Discrimination of Aspergillus spp., Botrytis cinerea and Penicillium expansum in Grape Berries by ATR-FT-IR Spectroscopy

Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy (ATR-FT-IR) in conjunction with chemometric modelling and machine learning algorithms, was successfully applied to objectively differentiate Aspergillus carbonarius, A. niger, Botrytis cinerea or Pencillium expansum fungal mycelium and mature wine-grape berries (Vitis vinifera, cultivar Chardonnay) infected with either of these bunch rot pathogens. The differentiation of B. cinerea infected grape berries from those infected with either Aspergillus or Penicillium species shows promise as a tool for the rapid detection of the pathogen when grapes are received at the winery for processing. Support vector modelling provided superior class prediction for pathogen and control samples over other modelling techniques, while random forest models were successful in classifying samples infected with Aspergillus spp., illustrating the potential for these techniques to be applied to the assessment of bunch rot pathogens. The use of ATR-FT-IR shows potential for assessing the phytosanitary aspects of grapes