Rapid determination of ethylene content in tomatoes using visible and short-wave near-infrared spectroscopy and wavelength selection

The plant hormone ethylene controls many aspects of development. In tomato, ethylene is an essential component for fruit ripening. In this paper, the study was concentrated on the visible/short-wave near-infrared (Vis/SW NIR) spectroscopy technique for the quantitative analysis of ethylene content in three varieties of tomatoes: non-transgenic tomatoes, and transgenic tomatoes with antisense LeETR1 and LeETR2. The results indicate that the determination of ethylene content in tomatoes could be successfully performed through Vis/SW NIR spectroscopy combined with chemometrics methods including partial least square regression (PLSR) and stepwise multiple linear regression (SMLR). The performances of models using four spectral pretreatment methods - standard normal variate (SNV), multiplicative scatter correction (MSC), first and second derivation - were compared. Wavelengths for ethylene analysis in tomatoes were proposed using SMLR models. The models of ethylene in tomato using the assigned wavelengths show reliable results. The prediction precision of PLSR and SMLR using selected wavelengths was compared. The results show that the modeling of PLSR using the visible region needed less time than that using full region, although the prediction precisions were a little lower. The overall results indicate that Vis/SW NIR spectroscopy combined with wavelength selection could be applied as a nondestructive and rapid tool for the determination of ethylene content in tomatoes. (c) 2009 Elsevier B.V. All rights reserved.