VIZR--an automated chemometric technique for metabolic profiling

Area of Science:

• Metabolomics
• Chemometrics
• Biostatistics

Background:

• Metabolic profiling requires robust methods for distinguishing subtle biological variations.
• Existing techniques may be sensitive to sample preparation and dilution effects.

Purpose of the Study:

• To introduce and validate a novel chemometric technique, visual interpretation of z-score ratios (VIZR), for metabolic difference identification.
• To demonstrate VIZR's capability in analyzing (1)H NMR data from urine samples.

Main Methods:

• Development of VIZR using the R programming language.
• Analysis of 49 urine samples ((1)H NMR spectroscopy): 41 controls and 8 test samples (dietary changes, dilution).
• Application of VIZR to discriminate test samples from controls and identify discriminatory spectral regions.

Main Results:

• VIZR successfully discriminated all 8 test samples from the 41 control samples.
• The method identified specific (1)H NMR spectral regions contributing to sample disparities.
• VIZR demonstrated robustness against dilution effects, crucial for urine analysis.

Conclusions:

• VIZR is a powerful, self-normalizing chemometric tool for identifying metabolic differences in biosamples.
• The technique is applicable to (1)H NMR data and potentially other digitized metabolic datasets.
• Potential applications include toxicological profiling, disease diagnosis, and biomarker discovery.