Fast, Cost-Effective and Fully Automated Structure Verification through Synergistic Use of Infrared and NMR Spectra

确定分子结构是药物发现的基础。机器人实验室、高通量实验和并行合成的发展极大地增加了分析数据量，以至于人工解释是这一过程中的瓶颈。需要对高通量分析数据进行快速、自动化的解释。当前的自动化结构验证方法在使用耗时和耗材的13C核磁共振数据时执行得最好。我们需要使用可以快速收集的不同数据，以匹配不断增长的合成产出。红外光谱学是一种灵敏的互补技术，它允许结构与光谱指纹匹配。在这个项目中，我们将结合红外和资源节约的~1H核磁共振数据，以显著提高自动化结构验证。一位有43个化合物的飞行员证明，预测准确率从40%(仅限~1H核磁共振)提高到70%(~1H结合红外)，当包括HSQC核磁共振数据时，准确率为88%。该项目将使用古德曼集团的创新分析平台(DP4和DP5)提供高通量的自动化结构验证方法。最终目标是自动化分析数据解释，包括红外、核磁共振和HSQC数据，适用于工业环境。

Determining molecular structure is fundamental to drug discovery. The development of robotic labs, high-throughput experimentation (HTE) and parallel synthesis dramatically increases the amount of analytical data to the extent that manual interpretation is a bottleneck in the process. Rapid and automated interpretation of high-throughput analytical data is needed. Current methods for automated structure verification perform best when using time- and material-consuming 13C NMR data. We need to use different data that can be collected at speed to match growing synthetic output. Infrared spectroscopy is a sensitive complementary technique which allows matching of structure to spectral fingerprints. In this project we will combine infrared and resource-economical 1H NMR data to enhance automated structure verification significantly. A pilot with 43 compounds has demonstrated an increase in correct prediction from 40% (1H NMR only) to 70% (1H combined with infrared) and 88% when HSQC NMR data is included. This project should deliver high-throughput methods for automated structure verification using the Goodman group's innovative analytical platform (DP4 & DP5). The ultimate goal is automated analytical data interpretation, including infrared, 1H NMR and HSQC data, suitable for an industrial setting.