Applications of chemometrics in 1D and 2D chromatography

CAST members Tijmen Bos, Wouter Knol, Leon Niezen and Stef Molenaar compiled an overview of applications and developments of chemometrics applied to one- and two-dimensional chromatography. The review was recently published in Journal of Separation Science.
Share on facebook
Share on twitter
Share on linkedin
Share on whatsapp
Share on google
Share on email

As we strive for more peak capacity to tackle the separation of the samples of tomorrow it is easy to forget that we should also still be able to retrieve the answer to our original question [1]. However, with the ever increasing complexity of our separation systems, the data we obtain from our experiments becomes similarly more sophisticated [2]. One modern example is an LC×LC-MS/MS system, which is capable of generating truly massive amounts of data per experiment.

As chromatographers, we tend to forget the data analysis and tend to rely on commercial software packages. However, as we continue to produce more efficient separation systems, the field of chemometrics is doing its best to keep up. With the technological capabilities of computer systems increasing by the day, the field of chemometrics is unsurprisingly very active.

CAST members Tijmen Bos, Wouter Knol, Leon Niezen and Stef Molenaar reviewed these recent works in their review recently published in Journal of Separation Sciences [3]. In their review, the young authors divided the work in literature into a number of categories data pre-processing (including retention-time alignment), data analysis (peak detection, information extraction, etc.), quantitative approaches and method optimization.

While the authors addressed the multivariate approaches used to tackle highly complex data, the review also focused on developments in the processing and use of day-to-day data such as obtained in 1D chromatography.

Most reported methods were developed to tackle a specific challenge in a data set and comparisons with other approaches supported by numerical data have rarely been reported.

Bos et al.

Within the subject of data preprocessing (background correction, signal smoothening, etc.) the authors found a rather large number (>10) of new methods published in the last few years. Yet, the authors surprisingly observed a complete lack of objective comparisons of these approaches. Consequently, it appears to be rather unclear which of the by now many existing approaches is suitable for the chromatographer in the routine lab. Another conclusion was that modern peak-alignment strategies are not robust for elution-order shifts.

The authors furthermore noted that this issue culminated into issues with data processing, information extraction and – ultimately – method optimization.

For the latter focal point within chemometrics, optimization strategies, the authors found that a shift in attention may be in order. While a large number of optimization strategies are reported, most studies do not take into account the validity of the actual optimization criteria. Reciting also the message from earlier works [4], the authors noted that for optimization in chromatography, more attention should be given to the quality descriptors. 

The authors devoted a lot of attention to also explain every core chemometric approach to help non-experts to understand the importance and significance of each of the featured methods. Image shows and example of an explanation of fundamental signal processing methods. Reproduced with permission from [3].

The review contains a massive table with all recent and important applications and developments within chemometrics for chromatography. The different works are sorted by category, including background correction, peak alignment, peak detection and quantification.

The review was initiated within the collaboration with Agilent Technologies through the University Relations program. The work was published open-access and is available download here.

Researchers Bos, Knol, Niezen and Molenaar are part of the UNMATCHED project, which is supported by BASF, DSM and Nouryon, and receives funding from the Netherlands Organization for Scientific Research (NWO) in the framework of the Innovation Fund for Chemistry and from the Ministry of Economic Affairs in the framework of the “PPS‐toeslagregeling”. 

References

[1] Practical Approaches for Overcoming the Challenges of Comprehensive Two-Dimensional Liquid Chromatography, B.W.J. Pirok and Peter J. Schoenmakers, LC-GC Europe, 2018, 31, 242–249, [LINK].

[2] Recent Developments in Two-Dimensional Liquid Chromatography: Fundamental Improvements for Practical Applications, B.W.J. Pirok, D.R. Stoll and P.J. Schoenmakers, Anal. Chem., 2019, 91(1), 240-263, DOI: 10.1021/acs.analchem.8b04841

[3] Recent applications of chemometrics in one- and two-dimensional chromatography, T.S. Bos, W.C. Knol, S.R.A. Molenaar, L.E. Niezen, P.J. Schoenmakers, G.W. Somsen, B.W.J. Pirok, J. Sep. Sci. 43(9-10), 2020, 1678-1727, DOI: 10.1002/jssc.202000011

[4] Optimizing separations in online comprehensive two-dimensional liquid chromatography, B.W.J. Pirok, A.F.G. Gargano and P.J. Schoenmakers, J. Sep. Sci., 2018, 41(1), 68–98, DOI: 10.1002/jssc.201700863

The Authors

Tijmen Bos

Wouter Knol

Leon Niezen

Stef Molenaar

Researchers Bos, Knol, Niezen and Molenaar are part of the UNMATCHED project, which is supported by BASF, DSM and Nouryon, and receives funding from the Netherlands Organization for Scientific Research (NWO) in the framework of the Innovation Fund for Chemistry and from the Ministry of Economic Affairs in the framework of the “PPS‐toeslagregeling”. You can read more about them and find their contact info on the Team page.

Related

Peak Tracking for Use in 2D-LC

Working with comprehensive 2D-LC (LC×LC) data can be highly challenging. This is particularly true for complex samples. Together with the group of Dwight Stoll, Stef Molenaar developed a pack-tracking algorithm to solve this.

Recent Posts

Peak Tracking for Use in 2D-LC

Working with comprehensive 2D-LC (LC×LC) data can be highly challenging. This is particularly true for complex samples. Together with the group of Dwight Stoll, Stef Molenaar developed a pack-tracking algorithm to solve this.