Integrated NMR for Complex Systems

R&D 3摘要 将核磁共振光谱中的每个共振指定给分子中的各个原子对于几乎每一个 核磁共振项目。自动化程序在分配较小的可溶蛋白质方面是成功的，但经常失败 对于较大的蛋白质和复合体，不对称低聚物，以及主要具有一个次级的体系 结构，如螺旋膜蛋白或β-折叠原纤维。对于大多数系统，分配过程- 实现了样品制备、实验选择、数据处理、信号识别和数据分析 作业--既昂贵又耗时，而且需要大量的人工劳动。此手动干预 需要专业知识才能做好，而任何步骤的糟糕执行都会使后续步骤变得更加困难。 在这里，我们建议开发算法和有指导的用户界面，使高等教育的处理自动化 量化谱，改进自动化峰值拾取，并定制核磁共振谱采集，以便分配到 在完全自动化的集成数据采集、分配和结构中最大限度地提高数据并最大限度地降低成本 确定平台。这一过程的自动化将提高核磁共振数据分析的重复性， 减少核磁共振研究的时间和成本，并使更广泛的研究社区更容易获得核磁共振。

TR&D 3 SUMMARY Assigning each resonance in an NMR spectrum to individual atoms in a molecule is essential to almost every NMR project. Automated programs are successful for assignment of smaller soluble proteins, but frequently fail for larger proteins and complexes, asymmetric oligomers, and systems with predominantly one secondary structure such as helical membrane proteins or beta-sheet fibrils. For most systems, the assignment process – sample preparation, experiment selection, data processing, signal identification and data analysis to achieve assignments – is expensive, time consuming and requires significant manual effort. This manual intervention requires expertise to be done well, and poor execution at any step makes subsequent steps more difficult. Here we propose to develop algorithms and guided user interfaces that will automate processing of higher dimensional spectra, improve automated peak picking, and tailor acquisition of NMR spectra for assignment to maximize data and minimize cost in a fully-automated, integrated data acquisition, assignment and structure determination platform. Automation of this process will improve the reproducibility of NMR data analysis, reduce the time and cost of NMR studies, and make NMR more accessible to the broad research community.