Skip to main content
Download PDF

Genetic imprints of grafting in wild iron walnut populations in southwestern China

BMC Plant Biology volume 23, Article number: 423 (2023) Cite this article

Abstract

Background

Anthropogenic activities are causing unprecedented loss of genetic diversity in many species. However, the effects on genetic diversity from large-scale grafting onto wild plants of crop species are largely undetermined. Iron walnut (Juglans sigillata Dode) is a deciduous nut tree crop endemic to southwestern China with a long history of cultivation. Due to the rapid expansion of the walnut industry, many natural populations are now being replaced by cultivars grafted onto wild rootstocks. However, little is known about the potential genetic consequences of such action on natural populations.

Results

We sampled the scion and the rootstock from each of 149 grafted individuals within nine wild populations of J. sigillata from Yunnan Province which is the center of walnut diversity and cultivation in China, and examined their genetic diversity and population structure using 31 microsatellite loci. Scions had lower genetic diversity than rootstocks, and this pattern was repeated in seven of the nine examined populations. Among those seven populations, AMOVA and clustering analyses showed a clear genetic separation between all rootstocks and all scions. However, the two remaining populations, both from northern Yunnan, showed genetic similarity between scions and rootstocks, possibly indicating that wild populations here are derived from feralized local cultivars. Moreover, our data indicated probable crop-to-wild gene flow between scions and rootstocks, across all populations.

Conclusions

Our results indicate that large-scale grafting has been causing genetic diversity erosion and genetic structure breakdown in the wild material of J. sigillata within Yunnan. To mitigate these effects, we caution against the overuse of grafting in wild populations of iron walnut and other crop species and recommend the preservation of natural genotypes through in situ  and ex situ conservation.

Peer Review reports

Background

As a fundamental dimension of biodiversity, genetic diversity represents evolutionary potential, determining long-term maintenance of species [1, 2]. In plants, the breeding system and mode of proliferation contribute significantly to shaping the genetic diversity and population structure [3, 4]. Sexual reproduction based on seed propagation, which usually occurs in natural populations, serves to preserve the evolutionary potential of species since the genetically variable offspring can survive and adapt to varying environmental conditions [5,6,7]. Natural populations of outcrossing species are usually characterized by high within-population diversity and low among-population differentiation [8]. Conversely, clonal propagation, being the commonest type of asexual reproduction [9], has been associated with reduced genetic diversity, especially when it occurs intensively, and can bring about a subsequent decrease in the levels of genetic differentiation within populations, or groups thereof derived from the same source [10,11,12]. This can also lead to genetic sub-structuring among populations [11].

For cultivated plants, clonal propagation, including the regeneration of whole plants via cuttings or grafts, allows for phenotype uniformity, with the desired agronomic traits faithfully passed on to subsequent generations [13]. This has benefits, including increased yield and quality of produce, plus guarantees of efficient nutrient acquisition, accumulation of biomass, and reduced mortality risk when a strong cultivar genotype is selected [9, 12]. This might explain the widespread shift from traditional seed-based breeding to vegetative propagation in many cultivated plant species. However, the resultant clonality often leads to reduced sexual recombination, or none at all [14]. This, coupled with the intensive cultivation of genetically identical cultivars, is likely to reduce the genetic variability within a species, in turn decreasing the species’ adaptability and resilience in the wake of biotic or abiotic stressors. Therefore, it is important to investigate the long-term effects of replacing seed-propagated individuals with grafted cultivars, particularly in long lived tree crops.

Grafting is the fusion of two plant parts so that vascular function can establish between them [15], and the resulting united individual can normally grow like a single plant [16]. Indeed, this has been a pivotal technology through much of human history and is now widely applied in agriculture, horticulture and biology [16] (e.g., rose, cucumber, eggplant, tomato, watermelon, apple, citrus, grape, and pear [17,18,19]). When, where and how fruit tree grafting began is still debated [16, 19], with the earliest reliable written evidence of grafting dating to 412 BCE [16], although the technique might have been used in ancient Mesopotamia around 1800 BCE [20]. Grafting in walnut is much more difficult than in many cultivated taxa because it is impeded by poor callus formation, and hence requires greater precision in terms of time, temperature, humidity, selection, and the expertise to handle scions and stocks. Therefore, traditional propagation of walnut remained through seed for a long time, but these challenges have gradually been resolved by introducing more developed grafting methods such as hot callus, epicotyl grafting, and water nutrient solution systems [21, 22]. Once grafting became feasible, market pressures hastened the replacement of natural seed-propagated walnut trees with grafted cultivars, creating great challenges for walnut resources conservation.

Yunnan Province is well known for its high biodiversity [23], and is the most important center of walnut genetic diversity, plantation, and production in China [24]. The common, Persian or English walnut (J. regia L.) is mainly cultivated in the northwestern parts, whereas both wild and cultivated forms of iron walnut (J. sigillata Dode) are widely distributed in subtropical region of Yunnan. Iron walnuts are highly valued for their nutritious nuts and high-quality timber, as well as its religious and cultural significance [25, 26]. Wild material of this species shows more resistance to pathogens and pests than do cultivars such as ‘Dapao’ [24]. Historically, cultivated populations were mainly maintained by grafting onto rootstocks grown from wild iron walnut seed. However, grafting has recently been widely applied to natural iron walnut forests (Figs. 1 and 2), as a consequence of the rapid expansion of the walnut industry over the past three decades. Consequently, sexual reproduction is increasingly being replaced by artificial asexual reproduction. Such wild populations now contain both grafted and naturally seed-propagated individuals, and can hence act as an ideal model to test the population genetic effects of grafting. Thus far, to the best of our knowledge, comparative analysis of genetic diversity of the scion and rootstock components has only been undertaken for Mediterranean old olives [27], with rootstocks showing higher genetic diversity than scions. However, that study [27] involved only cultivated trees, and we know of no other study examining the genetic impact of grafting within natural populations. Furthermore, little attention has been paid to the effect of grafting on the genetic diversity and population structure of wild stands of walnuts.

