物学功能不仅取决于原子的空间排列，还取决于其构象景观的动力学。高分辨率方法（x射线晶体学和低温电镜）的分子结构测定揭示了原子细节，但往往隐藏了对功能至关重要的动力学。另一方面，小角度x射线散射（SAXS）提供了溶液状态下的结构信息，但在分辨率方面受到限制。本研究将开发原子分子模拟与SAXS数据相结合的方法和工具，将SAXS实验的结构信息扩展到更高分辨率。使用SAXS对蛋白质结构进行从头算建模无法提供具有完全理解生物功能所需的原子级分辨率的结构。这个建议解决了这个问题。影响SAXS常规MD模拟的一个重要问题与运行计算的技术困难有关。每一步都需要使用相关软件的经验。然而，自动化是可能的。我们的Protocaller软件自动化了蛋白质MD模拟的工作流程，用于计算配体结合自由能。该软件已被纳入银河生物信息学平台，提高了分子模拟计算方法的可访问性、可共享性和可重复性。在这里，我们将扩展该软件，使先进的MD模拟自动应用于SAXS数据的结构分析。

A protein's biological function is not only determined by the spatial arrangement of atoms but also by the dynamics of its conformational landscape. Molecular structure determination from high-resolution methods (X-ray crystallography and cryoEM) reveal atomistic details but often hide the dynamics critical to function. Small angle X-ray scattering (SAXS), on the other hand, provides structural information in the solution state, but is limited in terms of resolution. In this proposal, methods and tools will be developed to combine atomistic molecular simulations with SAXS data, to extend the structural information from SAXS experiments to higher resolution. Ab initio modelling of protein structure using SAXS is unable to deliver structures with the atomistic-level resolution required to fully understand biological function. This proposal solves this problem. A significant problem affecting routine MD simulation with SAXS relates to the technical difficulties running the calculations. Each step requires experience in the relevant software. However, automation is possible. Our Protocaller software automates the workflow for protein MD simulations for calculating ligand binding free energies. This software has been incorporated into the Galaxy bioinformatics platform, improving the accessibility, shareability, and reproducibility of computational methods for molecular simulations. Here, we will extend this software to automate the application of advanced MD simulations to the structural analysis of SAXS data.