Diversity analysis of phenotypic characters
Results of statistical analysis of seven qualitative characters of 145 ancient tea germplasm resources are tabulated in Table S3. It can be observed that leaf tooth depth displayed the highest coefficient of variation (50.64%), whereas leaf quality showed the lowest coefficient of variation. The coefficients of variation for phenotypic characteristics of leaves such as their tooth depth, tip, color, size, shape, quality, as well as for that of tree shape ranged within the minimum and maximum values of coefficients of variation (Table S3). Further, the diversity indices for all phenotypic characters except leaf quality were found to be higher than 0.75.The diversity index for leaf quality was 0.55. Leaf color displayed the highest diversity index of 0.84.
A statistical analysis of six quantitative characters shown in Table S4 revealed that tree height had the highest (60.37%) and leaf vein logarithm had the lowest (17.76%) coefficients of variation, respectively. The coefficients of variation decreased in the following manner of quantitative characters-tree height, leaf area, the logarithm of leaf teeth, leaf length, leaf width, and logarithm of leaf vein. The diversity indices of all quantitative characters were above 1.5, the lowest Leaf vein logarithm showed the lowest (1.72), whereas the leaf area displayed the highest (2.74) diversity indices, respectively. Resultant values of the statistical analysis indicate the rich genetic diversity in the phenotypic characters of ancient tea germplasm in Sandu County, Guizhou province.
Phenotypic character diversity analysis of five ancient tea germplasm populations
The variation coefficients of ancient tea germplasm were observed to significantly vary among different populations and for phenotypic characteristics (Table S5). The tree type for the ancient tea germplasm collected from Landong and Guqi villages were shrubs, and hence the coefficient of variation was not included in the study. Among the ancient tea germplasm of arbors, the variation coefficient of leaf tip in Yangmeng village was observed to be the highest (41.7%), whereas that in Zenya village, was found to be the lowest (30.84%). There was no obvious difference in the coefficient of variation of leaf shape between 19.57% and 25.66%. The largest variation coefficient of leaf tooth depth of the ancient tea germplasm population in Zenya village was 57.77%, and the smallest was 34.54% of the ancient tea germplasm population in Landong village. The largest variation coefficient of leaf color was 55.8% of the ancient tea germplasm in Yangmeng village, and the smallest was 34.81% of the ancient tea germplasm in Landong village. The observed variation coefficients of leaf vein logarithm, leaf teeth logarithm, leaf length, and leaf width were small. The variation coefficient of tree height was the highest in the ancient tea population in Zenya village. Among the two shrub groups of ancient tea germplasm, the variation coefficient range of Guqi village was the widest and ranged from 13.05% to 55.36%.
It can be observed that the average genetic diversity index of the arboreal ancient tea germplasm populations is relatively high (Table S6). The genetic diversity index of leaf tip was the highest in Yangmeng village (1.0889) and the lowest in Landong village (0.6089). The highest genetic diversity of leaf shape and leaf size was observed in the ancient tea germplasm population of Landong village, and the values were 1.0209 and 0.7653, respectively. The highest diversity index of leaf tooth depth was observed for the ancient tea germplasm group in Yangmeng village. The highest genetic diversity index of leaf color was observed to be 1.0067 in Zenya village. The highest genetic diversity index of leaf veins was 2.6921 in Zenya village, and the lowest value, i.e., 1.8867, was observed for Yangmeng village. The highest genetic diversity index of leaf tooth logarithmic was observed for the ancient tea germplasm in Landong village, and the lowest was noted for Yangmeng village. The highest genetic diversity index of tree height belongs to the ancient tea germplasm in Landong village, which is 3.3431. The genetic diversity index of leaf length ranged from 2.3026 to 3.7226, with the highest diversity. The genetic diversity index of leaf width is the highest (3.3193) in Landong village. Further, the lowest was the ancient tea group in Yangmeng village. The genetic diversity index of leaf area was reported to be the highest in Zenya village, as 3.9890, followed by Landong village (arbor) and Yangmeng village. Between the two shrub ancient tea germplasm groups, the genetic diversity indices of leaf size, leaf color, leaf tooth logarithm, tree height, leaf length, leaf width, and leaf area in Guqi village were higher than those of the shrub ancient tea germplasm group in Landong village. This indicates that the genetic diversity index of the ancient tea germplasm in Guqi village is high.