Fig. 1
figure 1

A schematic illustration of grafting in iron walnut from the natural forest to the mature grafted individual. The natural forest reproduces sexually through seed propagation, while the cultivar is propagated asexually by cloning. The resultant tree will have different genotypes between the scion and rootstock. The figure was created by Drs. Richard Milne and Jie Liu

Full size image
Fig. 2
figure 2

Photographs of Juglans sigillata forest and trees in Goujie Xiang (耈街乡), Changning County (昌宁县), Yunnan Province. a the natural forest of J. sigillata in the valley near the Longtan village (龙潭组); b-c natural iron walnut forest disturbed by recent artificial grafting and plantation; d-e grafted individuals of natural iron walnut trees, with branches and trunks representing scions and rootstocks, respectively. Photos courtesy of Dr. Jie Liu and Mr. Tao Liu

Full size image

Simple sequence repeats (SSRs) are effective molecular markers, which have been widely and successfully used to infer the level of genetic diversity and population structure in different fruit trees including walnut (e.g. [28,29,30]). Therefore, to examine the genetic impacts of grafting on natural populations of walnuts, and more generally, we used 31 nuclear SSRs to examine nine populations of scions and rootstocks each, to investigate the effect of grafting on the genetic diversity and population structure of wild iron walnut trees in Yunnan. Our results will contribute towards data-driven breeding and conservation of walnut in Yunnan and beyond.

Results

Genetic diversity

A total of 298 samples (Table 1) were successfully screened using 31 microsatellite loci (Table S1). The set of SSR loci used in the study showed a high level of polymorphism (Table S2), minimum stuttering and allele drop-out, low null allele frequencies, and only a few population-locus combinations deviated significantly from Hardy–Weinberg equilibrium (HWE), as per our previous studies [28, 31]. Several clonemates were identified among scions, but very few were detected among rootstocks (Fig. S1). Genetic diversity indices are summarized for each locus (Table S2). The total number of alleles for each locus ranged from 3 (JR09 and JS28) to 12 (JS12) with an average value of 6.74, while the mean observed heterozygosity (HO) and expected heterozygosity (HE) were 0.58 and 0.56, respectively. Mean effective number of alleles (NE), Shannon’s information index (I), and unbiased expected heterozygosity (uHE) among loci were 2.53, 1.06, and 0.56, respectively.

Table 1 Sampling information of J. sigillata grafting system collected in Yunnan, China. The suffix of S and R represent scion and rootstock, respectively
Full size table

Three datasets were compiled, comprising only scions (S), only rootstocks (R), and all samples (T). For each of the  datasets and populations, genetic diversity parameters were calculated across loci (Table 2). A total of 209 alleles were identified for all samples, with the rootstocks exhibiting a higher total number of alleles (NT) (total of 194, and an average of 110 per population) than the scions (total of 171 with an average of 88 per population). The number of private alleles (NP) was higher in the rootstocks (28) than among the scions (13). Similarly, other genetic diversity parameters, namely mean number of alleles (NA) (5.52/6.26), effective number of alleles (NE) (2.25/2.65), HE (0.51/0.58) and allelic richness (AR) (5.55/6.45) were lower in scions than in rootstocks. However, scions had a slightly higher observed heterozygosity (HO = 0.63) than rootstocks (0.53). At the population level (except for populations LH and YS), all genetic diversity indices other than HO were higher in rootstock than in scions. All loci were polymorphic within each dataset (T, S, and R), but the proportion of polymorphic loci (PPL) ranged from 51.6% to 100% for populations. Seven scion populations had negative FIS values, while this was the case for only two of the rootstock populations. Finally, both the Wilcoxon and standardized difference tests independently indicated that two scion populations (YB-S and GB-S) had experienced significant bottlenecks (P < 0.05), whereas an additional four scion (LH-S, YB-S, GB-S and CN-S) and two rootstock (YS-R and CN-R) populations showed a bottleneck signal in one test but not the other (Table 2).

Table 2 Genetic diversity indices and bottleneck test results for whole dataset (T), Scions (S) and Rootstocks (R), and per population of scion and rootstock
Full size table

Patterns of genetic differentiation and population structure

According to AMOVA, when all scion and rootstock populations were placed in a single dataset (T), the among-population and within-population genetic variation for the whole dataset was 26% and 74% respectively (Table 3). When the scion and rootstock populations were analysed separately, the scions (dataset S) showed higher among-population genetic variation (35%) than the rootstocks (dataset R; 14%). Among the three datasets, the populations of scions show the highest genetic differentiation (FST = 0.350), then rootstocks (FST = 0.136), then all populations together (FST = 0.259) (Table 3). The analysis revealed a moderate genetic differentiation between scions and rootstocks (FST = 0.115). Generally, we found a low to moderate pairwise genetic differentiation (FST) between all pairs of sampled populations with LH-R showing genetic distinctiveness from the rest of the populations of both rootstocks and scions (Fig. 3d; Table S3).

Table 3 Analysis of molecular variance (AMOVA) quantifying the distribution of genetic variation for three datasets of Juglans sigillata. Dataset T = all populations together, S = scions only, and R = rootstocks only
Full size table
Fig. 3
figure 3

Genetic relationship and differentiation among scions and rootstocks of Juglans sigillata from nine sites. a Bayesian clustering of 149 scions (S) and 149 rootstocks (R) from 9 sites of J. sigillata from K = 2 to K = 4. b A radial distance neighbor-joining (NJ) tree showing relationships among all nine population pairs of scions (S) and rootstocks (R). c Principal Coordinates Analysis (PCoA) of all rootstocks and scions. The first and second components explain 17.68% and 8.89% of variation, respectively. Dots and triangles denote scions and rootstocks respectively. d Heat map depicting pairwise FST for both scion and rootstock populations; the magnitude of differentiation is indicated on the color ramp ranging from 0.00 to 0.80

Full size image

The STRUCTURE analysis (Fig. 3a) revealed that ΔK had its maximum value at K = 2, making this the optimal K value (Fig. S2). At K = 2, one cluster contains mainly scions and the other mostly rootstocks, except for three scion populations (LH-S, YS-S and GB-S) which grouped with the rootstocks. Three other scion populations, YG-S, FQ-S, and YD-S showed admixture between the two clusters. Most rootstock populations other than LH-R and GY-R showed some degree of admixture within populations, often with certain individuals belonging to the scion cluster. When K = 3, a third genetic cluster is formed containing both scions and rootstocks from populations LH and YS, plus scions of GB-S. At K = 4, another new admixed genetic cluster emerges containing mainly parts of individuals in YB-R, CN-R, YG-S and GB-S, plus some individuals of FQ-S.

