Computational strategies for metabolite identification in metabolomics

Area of Science:

• Metabolomics
• Computational Chemistry
• Analytical Chemistry

Background:

• Metabolomic datasets contain numerous spectral features, posing challenges for compound identification.
• Existing methods for spectral alignment and comparison are abundant, but compound identification remains a significant unmet need in metabolomics.

Purpose of the Study:

• To review emerging computational strategies for metabolite identification in biofluid mixtures.
• To discuss approaches for identifying both known and novel compounds using Nuclear Magnetic Resonance (NMR) and Mass Spectrometry (MS).

Main Methods:

• Matching spectral features of unknown compounds to curated spectral databases of reference compounds ('known unknowns').
• Employing computer-aided structure elucidation methods for identifying novel compounds ('unknown unknowns') from purified samples.

Main Results:

• Successful compound identification often relies on database matching for known metabolites.
• Identification of novel compounds requires advanced computational structure elucidation techniques.

Conclusions:

• Computational strategies are crucial for advancing metabolite identification in metabolomics.
• Future improvements in analytical technologies and computational methods are needed to address the challenges of identifying both known and novel compounds.