Chemical characteristics of loquat genotypes
In the research we found average score for taste 3.49, juiciness 4.09, soluble solide contents between 6.95–18.30%, acidity 0.22–0.82%, SSC/acidity ratio 16.39–67.68%, pH 2.85–4.94, dry matter ratio 47.29–54.73, fresh weigt 35.13-103.76 g, and dry weight 20.78–85.59 g (Table 3).
Table 3
Descriptive statistics of loquat genotypes for chemical characteristics
No | Character | Abbreviation | Unit | Min. | Max. | Mean | SD | CV (%) |
1 | Taste | Ta | code | 1.00 | 5.00 | 3.49 | 1.42 | 40.79 |
2 | Juiciness | Ju | code | 1.00 | 5.00 | 4.09 | 1.30 | 31.94 |
3 | Soluble Solid Contents | SSC | % | 6.95 | 18.30 | 12.90 | 2.11 | 16.37 |
4 | Acidity | Ac | % | 0.22 | 0.82 | 0.52 | 0.13 | 25.52 |
5 | SSC/Acidity ratio | SSCAcR | % | 16.39 | 67.68 | 27.66 | 10.70 | 38.68 |
6 | pH | - | | 2.85 | 4.94 | 3.60 | 0.43 | 11.95 |
7 | Dry Matter Ratio | DM | % | 47.29 | 54.73 | 50.02 | 1.13 | 2.26 |
8 | Fresh Weight | FreWe | g | 35.13 | 103.76 | 62.87 | 13.25 | 21.07 |
9 | Dry Weight | DW | g | 20.78 | 85.59 | 35.03 | 10.94 | 31.23 |
When the selected 70 loquat genotypes were examined in terms of taste, it was determined that 28 of them were highly sweet (40.0%), 31 of them were moderately sweet (44.2%) and 11 of them (15.8%) were less sweet (Fig. 1). It was determined that 45 of the loquat genotypes in terms of juiciness status were highly juicy (64.3%), 20 were moderately juicy (28.6%) and 5 of them (7.1%) were less juicy (Fig. 1).
Morphological characteristics of loquat genotypes
In the 70 genotype we used in the study we found leaf width between 5.41–12.71 cm, leaf length 15.63–41.22 cm, leaf length/width ratio 1.37–5.57%, petiole diameter 3.74-8.00 mm, petiole length 8.87–46.72 mm, leaf area 77.59-312.92 cm2, average tree vigor 6.17, tree habit 1.66, shape of leaf tip 1.56 (Table 4).
Table 4
Descriptive statistics of loquat genotypes for morphological characteristics
No | Character | Abbreviation | Unit | Min. | Max. | Mean | SD | CV (%) |
1 | Leaf Width | LWi | cm | 5.41 | 12.71 | 8.40 | 1.55 | 18.46 |
2 | Leaf Length | LL | cm | 15.63 | 41.22 | 24.43 | 4.14 | 16.94 |
3 | leaf Length/Width ratio | LLWR | % | 1.37 | 5.57 | 3.00 | 0.71 | 23.54 |
4 | Petiole Diameter | PD | mm | 3.74 | 8.00 | 5.76 | 0.75 | 13.03 |
5 | Petiole Length | PL | mm | 8.87 | 46.72 | 24.45 | 8.72 | 35.65 |
6 | Leaf Area | LA | cm2 | 77.59 | 312.92 | 137.68 | 37.92 | 27.54 |
7 | Tree vigor | TV | code | 5.00 | 7.00 | 6.17 | 0.99 | 16.08 |
8 | Tree habit | TH | code | 1.00 | 3.00 | 1.66 | 0.93 | 56.16 |
9 | shape of Leaf tip | SLT | code | 1.00 | 3.00 | 1.56 | 0.86 | 55.36 |
In case of tree vigor, it was determined that 41 (58.5%) of the 70 genotypes used in the study had strong growth strength and 29 (41.59%) had medium growth strength (Fig. 2). When the loquat genotypes were examined in terms of tree growth habit, it was found that 46 (65.7%) of the genotypes were upright, 2 (2.9%) were semi-upright and 22 (31.9%) were with spreading growth habit (Fig. 2). When the loquat genotypes were considered in terms of leaf tip shape, 49 (70%) of the genotypes were sharp acute, 17 (24%) were round, and 4 genotypes (6%) were blunt acute (Fig. 2).
