Visualising Analytes in Gas Chromatography by Staining and Substance Maps

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Data Visualization

Background:

• Chromatographic separations coupled with spectroscopic detectors are fundamental in analytical chemistry.
• Spectroscopic data is often not visualized in chromatograms, hindering rapid analysis and requiring extensive manual data processing.
• Existing methods lack efficient ways to visually represent structural similarities and substance classes of analytes.

Purpose of the Study:

• To develop a method for visualizing structural similarities of analytes in gas chromatography-mass spectrometry (GC-MS) data.
• To create a color-coding system that represents substance classes based on mass spectral data.
• To enable retention-time independent comparison and analysis of chromatographic data.

Main Methods:

• Mass spectra from gas chromatograms were sorted by similarity using a self-organizing map (SOM) trained on a spectral database.
• The position of each spectrum on the SOM was converted into a color using the hue, saturation, and lightness (HSL) color space.
• Generated substance maps provide a retention-time independent overview of sample composition.

Main Results:

• Color-coded gas chromatograms visually highlight the substance class of detected analytes.
• Substance maps serve as retention-time independent summaries, allowing direct comparison of different analytical setups.
• The method facilitates straightforward quantification of structurally similar compounds eluting at different times.
• Comparison of sample and reference measurements readily identifies additional compounds or missing components.

Conclusions:

• Color-coded gas chromatograms and substance maps offer significant benefits for visual inspection of chemical data.
• This visualization approach enhances the accessibility of information within spectroscopic data.
• The developed technique streamlines data analysis, comparison, and quantification in analytical chemistry.