Principal component analysis of phenotypic characters
In order to clarify the role of each character in the phenotypic diversity of ancient tea germplasm, 13 phenotypic characters were subjected to principal component analysis (Upadyayula et al. 2006). The contribution rate of the first six components reached 85.48%, indicating that they can reflect the features of 13 phenotypic characters (Table S7) (Tehrim et al. 2012). The correlation analysis between the original characteristic variables and the six principal components (Table 2) found that PC1 mainly represented the leaf size, leaf length, leaf width, and leaf area. Its principal component can be defined as the size factor of the leaf. PC2 mainly represents leaf shape and leaf shape index, while PC4 mainly represents leaf tip and leaf vein logarithm, both of which are related to the leaf shape. These two principal components are defined as the leaf shape factor. PC3 mainly represents the depth and logarithm of leaf teeth, which are defined as the leaf teeth factors. PC5 mainly represented leaf quality and was defined as a leaf quality factor. In a similar way, PC6 was defined as the vein factor on behalf of the leaf vein logarithm.
The principal component analysis of ancient tea germplasm (Table S8) indicated that the high positive value of PC1 reflects that the leaf size of ancient tea germplasm resources was dominated by middle and large leaves. LDT8, ZYT59, ZYT62, ZYT67, ZYT69, ZYT70, ZYT71, ZYT73, ZYT74, ZYT76, ZYT81, ZYT82, ZYT84, ZYT85, ZYT86, ZYT87, ZYT90, ZYT91, ZYT94, ZYT95, ZYT97, ZYT98, YMT99, and YMT101 belong to this category. On the other hand, a lower negative value of PC1 indicates that the leaves of this ancient tea germplasm are smaller. High positive values of PC2 indicated that the leaf shape of these materials tended to belong to elliptic or lanceolate. LDT14, LDT20, LDT33, LDT35, LDT36, LDT54, ZYT55, ZYT56, ZYT57, ZYT58, ZYT61, and ZYT63 can be classified into this category. A low negative value of PC2 indicates that the leaf shape of these materials tends to be round, and LDT21, LDT22, ZYT71, GQB121, GQB123, and GQB124 belong to this category. A higher positive PC3 value indicates that this kind of material has deeper leaf teeth and more logarithm of leaf teeth, ZYT69, YMT100, YMT101, GQB109, GQB110, GQB111, GQB112, GQB113, and GQB115 belong to this kind of material. A lower negative PC3 value indicated that the leaf teeth of this kind of material are shallow, and the logarithm of leaf teeth has a lower value. The high positive value of PC4 indicated that there were many vein logarithms of this material, and the leaf tip is obtuse, and LDT1, LDT2, LDT3, LDT4, LDT14, LDT35, YMT100, YMT101, YMT102, LDB139, and LDB143 can be assigned to this category. A lower PC4 negative value reflects that the leaf tip of these materials is sharp, and the logarithm of the leaf vein is less. The high positive values of PC5 indicated that the leave of such materials are relatively hard, such as LDT10, LDT16, LDT46, LDT47, ZYT58, ZYT64, ZYT66, ZYT68, ZYT82, ZYT83, ZYT96, YMT99, LDB134, LDB136, LDB137, etc. A low negative value of PC5 indicates that the leaves of these materials are soft, LDT5, LDT8, LDT9, LDT11, LDT12, LDT20, LDT29, LDT43, LDT44, LDT45, LDT48 belong to this category. A high positive value of PC6 indicates that this kind of material has more leaf vein logarithms, and LDT2, LDT5, LDT7, LDT9, LDT12, LDT20, LDT33 are in this category. A low negative value of PC6 indicates that the leaf vein logarithms are less, including LDT21, LDT29, LDT40, LDT43, ZYT63, and LDT52.
Cluster analysis of phenotypic characters
The cluster analysis with a genetic distance of 10, divided the 145 experimental samples used in this study into 4 categories. The first class comprised 31 ancient tea germplasm, which included ancient tea germplasm from trees in Landong village and ancient tea germplasm from all shrubs. The second class consisted of a total of 38 ancient tea germplasm, composed of arboreal ancient tea germplasm in Landong village and all ancient tea germplasm in Yangmeng village. Ancient tea germplasm of a total of 22 shrubs from the Guqi village formed the third class, whereas the fourth class comprised of 54 arboreal ancient tea germplasm in Zenya village.
