MCnebula: Critical Chemical Classes for the Classification and Boost Identification by Visualization for Untargeted...

Area of Science:

• Metabolomics
• Systems Biology
• Bioinformatics

Background:

• Untargeted mass spectrometry is crucial for biological analysis but suffers from lengthy data processing times, particularly in systems biology.
• Existing methods often struggle with the comprehensive analysis of complex biological samples.
• The need for efficient and intuitive tools for analyzing large-scale metabolomic data is significant.

Purpose of the Study:

• To develop and present Multiple-Chemical nebula (MCnebula), a novel framework designed to streamline liquid chromatography-mass spectrometry (LC-MS) data analysis.
• To enhance the classification and structural characterization of unknown compounds beyond spectral library limitations.
• To facilitate intuitive pathway analysis and biomarker discovery in complex biological systems.

Main Methods:

• MCnebula employs an abundance-based classes (ABC) selection algorithm for feature prioritization.
• It classifies "features" (compounds) into critical chemical classes.
• Data visualization is achieved through multi-dimensional Child-Nebulae network graphs with annotations and structural information.

Main Results:

• MCnebula successfully identified "Acyl carnitines" as biomarkers in a human serum metabolomics dataset, aligning with known references.
• The framework enabled rapid annotation and discovery of compounds in a plant-derived dataset from *E. ulmoides*.
• MCnebula demonstrated its utility in exploring classification and structural characteristics of unknown compounds.

Conclusions:

• MCnebula significantly reduces LC-MS data analysis time, making it a valuable tool for systems biology.
• The framework provides an intuitive platform for biomarker discovery and pathway analysis.
• MCnebula offers a robust solution for the classification and structural elucidation of unknown compounds in metabolomics.