Selected promising genotypes
The scores of the examined genotypes were obtained 90–210 in terms of fruit weight, 45–105 in terms of fruit pulp/seed ratio, 60–140 in terms of TSS, 30–70 in terms of TSS/acid ratio, 10–50 in terms of the number, and external quality (appeal) between 15–105 points. The total scores based on the “Weighed Rating” of the 70 loquat genotypes examined ranged from 320 to 580. The genotypes GN10, GN14, GN50, GN68 and GN69 were determined as superior based on high scores which they were received.
Correlation, PCA and HCA of chemical characteristics
Pearson correlation coefficients between chemical charateristics illustrated in the (Fig. 3). The correlation coefficients are calculated using total number of 70 genotypes. Correlation between SSCAcR and AC (0.836) obtained strongly negative. While DW and FREWE (0.636), pH and SSCAcR (0.544) showed significantly stronge positive correlation.
PCA of chemical characteristics illurstrated in the (Fig. 4). The contribution to PC 1 (29.10%) is shown on the x-axis, while the contribution to PC 2 (20.40%) is on the y-axis. The values between the brackets indicate the percentage of the variance explained by the individual PCs. Each characters’ contribution to the selected principle component is indicated by the length and color of the arrow. PCA was performed on raw data using the following characteristics; Ta, Ju, SSC, Ac, SSCAcR, pH, DM, FreWe, and DW.
Eigen values of component 1 to component 5 obtained respectively as the following (2.621, 1.833, 1.347, 0.976, and 0.920) and Scree plot of eigenvectors and their percentage of variences illustrated in the (Fig. 5). Percentage of variance explained by each principle component is shown from the most significant to the least respectively. For 5 first components respectively observed (29.13%, 20.37%, 14.96%, 10.84%, and 10.22%) which cumulatively encompassed 85.54% of variances, the remained percentage related to the components 6 to 8.
The dendrogram in the (Fig. 6) representing the clustering of genotypes based on chemical characteristics (Ta, Ju, SSC, Ac, SSCAcR, pH, DM, FreWe, and DW). Samples were clustered using Ward method. The presented dendrogram was calculated using raw data of 70 genotypes used in the study.
Correlation, PCA and heatmap of morphological characteristics
Pearson correlation coefficients between morphological charateristics illustrated in the (Fig. 7). The correlation coefficients are calculated using total number of 70 genotypes. Correlation between LWi and LLWR (-0.670), LL and LLWR (0.597), LWi and PD (0.311), LLWR and PD (-0.312), PD and PL (-0.260) obtained significant at P < 0.05.
PCA of morphological characteristics illurstrated in the (Fig. 8). The contribution to PC 1 (28.50%) is shown on the x-axis, while the contribution to PC 2 (22.20%) is on the y-axis. The values between the brackets indicate the percentage of the variance explained by the individual PCs. Each characters’ contribution to the selected principle component is indicated by the length and color of the arrow. PCA was performed on raw data using the following characteristics; LWi, LL, LLWR, PD, PL, LA, TV, TH, and SLT.
Eigen values of component 1 to component 4 obtained respectively as the following (2.568, 1.996, 1.446, and 0.964) and Scree plot of eigenvectors and their percentage of variences illustrated in the (Fig. 9). Percentage of variance explained by each principle component is shown from the most significant to the least respectively. For 4 first components respectively observed (28.54%, 22.18%, 16.06%, and 10.70%) which cumulatively encompassed 77.50% of variances, the remained percentage related to the components 5 to 9.
The heatmap of morphological characteristics which used for clustering presented in the (Fig. 10). Traits and individual samples are clustered using average method. Columns represent the mean of individual genotypes, while the rows are representing selected morphological characteristics (LWi, LL, LLWR, PD, PL, LA, TV, TH, and SLT). The presented heatmap was calculated using raw data of 70 genotypes in case of morphological traits. The average number of replicates applied in the analysis was 1 replication in case of each genotype. Yellow and blue colors represent high and low character value respectively. The values of individual samples are normalized per character using z-Fisher transformation.