The radial neighbor-joining (NJ) tree divided the 18 populations into three major groups (Fig. 3b). Of these, one comprised scions only (six populations, i.e. YG-S, YD-S, FQ-S, CN-S, GY-S and YB-S), the second comprised rootstocks only (seven populations, i.e. YB-R, CN-R, GY-R, GB-R, YG-R, YD-R, and FQ-R), while the third cluster comprised both scion and rootstock population of YS and LH, plus GB-S. This clustering pattern was generally consistent with that observed in STRUCTURE at K = 3 (Fig. 3a).

PCoA results indicated that most scions clustered closely, with a few mixing with rootstocks. Most rootstocks showed a much more dispersed distribution than scions (Fig. 3c). These three groupings were also generally consistent with STRUCTURE results at K = 3 (Fig. 3a).

Spatial genetic structure

Due to the high positive association among the various genetic diversity indices (Table 2), we mapped only the geographical distribution of two genetic diversity index (AR and HE) and genetic structure (at K = 3) (Fig. 4). The results suggest that for the seven sites in the southern region (i.e. YD, FQ, CN, GY, YB, YG, and GB), the genetic diversity of scions was lower than the value for rootstocks. However, an opposite trend was revealed in two northern sites (YS and LH). Regarding genetic structure, the genetic composition of six sites (i.e. YD, FQ, CN, GY, YB, and YG) from the southern region showed clear genetic differentiation between scions and rootstocks, but with some genetic overlap between them. The two northern sites showed similar genetic composition to one another, but within them, no obvious genetic differences were observed between scions and rootstocks. Population GB had a striking genetic difference between scions and rootstocks, with the scions (GB-S) being more similar to the scions of two northern sites, whereas the rootstocks (GB-R) were closest to the rootstocks from the rest of the southern populations (Figs. 3 and 4).

Fig. 4
figure 4

Spatial distribution of genetic diversity of the scions and rootstocks of iron walnut in Yunnan Province. The left-upper inset shows the geographical location of the study region in Asia. The dark points indicate sampling localities along with their population codes (see Table 1). For each site, the histogram represents the AR, HE and genetic structure (K = 3) for scion (S) and rootstock (R) from left to right. The genetic proportions of each cluster correspond to those in Fig. 3a

Full size image

Discussion

Natural forests of J. sigillata provide a reservoir of genetic diversity, which is potentially under threat from the use of wild trees as rootstocks to graft cultivars for walnut industry. Therefore, the iron walnut is an ideal model to examine how grafting affects genetic diversity and population structure in fruit crops, and possibly other wild plants.

Erosion of genetic diversity induced by grafting

The levels of genetic diversity in a plant species are often determined by intrinsic (e.g. life form, population history, and breeding system) and extrinsic (e.g. ecological factors, interspecific competition and human activity) factors [3, 4]. Globally, there has been an exceptionally rapid loss of genetic diversity due to intensive anthropogenically-driven environmental change [32, 33]. One such anthropogenic activity, the grafting of a few agronomically elite cultivars onto natural trees, may cause rapid genetic diversity loss (Fig. 1). Each such graft removes from the population one naturally sown tree, and hence reduces natural genetic diversity. Our results do indeed show that genetic diversity is being eroded by grafting within natural populations, because at least 23 alleles (NT) including 15 private alleles (NP) have been lost in the scions relative to the rootstocks, and all genetic diversity indices except HO are lower in scions than rootstocks (Table 2). The higher HO in scions indicates heterozygote excess, which is expected because recurrent artificial clonal propagation tends to preserve highly fit genotypes exhibiting heterosis and has also been observed in clonal material of Prunus amygdala “Halwani” [34] and P. avium [10]. At the population level, the same pattern of scions having lower diversity than rootstocks was repeated in all populations except the northern populations LH and YS (Fig. 4; Table 2; see below second point for more discussion).

The levels of genetic diversity observed within the rootstock populations (Table 2) were generally comparable to that of Citrus rootstocks in Tunisia [35], but lower than that for suckers derived from Olea europaea rootstocks in the Mediterranean [27] and even the seed propagated ‘Khachabi’ cultivar of Prunus amygdalus in Lebanon [34]. However, such direct comparison of genetic diversity among species should be done cautiously due to variations in life history traits, molecular markers and even the sources of rootstocks. The relatively low levels of genetic diversity of rootstocks in the current study may indicate that certain wild populations of J. sigillata may already contain rootstocks originating from naturalized cultivars. Consistent with this, the common walnut (J. regia) exists as naturalized rather than native populations in most of Europe [36] and China [37]. Therefore, a similar scenario might be true for J. sigillata. The genetic diversity among J. sigillata scions (Table 2) is also lower than that for Mediterranean Olea europaea scions [27], but is consistent with the generally low genetic diversity of J. sigillata cultivars from southwest China [28]. Such patterns clearly reflect both the narrow genetic background of cultivars, and the tendency to clonally propagate a few choice genotypes, for example by grafting [28]. One cause of low genetic diversity in cultivars is bottleneck events (Table 2) during selective breeding, as previously seen in Prunus avium [38].

Distortion of population structure in grafted populations

Our results revealed a moderate genetic differentiation between scions and rootstocks (FST = 0.115; Fig. 1), and hence grafting may have caused changes to population genetic structure in natural walnut forests. Clear genetic divergence between scion and rootstock populations was observed in clustering analysis for the seven sites in the southern region (Figs. 3 and 4; Table S1). No fertile shoots were observed from below the grafting mark on any sampled individual, indicating that population structure of wild iron walnut may have been permanently distorted by grafting. Moreover, there was considerable genetic admixture between scions and rootstocks (Fig. 3), which might suggest crop-to-wild gene flow via seed or pollen from scions. Interfertility between J. sigillata and J. regia [25, 39, 40] points to the absence of breeding barriers within J. sigillata. Therefore, our data indicate that wild populations of J. sigillata, as represented by rootstocks examined here, have already been influenced by gene flow from cultivars, and that this trend is likely to increase with the rapid expansion of walnut plantations.

