A chemometrics investigation of sugarcane plant properties based on the molecular composition of epicuticular wax

This paper describes a sequential chemometrics study of the use of a phenotypic property of sugarcane, its epicuticular wax, as a potential genetic marker and predictor of desirable plant traits. Epicuticular wax collected from 122 sugarcane clones was analysed by gas chromatography (GC). Fourteen principal GC peaks were identified and confirmed with the aid of Gas Chromatograph-Mass Spectrometer (GC-MS). The 122×14 GC data matrix was submitted to exploratory principal component analysis (PCA), which showed that the GC data separated according to the performance of the sugarcane clones with respect to a particular resistance trait as classified by conventional field-based rating systems. From the PCA biplot, it was possible for the first time to associate the important wax molecular components with the resistance characteristics of various sugarcane clones. nalysis stimulated research to rank the samples according to the GC data with the use of multicriteria decision-making methods (MCDM), PROMETHEE and GAIA. This experiment indicated that the ranking based on epicuticular wax components was similar to, but arguably more refined than, that obtained by the conventional field-based ratings scale. Such studies suggested that the performance of sugarcane clones could be predicted based on the molecular measurements obtained by GC. A partial least squares (PLS) calibration was constructed for the prediction of the conventional field-based performance ratings on the basis of the GC data matrix, and acceptable prediction results were obtained. Thus, it is now possible to express the sugarcane quality parameter based on accurate molecular measurements derived directly from the surface wax–a phenotypic property of the sugarcane. Additionally, this approach should reduce the time taken for breeders to rank plants.