Multivariate analyses of NP-TLC chromatographic retention data for grouping of structurally-related plant secondary metabolites

• PCA, HCA heat map DA can classify chemically related secondary metabolites based on TLC retention data.
• Use of RM values as the retention parameter is implemented as a relatively new approach.
• Approach described is simple and may assist with identification of metabolites in plant extracts.

The chromatographic behavior of 28 plant secondary metabolites belonging to four chemically similar classes (alkaloids, flavonoids, flavone glycosides and sesquiterpenes) was studied by normal-phase thin-layer chromatography (NP-TLC) under 5 different chromatographic systems commonly used in plant drug analysis with the aim to explore whether the retention properties of these metabolites can determine the chemical group they belong to. The use of RM values as the retention parameter is implemented as a relatively new approach in plant analysis. Principal component analysis (PCA), hierarchical clustering heat maps and discriminant analysis (DA), were used for statistical evaluation of the chromatographic data and extraction of similarities between chemically related compounds. The twenty eight metabolites were classified into four groups by principal component analysis. The heat map of hierarchical clustering revealed that all metabolites were clustered into four groups, except for caffeine, while linear discriminant analysis showed that 96.4% of metabolites are predicted correctly as the groupings identified by chemical class in original and cross-validated data. The main advantage of the approach described in current paper is its simplicity which can assist with preliminary identification of metabolites in complex plant extracts.