Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1181307 | Chemometrics and Intelligent Laboratory Systems | 2008 | 9 Pages |
Estimation of a large number of parameters to model higher order interaction terms limits the interpretability and therefore applicability of classic ANOVA models. Multiplicative models have been proposed to tackle this problem in data generated mainly by interactions. In this work a GEneralized Multiplicative ANalysis Of VAriance (GEMANOVA) method is applied to assess the bactericidal activity of novel antimicrobial agents isolated from plant extracts in different structure and oxidation forms and different concentrations on three genera of bacteria. While the applicability of ANOVA is restricted due to the complex interaction among the factors, GEMANOVA is shown to return robust and easily interpretable models which conform to the actual structure of the data. This study is the first application of GEMANOVA to model the data from the field of microbiology and the first GEMANOVA model in which more than two multi-way terms are used and interpreted.