Related Experiment Video
Updated: Aug 6, 2026
Mass Spectrometry-Guided Genome Mining as a Tool to Uncover Novel Natural Products
Published on: March 12, 2020
LLM-Assisted Development of a Locally Deployable Molecular Networking Toolkit: Enabling Customizable Analysis in
Shirou Feng1,2, Zihui Yang3,4, Yujun Liu5
1Key Laboratory of Chemical Biology and Traditional Chinese Medicine Research (Ministry of Education), Key Laboratory of Phytochemistry R&D of Hunan Province, and Key Laboratory of the Assembly and Application of Organic Functional Molecules of Hunan Province, Institute of Interdisciplinary Studies, College of Chemistry and Chemical Engineering, Hunan Normal University, Changsha, China, 410081.
MN-Suite offers a flexible, local workflow for natural product analysis using mass spectrometry/mass spectrometry (MS/MS). This toolkit, developed with AI assistance, aids in identifying potential alkaloid analogues for further study.
Area of Science:
- Computational chemistry
- Natural product analysis
- Bioinformatics
Background:
- Molecular networking is crucial for natural product discovery.
- Existing tools often require server-based infrastructure, limiting local accessibility.
- Developing specialized computational tools for complex data analysis is challenging.
Purpose of the Study:
- To introduce MN-Suite, an open-source, locally deployable toolkit for natural product MS/MS analysis.
- To provide a flexible, server-independent workflow for molecular networking.
- To demonstrate the utility of LLM-assisted software engineering in creating specialized bioinformatics tools.
Main Methods:
- MN-Suite integrates six similarity algorithms and three spectral modes (MS2, neutral loss (NL), and hybrid MS2+NL).
- A customizable GUI-based framework supports local preprocessing, network construction, and visualization.
- The neutral-loss entropy-similarity (NL-ES) strategy was evaluated on an Aconitum dataset.
Main Results:
- The NL-ES strategy achieved the highest internal RCF score (0.537) in the Aconitum dataset.
- MN-Suite, using diagnostic-ion/neutral-loss filtering and a seed-neighborhood strategy, prioritized 26 putative alkaloid analogues.
- The toolkit demonstrated practical utility for configurable molecular networking.
Conclusions:
- MN-Suite provides a practical and configurable local workflow for molecular networking.
- LLM-assisted software engineering, under human oversight, is effective for developing specialized computational tools.
- The toolkit facilitates the identification of natural products, such as alkaloid analogues.