The low genetic differentiation between scions and rootstocks of two northern sites (LH and YS) (Fig. 4) and between the two sites (Fig. 3), suggested a genetic similarity between grafted walnut trees at these two sites, possibly indicating a common origin. Hence both scions and rootstocks in these populations appear to share a common origin distinct from other populations, and perhaps some or all of the scions might have been derived from genetically superior wild individuals or their cultivated descendants from those same populations. An alternative but yet to be verified explanation is the possibility of a cryptic genetic contribution from more northern J. regia populations to populations LH and YS. However, quantification of such gene flow may need broader geographical sampling.

When interpreting unexpected genetic similarity between rootstocks and scions, as observed in the northern sites LH and YS (Figs. 3 and 4; Table S3), the possibility of mislabeling or misidentification must be considered and, if possible, such similarity be ignored. The mislabeling problem was also encountered in the analysis of olive trees from the Mediterranean [27], mango germplasm evaluation [41], and walnut cultivar fingerprinting [28]. However, in the current study, only trees with clear grafting marks were sampled, and the knowledge of local people was further used to confirm that each sampled individual had been grafted. Moreover, great care was taken to sample scions and rootstocks above and below the mark, respectively (Figs. 1, 2). Therefore, mislabeling can be ruled out, and instead the scion/rootstock similarity in northern Yunnan likely results from a long cultivation history in the region, and we postulate that either the cultivars here were derived from local wild material independently from this process in the south, or rootstock populations in the north in fact derive from feralised local cultivars. Both hypotheses might be true to a degree, although during our fieldwork we only encountered what appeared to be fully wild populations of J. sigillata in the south; the two northern populations LH and YS grew near farmlands, and appeared to be semi-wild. This observation tentatively supports the hypothesis of northern rootstocks deriving from feralised cultivars (see above first point), providing a handy local grafting resource for farmers, and explaining the genetic similarity of rootstocks to scions in these populations.

Throughout Yunnan, we detected a relatively low genetic differentiation among rootstock populations, but a relatively high level of genetic differentiation among scion populations (Figs. 3 and 4; Table 3). The former is a common scenario for most outcrossing species [8], which J. sigillata is [26], and so is an expected result for rootstocks, which originate as individuals within a wild population. Conversely, the relatively higher genetic divergence among scion populations is likely an artificial factor, reflecting the diverse origins of cultivars used for scions. Compared to rootstocks, less gene flow is expected among scions due to asexual propagation, and in fact, local adaptation and artificial selection would generate gradual divergence between cultivar populations used as scions. However, artificial movement of scion material between populations would enhance variation within populations while reducing differentiation between them, much like gene flow in natural populations. Such movement might, for instance, explain the strong similarity between populations LH and YS, or the scions of CN, GY and YB (Figs. 3 and 4). Overall, our research indicates that most scions form a distinct genetic population of cultivars that is distinct from the rootstocks, but that northern cultivars might have a separate origin, and/or that northern rootstocks represent naturalized rather than native material. Meanwhile, the question of the origin of cultivated iron walnut remains unresolved and needs more taxonomic, geographical, and genomic sampling in the future.

Implications for conservation and crop improvement

Genetic diversity is the fundamental basis for population fitness and maintenance of adaptive potential to respond to environmental changes, and hence has great conservation significance [1, 2, 42]. The extant genetic diversity of a species is a product of mutation, genetic drift, and selection over a long period [4, 43], but can be rapidly reduced by anthropogenic activities. Our data confirm that the replacement of wild individuals by grafted scions can have such an effect in J. sigillata, wherein scions exhibit lower genetic diversity than rootstocks (Fig. 4; Table 2), a scenario that might also be present in other crop species, e.g., grafted olive trees [27]. Loss of genetic diversity among scions might result in reduced ability to withstand biotic factors such as pests and pathogens, which are currently facing the walnut industry. Likewise, reduced genetic diversity in wild rootstock populations reduces the available natural genetic variation that could be used to buffer against such threats. These findings hence indicate that genetic diversity may be eroded from natural forests if large numbers of wild individuals are replaced by a few grafted cultivars. Therefore, in situ conservation of intact wild trees is recommended, because wild populations that retain genetic diversity can preserve co-adapted gene complexes as well as alleles not present in cultivated material, and hence have the capacity to evolve resilience in the wake of novel abiotic and biotic stresses [2]. With regard to iron walnut, we recommend that the remnant wild forest that includes the seven southern sites (Fig. 4) should be prioritised for in situ conservation. In addition, we recommend adoption of ex situ conservation, especially for material from sites with multiple private alleles. Since rootstocks used as grafting material no longer produce seeds (Fig. 2), seed collection from these rootstocks should focus on the remnant mature iron walnut trees. However, in populations where few or no such individuals remain, there might be a possibility to propagate wild material via offshoots from rootstocks. Moreover, breeding among ex situ populations should be done cautiously, because highly structured populations are prone to outbreeding depression [43, 44]. Finally, to ensure long term preservation and maintenance of genetic diversity, and hence overall crop health, we recommend that farmers adopt a mixed plantation approach encompassing both seed-based propagation of wild trees and the grafting of cultivars of iron walnut away from wild material.

Conclusions

Detailed knowledge regarding the genetic diversity and differentiation of scions and rootstocks of a species provides important information for the conservation of genetic resources to prevent future loss of genetic diversity caused by global environmental change. Our results indicated that rootstocks had higher levels of genetic diversity than scions, suggesting that grafting onto wild trees causes genetic erosion. Meanwhile, contrasting genetic composition between scions and rootstocks, hinted that the population structure was distorted due to grafting. Therefore, conservation action is urgently needed, and we recommend both ex situ and in situ conservation of natural J. sigillata populations, and caution against the overuse of grafting onto natural rootstocks. The same issues are likely to occur in other crop trees where wild trees are used as grafting for scions, in Yunnan (e.g. pear and persimmon) and elsewhere. For future studies, we recommend inclusion of range-wide sampling with genomic, phenotypic and environmental data, to examine differences between rootstocks and scions and examine the impact of grafting upon adaptability to stresses such as pathogens and climate change.

