Kernel PLS in nonlinear process monitoring

In recent Chemolab issue we can find an interesting article about Kernel PLS methods [1]. From cited references, I strongly recommend to read article of prof. Buydens team [2] as a first additional source.


  1. J.L. Godoy, D.A. Zumoffen, J.R. Vega, and J.L. Marchetti, "New contributions to nonlinear process monitoring through Kernel Partial Least Squares", Chemometrics and Intelligent Laboratory Systems, 2014.
  2. G. Postma, P. Krooshof, and L. Buydens, "Opening the kernel of kernel partial least squares and support vector machines", Analytica Chimica Acta, vol. 705, pp. 123-134, 2011.

Perspectives of chemometrics in metabolomics

A new "perspective" article in Journal of Chemometrics [1] can be recommended for all chemometricians interested in metabolomics - both beginners and advanced people. As some addendum it is worthy to read another earlier review [2] and trace back references.


  1. D.W. Cook, and S.C. Rutan, "Chemometrics for the analysis of chromatographic data in metabolomics investigations", Journal of Chemometrics, pp. n/a-n/a, 2014.
  2. J. Trygg, E. Holmes, and T. Lundstedt, "Chemometrics in Metabonomics", J. Proteome Res., vol. 6, pp. 469-479, 2007.

Limits of quantitation

All chemometricians interested in development of analytical parameters should read a recent article of Carlson et al. [1]. From cited references, I recommend quite old but very useful paper widening knowledge about it [2].


  1. J. Carlson, A. Wysoczanski, and E. Voigtman, "Limits of quantitation – yet another suggestion", Spectrochimica Acta Part B: Atomic Spectroscopy, 2014.
  2. A. Hubaux, and G. Vos, "Decision and detection limits for calibration curves", Anal. Chem., vol. 42, pp. 849-855, 1970.

How to make PLS-DA less magic?

The answer - to read a new article about it [1]. For beginners in discriminant analysis, it is interesting to read also [2] and [3].


  1. R.G. Brereton, and G.R. Lloyd, "Partial least squares discriminant analysis: taking the magic away", Journal of Chemometrics, pp. n/a-n/a, 2014.
  2. M. Barker, and W. Rayens, "Partial least squares for discrimination", Journal of Chemometrics, vol. 17, pp. 166-173, 2003.
  3. J.A. Westerhuis, H.C.J. Hoefsloot, S. Smit, D.J. Vis, A.K. Smilde, E.J.J. Velzen, J.P.M. Duijnhoven, and F.A. Dorsten, "Assessment of PLSDA cross validation", Metabolomics, vol. 4, pp. 81-89, 2008.

Atypical experimental design quite often needed, as not all experiments can be designed using conventional DoEs. I recommend new article in Chemolab [1] about new algorithm of generation DoEs in constrained spaces with some limitations. Someone new to DoE? Please start with these two articles [2] [3].


  1. A. Beal, M. Claeys-Bruno, and M. Sergent, "Constructing space-filling designs using an adaptive WSP algorithm for spaces with constraints", Chemometrics and Intelligent Laboratory Systems, vol. 133, pp. 84-91, 2014.
  2. R. Leardi, "Experimental design in chemistry: A tutorial", Analytica Chimica Acta, vol. 652, pp. 161-172, 2009.
  3. P.W. Araujo, and R.G. Brereton, "Experimental design II. Optimization", TrAC Trends in Analytical Chemistry, vol. 15, pp. 63-70, 1996.

rPLS - new variable selection method introduced in newest article in Journal of Chemometrics [1]. As a companion to this reading, one can take new paper from ACA [2] - presenting comparison of such methods in context of MS data.


  1. . Rinnan, M. Andersson, C. Ridder, and S.B. Engelsen, "Recursive weighted partial least squares (rPLS): an efficient variable selection method using PLS", Journal of Chemometrics, pp. n/a-n/a, 2013.
  2. P.S. Gromski, Y. Xu, E. Correa, D.I. Ellis, M.L. Turner, and R. Goodacre, "A comparative investigation of modern feature selection and classification approaches for the analysis of mass spectrometry data", Analytica Chimica Acta, 2014.

Area of feasible solutions in curve resolution

Everyone interested in curve resolution should read a new article about the area of feasible solutions in this chemometric problem [1]. This article is second part of the earlier study [2] published in the same journal. A good starting lecture in this topic could be frequently cited paper by R. Tauler [3].


  1. M. Sawall, and K. Neymeyr, " A fast polygon inflation algorithm to compute the area of feasible solutions for three-component systems. II: Theoretical foundation, inverse polygon inflation, and FAC-PACK implementation ", Journal of Chemometrics, pp. n/a-n/a, 2014.
  2. M. Sawall, C. Kubis, D. Selent, A. Börner, and K. Neymeyr, "A fast polygon inflation algorithm to compute the area of feasible solutions for three-component systems. I: concepts and applications", Journal of Chemometrics, vol. 27, pp. 106-116, 2013.
  3. R. Tauler, "Multivariate curve resolution applied to second order data", Chemometrics and Intelligent Laboratory Systems, vol. 30, pp. 133-146, 1995.

How to maintain calibration models?

Repeating calibration measurements (in order to maintain the calibration model) uses time, money and effort. A trial to reduce this cost can be found in recent paper by Du et al. [1].


  1. H. Du, Z. Chen, M. Song, Y. Chen, and R. Yu, "Novel calibration model maintenance strategy for solving the signal instability in quantitative liquid chromatography–mass spectrometry", Journal of Chromatography A, 2014.

Online updating of NIR models proposed in new Chemolab article [1]. The authors designed an online updating of NIR model with supervised locality preserving projection, wavelength selection method and local regression.


  1. K. He, F. Qian, and W. Du, "Online updating of NIR model and its industrial application via adaptive wavelength selection and local regression strategy", Chemometrics and Intelligent Laboratory Systems, 2014.