Geneticdiversity analysis of germplasm resources of ancient tea germplasm
Three sets of duplicate data were analyzed by POPGENE 32 (Table S9 and Table S10) (Liu et al. 2016). The results showed that the genetic consistency between 145 samples of ancient tea germplasm was 0.5765–0.9529. The genetic consistency of arbor lies between 0.5882–0.9529, and the largest one is for the ancient tea germplasm numbered 25 and 37. It indicates that the genetic background is similar, and the similarity is high. The lowest genetic consistency was numbered 48 and 102, indicating their genetic distance is relatively far. The genetic consistency of ancient tea germplasm of shrubs ranges from 0.6118 to 0.9294. The trees numbered 124 and 126 have the highest genetic consistency, indicating that their genetic backgrounds are quite similar. The two shrubs numbered 113 and 132 have the lowest genetic consistency, indicating that their genetic distance is relatively far. The genetic distance between trees and shrubs consistency is 0.5765–0.9176. The maximum genetic consistency was observed between the arbor numbered 105, and the shrub numbered 108. The minimum genetic consistency was noted between the arbor no. 48 and the shrub no. 127, indicating that the genetic distance between these two ancient tea germplasms was relatively farther. POPGENE 32 software was used to analyze the genetic diversity index (Table 3) of the 145 ancient tea resources. The results showed that the average observed allele number (Na) was 2.0000, the average effective allele number (Ne) was 1.3984, the average Nei’s genetic diversity index (H) was 0.2584, and the average Shannon information diversity index (I) was 0.4119. The results indicated that the genetic diversity among the 145 ancient tea germplasm materials is relatively high.
Genetic diversity parameter analysis of ancient tea germplasm at different sampling points
Comparing the same genetic parameters between the samples collected at different locations (Table S11) enabled the researchers to understand the differences between them. Compared with the arboreal ancient tea germplasm population in Landong village, the allele number (Na) of the arboreal ancient tea germplasm population in Zenya village increased by 2.21%, the Na of the shrubby ancient tea germplasm population in Guqi village decreased by 2.65%, and the Na of the shrubby ancient tea germplasm population in Landong village decreased by 7.52%. In contrast to the Shannon information index (I) of the ancient arboreal tea group in Landong village, the value of I for the ancient arboreal tea group in Zenya village was increased by 4.02%. The largest decrease was noted as 8.46% in the arboreal ancient tea group in Yangmeng village. On the other hand, in contrast to the polymorphism points of the arboreal ancient tea germplasm population in Landong village, in addition to the 3.80% increase in Zenya village, the others were decreased by 27.85%, 6.33%, and 16.46%, respectively. Among them, the largest decrease occurred in the ancient tea germplasm group in Yangmeng village. In contrast to with the percentage of polymorphism sites (PPB) in the arboreal ancient tea population in Landong village, the PPB in the arboreal ancient tea population in Zenya village increased by 4.61%, the PPB in the arboreal ancient tea population in Yangmeng Village decreased by 27.53%, the PPB in the shrubby ancient tea population in Guqi Village decreased by 5.53%, and the PPB in the shrubby ancient tea population in Landong Village decreased by 15.71%. Overall, the genetic diversity of arboreal ancient tea germplasm was richer than that of shrubby ancient tea germplasm.
UPGMA cluster analysis
Cluster analysis of 145 ancient tea germplasm was carried out with NTSYS-PC 2.1 (Rout et al. 2009), and the corresponding UPGMA tree diagram (Figure 2) was obtained. It was found that a genetic similarity coefficient of 0.734 divided the 145 ancient tea germplasm into four categories. The first category included 44 ancient tea germplasm of Landong village. The second category comprised 54 ancient tea germplasm in Zenya village, and the third category had nine ancient tea germplasm in Yangmeng village. The fourth category was further subdivided into two subclasses. The first subclass included 22 parts of the shrub ancient tea germplasm of Guqi village. The second subclass comprised 15 parts of the shrub ancient tea germplasm of Landong village as well as 1 part of the shrub ancient tea germplasm of Yangmeng village. The results from the cluster analysis suggest that the shrub ancient tea germplasm and the arboreal ancient tea germplasm have a greater genetic distance and far relationship, which conforms to the phenotypic clustering results. Moreover, the clustering results are closely related to the distribution region. Hence all ancient tea germplasm collected from a specific region were grouped under one category. Thus, it can be said that SSR markers could be used to classify and identify the germplasm resources of ancient tea germplasm and study the relationship between them.