Statistical Variable Selection: An Alternative Prioritization Strategy during the Nontarget Analysis of LC-HR-MS Data

Area of Science:

• Environmental Analytical Chemistry
• Mass Spectrometry
• Chemometrics

Background:

• Liquid chromatography coupled to high resolution mass spectrometry (LC-HR-MS) is crucial for analyzing small polar organic pollutants.
• LC-HR-MS generates vast datasets, necessitating data prioritization before structural elucidation.
• Current methods struggle with efficient prioritization of complex environmental data.

Purpose of the Study:

• To develop and validate a novel method for prioritizing analytes in LC-HR-MS data.
• To combine F-ratio statistical variable selection with apex detection algorithms for enhanced data analysis.
• To assess the method's performance against conventional approaches for environmental pollutant identification.

Main Methods:

• Integration of F-ratio statistical variable selection and apex detection algorithms.
• Validation using semisynthetic datasets combining real environmental data with known alkane signals.
• Performance evaluation across various false detection probabilities (0.01%–0.1%) and signal-to-noise ratios (S/N).

Main Results:

• The F-ratio method successfully prioritized features, with over 92% overlap with conventional peak list approaches (e.g., MZmine).
• This method demonstrated superior performance in sample classification and prioritization compared to pixel-by-pixel and peak list methods.
• The approach proved effective even with complex environmental matrices and varying S/N levels.

Conclusions:

• The combined F-ratio and apex detection method offers a robust solution for prioritizing analytes in LC-HR-MS data.
• This technique significantly enhances the efficiency and accuracy of identifying small polar organic pollutants in environmental samples.
• The study highlights the method's potential for routine application in environmental monitoring and research, with discussed limitations.