Methods

Population sampling

Between 2016 and 2018, a total of 149 grafted trees were sampled from nine sites containing natural J. sigillata populations that include grafted individuals, distributed across the five well known walnut production areas of Yunnan Province (Table 1). We sampled 19 to 22 individuals at all sites except LH and GB, where only 6 and 4 individuals could be sampled, respectively, due to annual intensive management that had removed all new shoots from most rootstocks. All sampled populations occupied semi-natural habitats that are currently facing considerable anthropogenic pressure (Fig. 2). To avoid sampling natural clones, all sampled trees were separated by at least 50 m. Based on information gathered in the field, all sampled individuals were first identified as having been grafted within the past 30 years. Grafted trees were identified by the presence of clear grafting marks separating the trunk from the shoot (Figs. 1 and 2), and confirmed by interviewing local people. From each individual, samples were collected from above and below the grafting mark separately, with the former sampling the scion and the latter the rootstock (Fig. 2d, e). Only clean and healthy leaves were sampled and subsequently dried in silica gel. Dr. Jie Liu identified voucher specimens, and further prepared and deposited them in the Herbarium of Kunming Institute of Botany, Chinese Academy of Sciences (KUN).

DNA extraction and SSR genotyping

Total genomic DNA was extracted from about 20 mg of silica gel-dried leaf tissue following a modified CTAB procedure [45, 46], and the concentrations of DNA solutions were adjusted to 30–50 ng/µL for downstream PCR amplification. The 298 samples from 149 grafted trees were genotyped using 31 SSR primer pairs used previously in our studies. This set of polymorphic primers was previously developed using an approach that integrates genome screening and cross-species amplification [31]. Subsequently, these loci have proved to be reliable in characterizing the genetic diversity and population structure of walnut [28,29,30]. These microsatellite loci included 20 with tri, seven with tetra, three with penta, and one with hexa repeat motifs (Table S1). All the forward primers were fluorescently labeled with FAM, HEX, or TAMRA dyes (Optimus Bio, Kunming, China) at the 5’ end. The 31 primer pairs were grouped in five multiplexes based on size and color following our previous studies [28, 29]. Multiplex PCR amplification was conducted on a Veriti® 96-Well Thermo-Cycler (Applied Biosystems, Foster City, California, USA) using a 15 µL multiplex PCR composed of 1 µL each of reverse and forward primers, 2 µL of the DNA template, and the remaining volume topped up with Golden Star T6 Super PCR Mix (Tsingke, Beijing, China). The following cycling regime was used: initial denaturation at 98 °C for 2 min, 35 cycles at 98 °C for 10 s, primer annealing temperatures (53 °C -61 °C) (Table S1) for 15 s, 72 °C for 10 s, then a final extension at 72 °C for 5 min, with a holding temperature of 4 °C. The fragment sizes of PCR products were determined using an ABI 3730xl DNA Analyzer (Applied Biosystems, Foster City, CA, USA).

Data organization

We used GENEMARKER v2.2.0 (SoftGenetics, State College, PA, USA) for data viewing and allele scoring. Null and stuttered alleles were examined by MICRO-CHECKER v2.2.3 [47], whereas GENODIVE v3.0 [48] was used to identify multi-locus genotypes (MLGs) for both the scions and rootstocks.

At the population level, scions and rootstocks at each site were treated as separate populations (e.g. CN-S and CN-R refer to the scion and rootstock populations from site CN (Goujie Xiang (耈街乡), Changning county (昌宁县)), respectively) (Tables 1 and 2). To allow for clear comparison and analysis, three datasets were compiled: dataset T comprised all populations, whereas datasets S and R comprised all scion, and all rootstock populations respectively.

Estimation of genetic diversity

The level of polymorphism of microsatellite loci for the current study was tested by calculating the number of alleles (NA), effective number of alleles (NE), Shannon’s information index (I), observed (HO) and expected heterozygosity (HE), and unbiased expected (uHE) heterozygosity using GENALEX v6.51b2 [49]. For datasets T, S, and R, and each of the populations, the genetic diversity analyses were conducted in GENALEX for both the scion and rootstock populations. In addition to the diversity parameters estimated for SSR loci, other indices including the total number of alleles (NT), private alleles (NP), and percentage of polymorphic loci (PPL) were also calculated in GENALEX. Rarefied allelic richness (AR) was computed using Hierfstat R package [50], with a minimum sample size of four genotypes. Inbreeding coefficient (FIS) was calculated using FSTAT v2.9 [51]. To identify possible population bottlenecks based on the allele frequency spectrum, BOTTLENECK v1.2.02 [52] was used, applying the default settings and tests for microsatellite data (two-phased model: 90% one-step mutations and 30% multi-step changes, evaluation with standardized differences and Wilcoxon tests; [53]).

Genetic differentiation and population structure analysis

The estimation of genetic differentiation was conducted for each of the three datasets. To minimize the impact of null alleles on this [54], estimation of pairwise differentiation (FST) among populations involved a correction for null alleles using FreeNA [54], with a bootstrap resampling of 10 000, and results presented in the form of a heatmap using the pheatmap R package [55]. Analysis of molecular variance (AMOVA) was performed in GENALEX to quantify the partitioning of genetic variation both between and within populations. Additionally, a model-based clustering approach implemented in STRUCTURE v2.3.4 [56] was used to infer the clustering of individuals. This was run with different values of the numbers of clusters (K) varying from 1 to 15 under the admixture model with correlated allele frequencies and without a priori grouping assumptions. To confirm the reliability of the results, we performed 10 independent runs per K value with a 10 000 burn-in period and 100 000 MCMC replications. The best K value was then determined in the web-based STRUCTURE HARVESTER v0.6.94 [57]. Furthermore, principal coordinates analysis (PCoA) was used to infer genetic clustering among scions and rootstocks. The PCoA analysis was implemented in GENALEX with a standardized covariance matrix of genetic distances generated according to Smouse and Peakall [58]. Finally, an unrooted neighbor-joining tree (Nei’s genetic distance; 1 000 bootstrap replicates) was constructed using POPULATIONS v1.2.31 [59] and visualized with the online tool ITOL (https://itol.embl.de).

Spatial distribution of genetic variation

To spatially link genetic diversity and genetic structure of scion and rootstock populations with geography, the genetic diversity indices (AR and HE) and genetic structure (STRUCTURE result for K = 3) were subsequently mapped using ArcGIS v10.7 (ESRI, Redlands, CA, USA).

