Use of Fourier transform infrared spectroscopy and chemometric data analysis to evaluate damage and age in mushrooms (Agaricus bisporus) grown in Ireland.

The aim of this research was to investigate whether the chemical changes induced by mechanical damage and aging of mushrooms can be (a) detected in the midinfrared absorption region and (b) identified using chemometric data analysis. Mushrooms grown under controlled conditions were bruise-damaged by vibration to simulate damage during normal transportation. Damaged and nondamaged mushrooms were stored for up to 7 days postharvest. Principal component analysis of Fourier transform infrared (FTIR) spectra showed evidence that physical damage had an effect on the tissue structure and the aging process. Random forest classification models were used to predict damage in mushrooms producing models with error rates of 5.9 and 9.8% with specific wavenumbers identified as important variables for identifying damage, and partial least-squares (PLS) models were developed producing models with low levels of misclassification. Modeling postharvest age in mushrooms using random forests and PLS resulted in high error rates and misclassification; however, random forest models had the ability to correctly classify 82% of day zero samples, which may be a useful tool in discriminating between "fresh" and old mushrooms. This study highlights the usefulness of FTIR spectroscopy coupled with chemometric data analysis in particular for evaluating damage in mushrooms and with the possibility of developing a monitoring system for damaged mushrooms using the FTIR "fingerprint" region.