Early Detection Of Apple Valsa Canker Based On SERS

Apple tree is one of the major cash crops in China.However,apple Valsa canker caused by the fungus(valsa mali)is a serious threat to the growth and fruit quality of fruit trees,causing serious economic losses to our plantation industry.At the initial infection stage,the fungal pathogen can survive in damaged or dead subcutaneous bast tissues for more than six months without visible symptoms;when visible symptoms appear,it is difficult to prevent the spread of the rot throughout the orchard by conventional treatments.Therefore,it is necessary to conduct early disease detection on infected trees for early disease control.The existing methods for detecting and diagnosing apple Valsa canker are mainly labor-intensive and complex biomolecular methods,and there is a lack of fast and easy detection techniques.In order to achieve early and rapid detection,this study used surface-enhanced Raman spectroscopy(SERS)as a tool,combined with chemometric methods to establish an early diagnosis model based on apple Valsa canker,and used microscopic imaging techniques to generate a map of the chemical composition of the healthy junction of the bast at the early stage of disease infestation.This study was conducted to investigate the dynamic development of the disease infestation.This study provides a proven practical application for the early detection of apple Valsa canker.The main findings and conclusions of this paper are as follows:(1)Nanosubstrates for Raman spectroscopy signal enhancement were prepared and characterized,and the test parameters were optimized.In this study,Lee-Meisel’s classical method was used to prepare silver nano-enhanced substrates for enhancing the pristine Raman signal.The silver nanoparticles were characterized by using transmission electron microscopy imaging,Raman spectroscopy,and UV-vis absorption spectroscopy,and the best formulation was finally determined to be prepared with a reduction reaction time of25 min and a trisodium citrate mass concentration of 1%.The silver nanoparticles at this time had a particle size range of 60-120 nm and showed good homogeneity and dispersion.The Raman spectra of silver nanoparticles showed no interference of spurious peaks in the substrate.Meanwhile,the most suitable conditions and acquisition parameters for SERS detection of apple Valsa canker were found,which were optimized as follows: excitation wavelength of 785 nm,exposure time of 3.57 ms,laser power of 2.6 m W,and the number of scans was 30.The results of the study laid a good foundation for the subsequent study.(2)A model for early diagnosis and analysis of apple Valsa canker based on SERS technology was constructed.Three algorithms,Multiple Spectra Baseline Correction(MSBC),Asymmetric Least Squares(As LS)and Adaptive iterative reweighted Penalized Least Squares(air-PLS),were used to pre-process the raw spectra to eliminate fluorescence baselines,and confirm that air-PLS is the most effective.Principal Component Analysis(PCA)provides a significant clustering effect to visualize the distribution of the samples.The optimal variables were selected using two strategies,algorithmic selection and manual selection.Using Back Propagation Neural Network(BP-NN),Extreme Learning Machine(ELM),Random Forest,and Least squares support vector machine(LS-SVM)algorithms combined with full-spectrum and feature bands to build discriminative models with detection accuracies of 91.80%,95.39%,99.57% and 98.04%,respectively.The microspectral imaging results showed that cellulose and lignin were significantly reduced at the disease-health junction of the bast under apple Valsa canker stress,and the results provided a spatial and temporal dynamic characterization of the changes in the content of biological components in the early stage of the disease.(3)In order to resolve the Raman spectra and spectral peak attribution,density functional theory(DFT)was applied to optimize the molecular structure and simulate theoretical spectra of apple tree bast cell wall macromolecules.Cellulose and lignin are the most important constituents of plant cell walls and the most important sources of spectral signals.The Raman characteristic peaks of cellulose and lignin were determined by simulating the vibrational attribution of the Raman spectra of the samples under the B3LYP/6-31 ++G(d,p)basis set conditions.The Raman spectra simulated by DFT theory and the detected Raman spectra have a good match.The results of the study provide a resolution and validation of the above diagnostic and analytical modeling.