A novel procedure for strain classification of fungal mycelium by cluster and artificial neural network analysis of Fourier transform infrared (FTIR) spectra

Abstract

Fourier transform infrared spectroscopy (FTIR) was used to discriminate important wood-destroying fungi. Mycelia of 26 fungal strains belonging to 24 different species were grown on agar plates and subjected to FTIR attenuated total reflection (ATR) measurements. To classify the FTIR spectra, cluster analysis – an unsupervised multivariate data analysis method – was compared with artificial neural network (ANN) analysis – a supervised approach. By internal validation, both methods classified 99% of the spectra correctly. External validation with independent test set spectra resulted in 95% correctly classified spectra, demonstrating the high potential of this method for fungal strain identification.