基于OPLS-DA模型的雪茄烟叶香气差异分析OA北大核心CSTPCD

To investigate the differences in the aroma quality of cigar tobacco leaves from different varieties,the aroma composition of Dexue 1,Dexue 3,and Dexue 5 were analyzed by using gas chromatography-mass spectrometry(GC-MS),and were used to construct an orthogonal partial least squares discriminant analysis(OPLS-DA)model.The results showed that the types of aroma substances in tobacco leaves were consistent among varieties,but there were differences in the content of aroma substances.In Dexue 1,the contents of megastigmatrienone,β-cyclocitral,benzaldehyde and 2,6-nonadienal were higher,while the contents of dihydroactinidiolide,2-acetylfuran,solanone and neophytadiene were lower.The contents of furan compounds and neophytadiene were higher in Dexue 3.The contents of dihydroactinidiolide,solanone and guaiacol were higher in Dexue 5.An OPLS-DA model was constructed using 30 aroma components,which the fit index of independent variable(R2X),fit index of dependent variable(R2Y),and model prediction index(Q2)were 0.806,0.993 and 0.988,respectively.The cross validation analysis of variance(CV-ANOVA)test result showed that F was 179.76(P<0.05).Through 200 substitution verification,it was found that the model did not exhibit over fitting,which indicated that the model was highly significant.The model screened 8 different aroma components from 30 aroma components,which could be used to distinguish cigar tobacco varieties.Above results provided useful information for elucidating the aroma quality of cigar tobacco leaves from different varieties.