Data Processing Optimization in Untargeted Metabolomics of Urine Using Voigt Lineshape Model Non-Linear Regression...

Area of Science:

• Analytical Chemistry
• Biochemistry
• Spectroscopy

Background:

• Nuclear magnetic resonance (NMR) spectroscopy is a powerful tool for untargeted metabolomics.
• Urine NMR spectra present challenges due to signal overlap, peak shifting from pH/salt variations, and baseline noise.
• Existing data processing methods can be time-consuming or require spectral libraries.

Purpose of the Study:

• To develop a straightforward data processing tool for analyzing complex urine metabolomics datasets.
• To address signal overlap and peak shifting issues in NMR spectra.
• To enable rapid untargeted metabolomics analysis without 2D-spectra or spectral libraries.

Main Methods:

• A non-linear curve fitting model utilizing Voigt function line shape was applied.
• Integration of underlying peak areas was performed.
• The method was validated using spiking experiments and tested on human urine 1H datasets.

Main Results:

• The developed tool effectively addresses signal overlap and peak shifting in urine NMR spectra.
• It allows for rapid untargeted analysis of metabolomics data.
• Performance was comparable to conventionally used methods.

Conclusions:

• The novel data processing tool facilitates more efficient and accurate analysis of urine metabolomics data.
• This approach simplifies the analysis of complex NMR spectra.
• It offers a valuable alternative to existing time-consuming methods.