Morphological, pathogenic and genetic diversity in Diplodia seriata associated with black rot canker of apple in India

Apple cankers are extremely destructive diseases threatening the global apple industry through direct and indirect losses. The population structure of the pathogens is of paramount significance for the development of efficient management strategies. Therefore, phenotypic, pathogenic, and genetic diversity of Diplodia seriata causing black rot canker of apple was investigated in this study. All the isolates were included for investigating the in vitro mycelial growth, conidial dimensions, and pathogenic variability on two-year-old potted apple seedlings. The ISSR approach was used to investigate the molecular diversity of D. seriata. Mycelial growth rates were found to vary significantly amongst the isolates; however, there were no major variations seen between the different geographical groupings of isolates. Pathogenicity tests revealed variations in the size of cankers among the isolates indicating the presence of virulence variability. The isolates were segregated into three virulence groups based on canker length. The Bayesian analyses of ISSR data divided the isolates into two genetic clusters. The genetic clustering of the isolates revealed no relationship with geographical origin of the isolates. Furthermore, no direct relationship of genetic clustering was observed with morphological or pathogenic variability. The ISSR primers revealed very high level of variability in D. seriata; however, no distinct populations of the pathogen existed which is an indication of high level of gene flow between the diverse geographical populations. According to our knowledge, this is the first thorough investigation on the diversity of D. seriata associated with apple black rot canker in India.

the grouping of isolates based on geographical origin as well as with virulence groups.The statistically significant differences in growth rates were not observed among the geographical groups of isolates (Table 1).

Pathogenic variability.
All the isolates of D. seriata were tested for pathogenicity on potted apple seedlings.Each of the isolates produced smoky canker symptoms characterized by water soaked, yellow to brown discoloration at the site of incision.The lesion showed upward and downward extension, girdling and rupturing of bark.Characteristic concentric ring development was also observed in the lesions (Fig. 2).During the investigation, significant differences in canker length between the isolates were observed in statistical analysis.All the isolates produced cankers considerably larger than control (Fig. 3) with mean lesion length varying from 7.08 cm (isolate 12) to 17.43 cm (isolate 41) and an overall mean lesion length of 10.35 cm.In each geographical group of isolates the mean lesion length was considerably longer than that of control but no considerable differences were observed amongst the groups (Table 1).The mycelial growth rates recorded under in vitro conditions and the length of canker lesions recorded during pathogenicity tests were not found to have any relationships.The overall average positive re-isolation from the inoculated twigs was 2.94 out of 3 twigs, i.e., 98% re-isolation frequency; thus, proving that all the isolates were pathogenic in nature.
The isolates exhibited a considerable variability in incubation periods during pathogenicity tests.Based on incubation period the isolate were divided into three groups.The first group comprised of isolates for which the mean incubation period was 31.93 ± 2.60 days to develop canker symptoms on the inoculated plants.A total of 37.50% isolates belonged to this group.The second group comprised of 35.41% isolates which exhibited an incubation period of 44.59 ± 2.61 days, while as the third group comprising of 27.08% isolates was found to show the longest incubation period of 59.79 ± 2.49 days (Table 3).
The isolates also varied in their virulence demonstration during pathogenicity trials.Based on average lesion size, the isolates were divided into three virulence groups viz.Group I-least virulent which produced a canker lesion of < 100 mm length; Group II-moderately virulent which produced a lesion size of 100-150 mm length and Group III-highly virulent which produced a canker lesion of > 150 mm length (Table 4).Cankers show more extension longitudinally than radial expansion, for that reason canker length was used as a standard for comparing canker sizes.Group I accommodated 47.91 per cent of isolates (23 isolates) and was categorized as least virulent group.Group II accommodated 41.66 per cent isolates (20 isolates) and was considered as moderately virulent.Group III accommodated only 10.41 per cent of isolates (5 isolates) and was considered as highly virulent group (Table 4).

Genetic diversity.
Initially 17 ISSR primers were screened for standardization using 10 randomly selected isolates of the pathogen.Among these 12 ISSR primers failed to produce polymorphic, reproducible, and scorable bands and hence did not yield any information.The remaining 5 primers resulted in polymorphic and reproducible bands, hence, were adopted for studying genetic diversity between the isolates of D. seriata.DNA fingerprinting of D. seriata isolates derived from diverse geographical locations clearly showed distinct banding pattern.The amplification products produced 65 bands in total ranging from 300 to 2000 bp, out of which 63 (96.92%) were polymorphic (Table 5).www.nature.com/scientificreports/Under similar experimental conditions, all the primers amplified the DNA from ten randomly chosen isolates while testing for reproducibility of the results.Out of 50 total combinations (10 isolates × 5 primers) band patterns identical to band arrangements in ISSR characterization were detected in 21 and the proportion of coincident bands exceeding 90% were observed in 35 combinations.Among all the combinations the coincident bands were estimated to be present with an overall percentage of 94%.