Availability of data and materials

All data in the study are included in this published article. Specifically, the genotyping data of SSR are available in Table S4.

Abbreviations

SSRs:

Simple sequence repeats

KUN:

Herbarium of Kunming Institute of Botany, Chinese Academy of Sciences

A R :

Allelic richness

F IS :

Inbreeding coefficient

F ST :

Genetic differentiation index

H E :

Expected heterozygosity

H O :

Observed heterozygosity

I :

Shannon’s information index

N A :

Number of alleles

N E :

Effective number of alleles

N P :

Private alleles

N T :

Total number of alleles

PPL :

Percentage of polymorphic loci

uH E :

Unbiased expected heterozygosity

AMOVA:

Analysis of molecular variance

PCoA:

Principal coordinates analysis

NJ:

Neighbor-joining

References

  1. Laikre L, Allendorf FW, Aroner LC, Baker CS, Gregovich DP, Hansen MM, et al. Neglect of genetic diversity in implementation of the convention on biological diversity. Conserv Biol. 2010;24(1):86–8.

    PubMed  Google Scholar 

  2. Frankham R, Ballou JD, Ralls K, Eldridge M, Dudash MR, Fenster CB, et al. Genetic management of fragmented animal and plant populations. Oxford: Oxford University Press; 2017.

    Google Scholar 

  3. Loveless MD, Hamrick JL. Ecological determinants of genetic structure in plant populations. Annu Rev Ecol Syst. 1984;15:65–95.

    Google Scholar 

  4. Ellegren H, Galtier N. Determinants of genetic diversity. Nat Rev Genet. 2016;17(7):422–33.

    CAS  PubMed  Google Scholar 

  5. Roze D. Disentangling the benefits of sex. PLoS Biol. 2012;10(5):e1001321.

    CAS  PubMed  PubMed Central  Google Scholar 

  6. Williams GC. Adaptation and natural selection: a critique of some current evolutionary thought. Princeton: Princeton University Press; 2018.

    Google Scholar 

  7. Crow JF. Advantages of sexual reproduction. Dev Genet. 1994;15(3):205–13.

    CAS  PubMed  Google Scholar 

  8. Nybom H. Comparison of different nuclear DNA markers for estimating intraspecific genetic diversity in plants. Mol Ecol. 2004;13(5):1143–55.

    CAS  PubMed  Google Scholar 

  9. Barrett SCH. Influences of clonality on plant sexual reproduction. Proc Natl Acad Sci USA. 2015;112(29):8859–66.

    CAS  PubMed  PubMed Central  Google Scholar 

  10. Stoeckel S, Grange J, Fernández-Manjarres JF, Bilger I, Frascaria-Lacoste N, Mariette S. Heterozygote excess in a self-incompatible and partially clonal forest tree species —Prunusavium L. Mol Ecol. 2006;15(8):2109–18.

    CAS  PubMed  Google Scholar 

  11. Balloux F, Lehmann L, de Meeûs T. The population genetics of clonal and partially clonal diploids. Genetics. 2003;164(4):1635–44.

    PubMed  PubMed Central  Google Scholar 

  12. McKey D, Elias M, Pujol B, Duputié A. The evolutionary ecology of clonally propagated domesticated plants. New Phytol. 2010;186(2):318–32.

    PubMed  Google Scholar 

  13. Altman A. Agricultural biotechnology. Boca Raton: CRC Press; 1997.

    Google Scholar 

  14. Miller AJ, Gross BL. From forest to field: perennial fruit crop domestication. Am J Bot. 2011;98(9):1389–414.

    PubMed  Google Scholar 

  15. Pina A, Errea P. A review of new advances in mechanism of graft compatibility–incompatibility. Sci Hortic. 2005;106(1):1–11.

    Google Scholar 

  16. Mudge K, Janick J, Scofield S, Goldschmidt EE. A history of grafting. Hortic Rev. 2009;35:437.

    Google Scholar 

  17. Lee J-M, Kubota C, Tsao SJ, Bie Z, Echevarria PH, Morra L, et al. Current status of vegetable grafting: diffusion, grafting techniques, automation. Sci Hortic. 2010;127(2):93–105.

    Google Scholar 

  18. Nawaz MA, Imtiaz M, Kong Q, Cheng F, Ahmed W, Huang Y, et al. Grafting: A technique to modify ion accumulation in horticultural crops. Front Plant Sci. 2016;7:1457.

    PubMed  PubMed Central  Google Scholar 

  19. Janick J. The origins of fruits, fruit growing, and fruit breeding. In: Plant breeding reviews, vol. 25. Hoboken: John Wiley and Sons; 2005. p. 255–320.

    Google Scholar 

  20. Harris SA, Robinson JP, Juniper BE. Genetic clues to the origin of the apple. Trends Genet. 2002;18(8):426–30.

    CAS  PubMed  Google Scholar 

  21. Vahdati K, Sadeghi-Majd R, Sestras AF, Licea-Moreno RJ, Peixe A, Sestras RE. Clonal propagation of walnuts (Juglans spp.): a review on evolution from traditional techniques to application of biotechnology. Plants. 2022;11(22):3040.

    CAS  PubMed  PubMed Central  Google Scholar 

  22. Hassani D, Sarikhani S, Dastjerdi R, Mahmoudi R, Soleimani A, Vahdati K. Situation and recent trends on cultivation and breeding of Persian walnut in Iran. Sci Hortic. 2020;270:109369.

    CAS  Google Scholar 

  23. Yang Y, Tian K, Hao J, Pei S. Biodiversity and biodiversity conservation in Yunnan. China Biodivers Conserv. 2004;13(4):813–26.

    Google Scholar 

  24. Xi RT, Zhang YP. Walnut flora. In: Zhao HC, Tian BF, editors. China Fruit-plant Monograph. Beijing: China Forestry Publishing House; 1996.

    Google Scholar 

  25. Gunn BF, Aradhya M, Salick JM, Miller AJ, Yang YP, Liu L, et al. Genetic variation in walnuts (Juglansregia and J.sigillata; Juglandaceae): Species distinctions, human impacts, and the conservation of agrobiodiversity in Yunnan, China. Am J Bot. 2010;97(4):660–71.

