Methods for the exploratory analysis of two-dimensional chromatographic signals

Area of Science:

• Analytical Chemistry
• Chemometrics

Background:

• Two-dimensional chromatographic data, or fingerprints, require specialized exploratory analysis techniques.
• High-performance liquid chromatography with photodiode-array detection (HPLC-DAD) generates complex, multi-dimensional datasets.

Purpose of the Study:

• To present and illustrate the utility of diverse exploratory analysis approaches for two-dimensional chromatographic signals.
• To compare various statistical and N-way methods for their effectiveness in analyzing HPLC-DAD data.

Main Methods:

• Exploratory analysis techniques including Principal Component Analysis (PCA).
• Hierarchical clustering methods.
• N-way techniques such as PARAFAC, PARAFAC2, and Tucker3.
• Other comparative methods like the Rv coefficient and STATIS approach.

Main Results:

• Demonstration of the applicability of presented methods on experimental HPLC-DAD data.
• Evaluation of techniques for handling chromatographic data, including those with peak shifts.
• Comparison of different multivariate data analysis approaches for fingerprint analysis.

Conclusions:

• Various multivariate methods are effective for exploratory analysis of two-dimensional chromatographic data.
• The choice of method depends on the specific characteristics of the chromatographic data and the analytical goals.
• These techniques provide valuable insights into complex sample comparisons and data interpretation.