Chemometrics and Analytical chemistry


The main research areas of the chemometrics and analytical chemistry group at the University of Bergen are the following:

Main content

  • early diagnosis and biomarker discovery
  • metabolomics 
  • multivariate statistical process control 
  • curve resolution
  • infrared and near infrared spectroscopy
  • environmental monitoring
  • multivariate calibration and regression
  • treatment for analytical raw data
  • compound identification based on multivariate methodology
  • optimization and validation of analytical methods
  • analytical applications

Early diagnosis and biomarker discovery using mass spectral profiles has been applied to multiple sclerosis, and  tools for exploration of metabolomics data isdeveloped.

MSPC is about monitoring, understanding and controlling chemical processes in the industry. The introduction of on-line spectroscopic measurements, like NIR, is a particularly exciting field.

Curve resolution problems deal with modern, advanced instrumentation like the GC-MS and LC-DAD systems. A chromatograph is coupled with a spectrometer, thus giving a complete spectrum for each time. This instrumentation enables us to employ sophisticated methods to separate the individual components in the sample mathematically, as opposed to the traditional way. This is also possible for spectroscopic measurements alone – one can, e.g, monitor a reaction over time. Spectra of the different reactants, intermediate products and end products can be found mathematically. We have developed commercial software for this purpose. Our laboratory has excellent equipment for infrared spectroscopy.


The group has done a lot of work in the field of multivariate regression and calibration. Methods like PLS and PCR are central here. Spectroscopic measurements, like the NIR, are often a speedy alternative to slower and more expensive measurements. The spectroscopic measurements can be used to predict a variety of other features, like particle size distributions, concentrations, etc. Spectroscopic measurements are of course just an example of the area of use. We have developed professional software for this subject.