Practical Workflow for Building Local Mass Spectral Libraries for Untargeted Metabolomics

Area of Science:

• Analytical Chemistry
• Biochemistry
• Metabolomics

Background:

• Metabolite identification is challenging in untargeted metabolomics due to mass spectral feature ambiguity.
• High-confidence metabolite annotation requires experimental validation with authentic standards.
• Current methods often rely on in silico predictions or database matching, which lack sufficient reliability.

Purpose of the Study:

• To develop a practical and scalable workflow for creating a high-quality mass spectral library.
• To utilize a commercially available analytical standards kit for library construction.
• To enhance the reliability and specificity of metabolite identification in metabolomics studies.

Main Methods:

• Organized 603 metabolites into 42 mixtures based on molecular mass and logD values for optimal chromatographic separation.
• Minimized injections by strategic mixture design while maintaining spectral quality.
• Validated spectral data against online resources and in silico fragmentation predictions.

Main Results:

• Successfully detected and incorporated 471 metabolites (78%) into the spectral library.
• Demonstrated improved spectral quality and reliability compared to existing resources.
• Enabled identification of endogenous metabolites in human serum samples using the constructed library.

Conclusions:

• The developed workflow offers a scalable strategy for mass spectral library construction.
• It balances spectral quality with analytical throughput for untargeted metabolomics.
• Utilizes rational mixture design and authentic standards for reliable metabolite identification with minimized experimental effort.