Skip to main content
. 2019 Feb 1;24(3):536. doi: 10.3390/molecules24030536

Table 3.

‘Characteristic’ VOCs emitted from crushed berries of Frontenac, Marquette, St. Croix, and La Crescent grapes grown in Iowa and South Dakota. These VOCs were indicated to be the most representative variable from hierarchical clustering analysis after the PCA (JMP Pro 12.0.1, SAS Institute Inc., Cary, NC, USA).

Sample Cluster No. of Members B Most Representative Variable C Cluster Proportion of Variation Explained D Total Proportion of Variation Explained E
IA Frontenac (0.803) A 1 7 3-Methyl-1-butanol 0.993 0.257
2 9 Cyclohexanol 0.770 0.257
4 4 Isoamyl acetate 0.745 0.110
3 4 Isovaleraldehyde 0.622 0.092
5 3 Toluene 0.774 0.086
SD Frontenac (0.627) A 2 8 Styrene 0.636 0.083
10 7 Acetaldehyde 0.546 0.063
5 6 2-Octanone 0.608 0.060
1 6 Acetone 0.602 0.059
6 5 1-Hexanol 0.645 0.053
4 5 Nonane 0.624 0.051
8 4 Ethyl hexanoate 0.732 0.048
3 4 Ethyl palmitate 0.694 0.045
7 3 Hexanoic acid 0.740 0.036
9 4 N-benzyl-2-phenethylamine 0.509 0.033
12 2 2-Methyl-1-propanol 0.953 0.031
11 3 Benzoic acid, methyl ester 0.560 0.028
14 2 Isophorone 0.569 0.019
13 2 Octanal 0.524 0.017
IA Marquette (0.863) A 1 9 Hexanal 0.925 0.347
3 5 Isoamyl acetate 0.878 0.183
2 4 Styrene 0.776 0.129
5 2 Ethanol 0.933 0.078
6 2 Benzophenone 0.813 0.068
4 2 Allyl alcohol 0.703 0.059
SD Marquette (0.654) A 7 7 Acetaldehyde 0.617 0.062
6 6 Methyl ethyl ketone 0.639 0.055
3 5 Decane 0.760 0.054
19 5 Nonanal 0.667 0.048
1 4 Styrene 0.777 0.044
5 5 Amyl acetate 0.621 0.044
4 4 (E)-2-Hexenoic acid 0.704 0.040
10 4 Cyclohexanol 0.692 0.040
9 5 Octanal 0.480 0.034
2 3 1-Pentanol 0.673 0.029
8 2 Nonane 0.966 0.028
18 3 Valeraldehyde 0.630 0.027
11 3 1-Heptanol 0.629 0.027
14 4 beta-Damascenone 0.470 0.027
12 4 Allyl alcohol 0.435 0.025
13 2 p-Cymene 0.835 0.024
16 2 Methyl disulfide 0.635 0.018
15 1 beta-Cyclocitral 1.000 0.014
17 1 Nerol acetate 1.000 0.014
IA St. Croix (0.772) A 1 9 Formic acid, octyl ester 0.832 0.150
4 8 Ethyl decanoate 0.901 0.144
2 7 Isobutyraldehyde 0.674 0.094
3 5 Aspirin methyl ester 0.813 0.081
5 3 Benzeneacetaldehyde 0.858 0.052
10 3 Ethanol 0.771 0.046
8 3 Methacrolein 0.682 0.041
12 2 Isoamyl acetate 0.841 0.034
6 2 1-Butanol 0.790 0.032
7 3 Ethyl butyrate 0.493 0.030
9 2 1-Hexanol 0.649 0.026
11 2 beta-Damascenone 0.576 0.023
13 1 Valeraldehyde 1.000 0.020
SD St. Croix (0.692) A 2 8 Acetophenone 0.685 0.081
3 6 Linalool 0.786 0.069
6 6 Benzaldehyde 0.727 0.064
7 6 Methyl salicylate 0.681 0.060
5 6 Cyclohexanol 0.662 0.058
4 6 2-Heptanone 0.627 0.055
1 5 2-Phenylethanol 0.728 0.054
10 5 1-Pentanol 0.581 0.043
8 5 Benzyl alcohol 0.557 0.041
11 3 Safrol 0.855 0.038
9 3 Benzoic acid, methyl ester 0.797 0.035
12 3 Ethyl acetate 0.622 0.027
14 2 Aspirin methyl ester 0.840 0.025
13 2 Propionaldehyde 0.797 0.023
15 2 N-Benzyl-2-phenethylamine 0.628 0.018
IA La Crescent (0.699) A 1 11 beta-Cyclocitral 0.682 0.121
2 8 beta-Pinene 0.836 0.108
3 9 Ethyl butyrate 0.710 0.103
8 4 p-Cymene 0.663 0.043
9 4 Propanoic acid 0.648 0.042
12 3 1-Hexanol 0.792 0.038
6 3 Nerol acetate 0.776 0.038
4 4 Methacrolein 0.565 0.036
5 4 Beta-damascenone 0.521 0.034
7 3 (+)-4-Carene 0.650 0.031
11 3 Valeric acid 0.641 0.031
13 3 3-Methyl-1-butanol 0.638 0.031
10 2 Acetic acid 0.845 0.027
14 1 Propyl-benzene 1.000 0.016
SD La Crescent (0.741) A 3 8 Allyl alcohol 0.845 0.086
2 8 beta-Pinene 0.837 0.085
1 7 Toluene 0.691 0.061
11 5 Isoamyl acetate 0.915 0.058
6 6 Isophorone 0.637 0.048
7 6 Ethyl butyrate 0.567 0.043
8 5 Hexanal 0.669 0.042
4 5 Benzaldehyde 0.657 0.042
13 5 Styrene 0.618 0.039
15 4 Carbon disulfide 0.771 0.039
9 3 Ethyl vinyl ketone 1.000 0.038
5 3 Camphene 0.900 0.034
18 3 Linalyl acetate 0.806 0.031
17 3 Geraniol 0.730 0.028
10 2 Furfural 0.908 0.023
12 3 Isobutyraldehyde 0.499 0.019
16 2 2-Ethyl-1-hexanol 0.500 0.013
14 1 Propyl-benzene 1.000 0.013

A The total proportion of variation explained by all the cluster components. B The number of variables in the cluster. C The cluster variable that has the largest squared correlation with its cluster component. D The cluster’s proportion of variance explained by the first principal component amount the variables in the cluster, based only on variables within the cluster. E The overall proportion of variance explained by the cluster component, using only the variables within each cluster to calculate the first principal component.