Integrated Approach for High-Confidence Compound Identification: Combining DAFdiscovery and NMRfilter

The identification of structures in complex mixtures presents a significant challenge in natural product and metabolomics research. While the choice between Nuclear Magnetic Resonance (NMR) and Mass Spectrometry (MS) as primary analytical tools is common, the complexity of these systems often demands a multi-technique approach. This presentation introduces the combined use of DAFdiscovery and NMRfilter, offering a tentative solution for high-confidence compound identification in mixtures.
DAFdiscovery is a powerful pipeline that integrates MS, NMR, and bioactivity data within a user-friendly Jupyter Notebook environment. By leveraging Statistical Total Correlation Spectroscopy (STOCSY) and Statistical HeteroSpectroscopy (SHY) calculations, DAFdiscovery enables the integration and analysis of diverse data types. NMRfilter, on the other hand, is specifically designed to enhance compound identification in mixtures. The algorithm begins with a list of candidate compounds potentially present in the sample. It simulates the chemical shift data of these candidates and compares them with the user's experimental NMR data. Additionally, it performs a search for heteronuclear and homonuclear scalar coupling networks using HMBC and HSQC-TOCSY data, determining if the signals matched in the HSQC share a unique chemical structure.
The integration of MS and NMR data into a single dataset is promising, providing broader coverage of the chemical space and facilitating confident compound identification. By combining the strengths of DAFdiscovery and NMRfilter, researchers can overcome the limitations of individual techniques and achieve high-confidence compound identification in complex mixtures.