A molecular networking-assisted automatic database screening strategy for comprehensive annotation of small molecules...

Area of Science:

• Analytical Chemistry
• Metabolomics
• Natural Product Chemistry

Background:

• Annotating small molecules in complex matrices using LCHRMS is challenging due to data complexity and limited reference spectra.
• Existing methods struggle with comprehensive identification of known and unknown compounds.

Purpose of the Study:

• To develop and validate a novel strategy, MN/auto-DBS, for efficient and accurate small molecule annotation in complex samples.
• To improve the identification of both reported and unreported compounds.

Main Methods:

• Developed a Python software to create an in-house database (DB) for automatic screening.
• Employed MS1 exact mass screening combined with MS2 similarity analysis via feature-based molecular networking (FBMN).
• Integrated results from automatic screening and molecular networking, followed by manual curation.

Main Results:

• Successfully annotated 223 compounds in the Huangqi-Danshen (HD) herb pair.
• Identified 65 compounds not previously reported in HD, including aromatic polyketides, coumarins, and diarylheptanoids.
• Demonstrated the MN/auto-DBS strategy's effectiveness in profiling complex matrices and discovering new compounds.

Conclusions:

• The MN/auto-DBS strategy significantly enhances the annotation of small molecules in complex natural product samples.
• This approach facilitates the discovery of novel compounds and provides a comprehensive chemical profile.
• MN/auto-DBS is applicable for mining a wide range of complex matrices for potentially new chemical entities.