    PubMed  Google Scholar 

  26. Lu AM, Stone DE, Grauke LJ. Juglandaceae. In: Wu ZY, Peter RH, editors. Flora of China. Beijing & St. Louis: Science Press & Missouri Botanical Garden Press; 1999. p. 277–85.

    Google Scholar 

  27. Barazani O, Westberg E, Hanin N, Dag A, Kerem Z, Tugendhaft Y, et al. A comparative analysis of genetic variation in rootstocks and scions of old olive trees - a window into the history of olive cultivation practices and past genetic variation. BMC Plant Biol. 2014;14(1):146.

    PubMed  PubMed Central  Google Scholar 

  28. Wambulwa MC, Fan PZ, Milne R, Wu ZY, Luo YH, Wang YH, et al. Genetic analysis of walnut cultivars from southwest China: implications for germplasm improvement. Plant Divers. 2022;44(6):530–41.

    PubMed  Google Scholar 

  29. Magige EA, Fan PZ, Wambulwa MC, Milne R, Wu ZY, Luo YH, et al. Genetic diversity and structure of Persian walnut (Juglansregia L.) in Pakistan: Implications for conservation. Plants. 2022;11(13):1652.

    PubMed  PubMed Central  Google Scholar 

  30. Shahi Shavvon R, Qi HL, Mafakheri M, Fan PZ, Wu HY, Bazdid Vahdati F, et al. Unravelling the genetic diversity and population structure of common walnut in the Iranian Plateau. BMC Plant Biol. 2023;23(1):201.

    PubMed  PubMed Central  Google Scholar 

  31. Xu ZC, Jin YC, Milne RI, Xiahou ZY, Qin HT, Ye LJ, et al. Development of 32 novel microsatellite loci in Juglanssigillata using genomic data. Appl Plant Sci. 2020;8(3):e11328.

    PubMed  PubMed Central  Google Scholar 

  32. Vitousek PM, Mooney HA, Lubchenco J, Melillo JM. Human domination of Earth’s ecosystems. Science. 1997;277(5325):494–9.

    CAS  Google Scholar 

  33. Tilman D, Clark M, Williams DR, Kimmel K, Polasky S, Packer C. Future threats to biodiversity and pathways to their prevention. Nature. 2017;546(7656):73.

    CAS  PubMed  Google Scholar 

  34. Hamadeh B, Chalak L, Coppens d’Eeckenbrugge G, Benoit L, Joly HI. Evolution of almond genetic diversity and farmer practices in Lebanon: Impacts of the diffusion of a graft-propagated cultivar in a traditional system based on seed-propagation. BMC Plant Biol. 2018;18(1):155.

    PubMed  PubMed Central  Google Scholar 

  35. Snoussi H, Duval M-F, Garcia-Lor A, Belfalah Z, Froelicher Y, Risterucci A-M, et al. Assessment of the genetic diversity of the Tunisian citrus rootstock germplasm. BMC Genet. 2012;13(1):16.

    CAS  PubMed  PubMed Central  Google Scholar 

  36. Křivánek M, Pyšek P, Jarošík V. Planting history and propagule pressure as predictors of invasion by woody species in a temperate region. Conserv Biol. 2006;20(5):1487–98.

    PubMed  Google Scholar 

  37. McGranahan G, Leslie C. Breeding walnuts (Juglansregia). In: Jain SM, Priyadarshan PM, editors. Breeding plantation tree crops: tropical species. New York: Springer; 2009. p. 249–73.

    Google Scholar 

  38. Mariette S, Tavaud M, Arunyawat U, Capdeville G, Millan M, Salin F. Population structure and genetic bottleneck in sweet cherry estimated with SSRs and the gametophytic self-incompatibility locus. BMC Genet. 2010;11:77.

    PubMed  PubMed Central  Google Scholar 

  39. Ding YM, Cao Y, Zhang WP, Chen J, Liu J, Li P, et al. Population-genomic analyses reveal bottlenecks and asymmetric introgression from Persian into iron walnut during domestication. Genome Biol. 2022;23(1):145.

    CAS  PubMed  PubMed Central  Google Scholar 

  40. Wang H, Pan G, Ma Q, Zhang J, Pei D. The genetic diversity and introgression of Juglansregia and Juglanssigillata in Tibet as revealed by SSR markers. Tree Genet Genomes. 2015;11(1):804.

    Google Scholar 

  41. Kuhn DN, Dillon N, Bally I, Groh A, Rahaman J, Warschefsky E, et al. Estimation of genetic diversity and relatedness in a mango germplasm collection using SNP markers and a simplified visual analysis method. Sci Hortic. 2019;252:156–68.

    Google Scholar 

  42. Hoban S, Archer FI, Bertola LD, Bragg JG, Breed MF, Bruford MW, et al. Global genetic diversity status and trends: towards a suite of Essential Biodiversity Variables (EBVs) for genetic composition. Biol Rev. 2022;97(4):1511–38.

    PubMed  Google Scholar 

  43. Allendorf FW, Funk WC, Aitken SN, Byrne M, Luikart G. Conservation and the genomics of populations. Oxford: Oxford University Press; 2022.

    Google Scholar 

  44. Schierup MH, Christiansen FB. Inbreeding depression and outbreeding depression in plants. Heredity. 1996;77(5):461–8.

    Google Scholar 

  45. Liu J, Gao LM. Comparative analysis of three different methods of total DNA extraction used in Taxus. Guihaia. 2011;31(2):244–9.

    CAS  Google Scholar 

  46. Doyle JJ, Doyle JL. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem Bull. 1987;19(1):11–5.

    Google Scholar 

  47. Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P. MICRO-CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Mol Ecol Notes. 2004;4(3):535–8.

    Google Scholar 

  48. Meirmans PG. genodive version 3.0: Easy-to-use software for the analysis of genetic data of diploids and polyploids. Mol Ecol Resour. 2020;20(4):1126–31.

    CAS  PubMed  PubMed Central  Google Scholar 

  49. Peakall R, Smouse PE. GenAlEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research-an update. Bioinformatics. 2012;28(19):2537–9.

    CAS  PubMed  PubMed Central  Google Scholar 

  50. Goudet J. hierfstat, a package for R to compute and test hierarchical F-statistics. Mol Ecol Notes. 2005;5(1):184–6.

    Google Scholar 

  51. Goudet J. FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3). 2001. http://www2.unil.ch/popgen/softwares/fstat.htm. Accessed 6 June 2023.

