Dpt. for Analytical Chemistry, Faculty of Chemistry, University of Belgrade
Development of efficient and reliable high-performance thin-layer chromatography (HPTLC) fingerprint method requires chemometric approach at several levels starting with application of experimental design and optimization techniques for the separation step, followed by data acquisition, signal manipulation, and finally solving classification and modeling problem. Serious lack in application of aforementioned techniques could be observed, particularly in the part of image processing. However, multivariate image analysis is crucial in the light of proper data acquisition. Due to an increasing scientific interest for the use of HPTLC in combination with multivariate analysis, as a tool for fingerprint of different natural products, it is important to indicate the procedure for image processing.
Full multivariate image processing, using Image J software, will be described. All steps of signal acquisition and data pretreatment including noise reduction, smoothing, filtering, baseline removal, and peak identification will be practically explained.
Thin-layer chromatogram is a rich source of data. Great amounts of information (variables or features) for a large number of samples (objects) require the use of statistical procedures in order to efficiently extract the maximum useful information from the data. Based on the similarity/dissimilarity analysis or correlation matrix, a number of unsupervised and supervised chemometric methods could be performed. The most commonly used methods for evaluation of chromatographic fingerprint data sets, such as Principal Component Analysis (PCA), Hierarchical Cluster Analysis (HCA), and Partial Least Square Discriminant Analysis (PLS-DA), will be explained.