Atomistic structures and dynamics of biomolecules from low-resolution scattering data

实验和模拟之间的相互作用已经发展成为生物分子研究的重要推动力，但随着近年来的发展，新的挑战也出现了。例如，对大型生物分子组装或膜蛋白结构的兴趣增加了解释低分辨率散射数据的需求，因为这种结构的晶体经常衍射不良。同样，新兴的时间分辨广角x射线散射（TR-WAXS）技术原则上可以在近原子分辨率下测量溶液中蛋白质的构象转变，但还原后的TR-WAXS光谱难以在分子水平上解释。在Emmy-Noether小组中，我们将应用分子动力学（MD）模拟和相关的计算方法来解释这种单独提供仅减少信息的x射线散射数据。特别是，我们将开发新的细化算法，旨在克服当前低分辨率细化协议的局限性。这些算法的关键本质将是使用相关分析降低搜索空间的维数，以便将采样集中在低维配置空间中，其中成功的细化更有希望。作为改进静态结构的补充，我们将把测量到的TR-WAXS光谱作为MD模拟的额外驱动力，以在原子细节上对WAXS数据进行无偏解释。我们将重点分析质子泵，如视紫红质，因此我们的目标是得出这些重要的膜蛋白的结构周期仅基于一个实验。除了关注x射线数据的解释外，我的目标是在我之前的研究基础上，在艾米-诺特资助下对膜通道的选择性进行研究。通过广泛的MD模拟，我们首先旨在确定一组膜通道的选择性过滤器和机制，包括尿素转运体、甲酸通道FocA和β -桶孔。在第二步中，我们将利用对分子滤波器的理解，从零开始合理设计具有预选选择性特性的分子滤波器。

The interplay between experiment and simulation has developed to an important driving force for biomolecular research, yet with recent developments new challenges arise. For instance, the interest in structures of large biomolecular assemblies or membrane proteins has augmented the requirement to interpret low-resolution scattering data, because crystals of such structures frequently diffract poorly. Likewise, the emerging technique of time-resolved wide-angle X-ray scattering (TR-WAXS) can in principle measure conformational transitions of proteins in solution at near-atomic resolution, yet the reduced TR-WAXS spectra are difficult to interpret at a molecular level. Within the Emmy-Noether group, we will apply molecular dynamics (MD) simulations and related computational methods to interpret such X-ray scattering data that alone provide only reduced information. In particular, we will develop new refinement algorithms, that aim to overcome the limitations of current refinement protocols at low resolution. The key essence of these algorithms will be to reduce the dimensionality of the search space using a correlation analysis, in order to concentrate the sampling within a low-dimensional configurational space, where a successful refinement is more promising. Complementary to these efforts to refine static structures, we will apply measured TR-WAXS spectra as additional driving forces into MD simulations, to yield an unbiased interpretation of the WAXS data at atomic detail. We will focus that analysis on proton pumps such as rhodopsin, and we thus aim to derive the structural cycle of these important membrane proteins based on only a single experiment. Apart from the focus on the interpretation of X-ray data, I aim to build upon my previous studies on the selectivity of membrane channels within this Emmy-Noether grant. Using extensive MD simulations, we first aim to identify the selectivity filters and mechanisms of a set of membrane channels, including a urea transporter, the formate channel FocA, and a beta-barrel pore. In a second step we will use the understanding of molecular filters, in order to rationally design a molecular filter from scratch with preselected selectivity characteristics.