Chemometric Methods for GC-MS and LC-MS

Course Description

Chemometric Methods for untargeted GC-MS covers methods for dealing with complex data produced by mass spectrometry coupled with gas chromatography. Participants will learn how overlapping peaks can be resolved into separate peaks for each of the components and their associated single component spectra. It will be shown how chemometric methods can overcome not only severe overlapping, but also retention time variations, very low signal to noise ratio, and base-line artefacts. The course includes hands-on computer time for participants to work example problems using PLS_Toolbox.


Linear Algebra for Chemometricians, MATLAB for Chemometricians or equivalent experience.

Course Outline

  1. Self-modeling Mixture Analysis and Self-modeling Curve Resolution
    - Pure variable method
    - Pure spectrum method
  2. Analysis of multiple samples simultaneously with PARAFAC
    - the PARAFAC model
    - interpreting results
    - potential problems due to retention shifts, etc.
  3. CODA (COmponent Detection Algorithm)
    - the Durbin and Watson criterion
    - extracting high quality mass chromatograms
  4. COMPARELCMS for Extracting Differences Between Similar Samples
    - A new simpler algorithm
  5. Handling Retention Time Shifts
    - Methods for alignment, COW, DTW, etc.
    - PARAFAC2
  6. Homework and Example Problems