    Google Scholar 

  52. Luikart G, Allendorf FW, Cornuet J-M, Sherwin WB. Distortion of allele frequency distributions provides a test for recent population bottlenecks. J Hered. 1998;89(3):238–47.

    CAS  PubMed  Google Scholar 

  53. Rienzo AD, Peterson AC, Garza JC, Valdes AM, Slatkin M, Freimer NB. Mutational processes of simple-sequence repeat loci in human populations. Proc Natl Acad Sci USA. 1994;91(8):3166–70.

    PubMed  PubMed Central  Google Scholar 

  54. Chapuis MP, Estoup A. Microsatellite null alleles and estimation of population differentiation. Mol Biol Evol. 2007;24(3):621–31.

    CAS  PubMed  Google Scholar 

  55. Kolde R. Pheatmap: pretty heatmaps. R package version 1.0.12. 2019. https://CRAN.R-project.org/package=pheatmap. Accessed 6 June 2023.

    Google Scholar 

  56. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155(2):945–59.

    CAS  PubMed  PubMed Central  Google Scholar 

  57. Earl DA, vonHoldt BM. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv Genet Resour. 2012;4(2):359–61.

    Google Scholar 

  58. Smouse PE, Peakall R. Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity. 1999;82(5):561–73.

    PubMed  Google Scholar 

  59. Langella O. Populations version 1.2.31: population genetic software (individuals or populations distances, phylogenetic trees) CNRS UPR9034. 1999. http://bioinformatics.org/~tryphon/populations. Accessed 6 June 2023.

    Google Scholar 

Download references

Acknowledgements

Dr. Jie Liu is grateful to his father Xue-Wen Liu and younger brother Tao Liu for their insightful discussion of grafting techniques and great effort in field sampling. We thank Mr. Zuo-Ying Xiahou and Miss. Ying Chen for their help in the lab. Molecular experiments were performed at the Laboratory of Molecular Biology at the Germplasm Bank of Wild Species, Kunming Institute of Botany, Chinese Academy of Sciences.

Funding

This research was funded by the National Natural Science Foundation of China (32170398, 41971071, 31770367), Top-notch Young Talents Project of Yunnan Provincial “Ten Thousand Talents Program” (YNWR-QNBJ-2018–146), CAS “Light of West China” Program, Natural Science Foundation of Yunnan (202201AT070222), Key Research Program of Frontier Sciences, CAS (ZDBS-LY-7001), and Yunnan Fundamental Research Projects (202201BC070001). Richard Milne thanks the CAS President’s International Fellowship Initiative (2022VBA0004). Moses Wambulwa was supported by the Postdoctoral International Exchange Program of the Office of China Postdoctoral Council, the Postdoctoral Targeted Funding, and Postdoctoral Research Fund of Yunnan Province.

Author information

Author notes

  1. Jie Liu, Ephie A. Magige, and Peng-Zhen Fan contributed equally to this work.

Authors and Affiliations

  1. CAS Key Laboratory for Plant Diversity and Biogeography of East Asia, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan, 650201, China

    Jie Liu, Ephie A. Magige, Moses C. Wambulwa, Ya-Huang Luo & Lian-Ming Gao

  2. Germplasm of Bank of Wild Species, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan, 650201, China

    Jie Liu, Ephie A. Magige, Peng-Zhen Fan, Moses C. Wambulwa, Ya-Huang Luo & Lian-Ming Gao

  3. University of the Chinese Academy of Sciences, Beijing, 100049, China

    Ephie A. Magige & Peng-Zhen Fan

  4. Department of Life Sciences, School of Science and Computing, South Eastern Kenya University, Kitui, 170-90200, Kenya

    Moses C. Wambulwa

  5. Lijiang Forest Biodiversity National Observation and Research Station, Kunming Institute of Botany, Chinese Academy of Sciences, Lijiang, Yunnan, 674100, China

    Ya-Huang Luo & Lian-Ming Gao

  6. School of Ecology and Environmental Science, Yunnan University, Kunming, Yunnan, 650091, China

    Hai-Ling Qi

  7. Institute of Molecular Plant Sciences, School of Biological Sciences, University of Edinburgh, Edinburgh, UK

    Richard I. Milne

Authors
  1. Jie Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ephie A. Magige
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Peng-Zhen Fan
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Moses C. Wambulwa
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Ya-Huang Luo
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Hai-Ling Qi
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Lian-Ming Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Richard I. Milne
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

JL and LMG conceived and designed the study. JL and YHL carried out field work. EAM performed laboratory work. PZF, HLQ, and JL analyzed the data. JL and EAM wrote the first draft, with inputs from RM and MCW. All authors contributed to the revision and approved the final version of the manuscript.

Corresponding authors

Correspondence to Jie Liu or Lian-Ming Gao.

Ethics declarations

Ethics approval and consent to participate

This study has been approved by the government and carried out according to the laws of the People’s Republic of China. All participants had the requisite approvals by the Kunming Institute of Botany, Chinese Academy of Sciences. Permission to collect samples was obtained from the local forestry bureau. The sampling and laboratory methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Supplementary tables. Table S1.

Properties of 31 microsatellite loci used to characterize 298 walnut individuals. Table S2. Genetic diversity of the 31 microsatellite loci. Table S3. Population pairwise FST for both scion and rootstock populations. Table S4. The population and individual ID, voucher specimens and SSR genotyping data used in the current study.

Additional file 2: Supplementary figures. 

Fig. S1. Distribution of the pairwise number of allele differences among MLG (threshold=10): (a) Bimodal curve for scions, (b) Unimodal curve for rootstock genotypes. Fig. S2. Optimal K value of STRUCTURE selection graphs. (a) Delta (Δ)K for different numbers of subpopulations (K), maximum number of sub-populations were inferred at K=2 for STUCTURE analysis; (b) The average of log-likelihood value of K.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, J., Magige, E.A., Fan, PZ. et al. Genetic imprints of grafting in wild iron walnut populations in southwestern China. BMC Plant Biol 23, 423 (2023). https://doi.org/10.1186/s12870-023-04428-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12870-023-04428-z

Keywords

BMC Plant Biology

ISSN: 1471-2229

Contact us