Genetic variation analysis.
Four distinct clustering techniques were used to determine the most probable number of communities that the samples obtained would represent, assign individuals to these populations, and estimate the degrees of variance.Except for one isolate, i.e.DS38, which split as an independent lineage, the cluster analysis based on the unweighted Neighbor-joining approach yielded a dendrogram that grouped  www.nature.com/scientificreports/all the isolates representing four geographical populations into two primary clusters.The cluster first comprised of 26 isolates while as cluster second comprised of 21 isolates.The assignment of isolates into these two clusters was independent of their geographical background as both the clusters contained individuals from all the four geographical populations.Both the major clusters were further divided into smaller sub-clusters which were also independent of the geographical origin of the isolates.This demonstrates that genetic diversity exists both inside and among population isolates (Fig. 4).The average number of observed alleles (Na) and effective alleles (Ne) across all the loci varied between 1.385 and 1.662 and 1.345 and 1.393, respectively, across the four distinct regional populations (PL, SH, BR, and GN).The highest number of observed alleles (1.662) were found in the population SH belonging to district Shopian of the valley, while as lowest number of observed alleles (1.345) were found in the population GN belonging to district Ganderbal, with mean value of 1.504.Again, the highest effective allele number (1.393) was found in the population SH while as the lowest number of effective alleles (1.345) were found in the population PL belonging to district Pulwama, with an average of 1.372.Similarly, the values of gene diversity (h) and unbiased gene diversity (uh) ranged from 0.218 to 0.242 and 0.238 to 0.264, respectively.The highest values of h (0.242) were recorded for the population SH pertaining to geographic locations of district Shopian and lowest values of h (0.218) were recorded for the population PL pertaining to geographic locations of district Pulwama, with an average of 0.228.Similarly, the highest values of uh (0.264) were recorded for the population SH belonging to geographic locations of district Shopian and lowest values (0.238) were recorded for the population PL, with an average of 0.248.Based on the Shannon's information index (I), the highest diversity was recorded in the population SH (I = 0.377) and lowest diversity was recorded in the population PL and GN (I = 0.341), with an average of 0.350.The percentage of polymorphic loci among the populations ranged from 66.15% in GN to 81.54% in SH with an average of 72.31% (Table 6).
The average genetic distance and identity of Nei among D. seriata populations was 0.018-0.023and 0.977-0.982,respectively (Table 7).The BR population of Baramullah, the SH population of Shopian, and the GN population of Ganderbal were found to have the highest genetic differences.The SH populations of Shopian and the PL populations of Pulwama were found to have the least genetic distance.Likewise, the SH population of Shopian and the PL population of Pulwama demonstrated the highest levels of genetic similarity.Ganderbal's  GN and Baramulla's BR populations had the fewest genetic characteristics.The findings of Nei's genetic distance analysis showed that groups that were geographically close to one another had the lowest genetic distances, whereas populations that originated in various locations had the greatest genetic distances.The geographically defined populations had considerable genetic diversity, according to the analysis of molecular variance (AMOVA).Although just 18% of the overall genetic diversity was found across populations from various geographical locations, most of the genetic variation (82%) was found within populations (among individual isolates) (Table 8; Fig. 5).
It was noted that there were statistically significant differences between the populations (P = 0.01).Pairwise PhiPT values estimated the degree of genetic divergence between the four populations.The average degree of genetic similarity was calculated to be 0.178 (average PhiPT value) (Table 9).The populations PL and GN showed the largest genetic difference (0.230), whereas the populations GN and BR showed the lowest genetic differentiation (0.076) (Table 9).The pairwise PhiPT values between the populations of South Kashmir (SH and PL; PhiPT = 0.103) were much lower than those between the populations of South and North Kashmir (PL-BR, SH-GN, and SH-BR), indicating that there was less genetic difference between these groups.The Mantel test confirmed any potential association between the geographic and genetic distances between the corresponding www.nature.com/scientificreports/groups.The test's results revealed no connection between genetic and geographic distances (r = 0.591; p = 0.210) (Fig. 6).
The analysis performed with the help of structure software divided the 48 D. seriata isolates into two genetically distinct clusters.The log likelihood values (Ln P(D)) increased from K = 1 to K = 2 and was maximum at K = 2 (ln P(D) = 2145•68).The standard deviation showed a slow increase from K = 1 to K = 2 and thereafter it showed a rapid increase from K = 3 to K = 10.This showed that the population structure of the D. seriata pathogen population is best explained by K = 2 clusters.This is further strengthened by the value of ∆K which was highest at K = 2 (∆K = 101•39) and after that it decreased at a very fast rate.The highest value of ∆K that determines the best fit of K was derived by employing STRU CTU RE HARVESTER (Fig. 7).
The clusters formed in the structure analysis are shown in the figure below.Most of the individual isolates were find out to possess 80% of their heredity from one of the two distinct populations (Fig. 8), while as 17 isolates were found to have mixed ancestry with less than 80% posterior probability to any of the two clusters.The cluster in red consisted of 14 D. seriata isolates while as the cluster in green consisted of 17 isolates.The distribution of isolates into different clusters was irrespective of their geographical origin, hence both the clusters contained isolates from all the four predetermined geographical populations.
The principal coordinate analysis (PCoA) was carried out with Genalex software on the whole data set to verify the consistency of genetic differentiation clustering performed by different clustering methods.The two-dimensional PCoA plot revealed that 15.45% and 9.20% of the total genetic variation is represented by the coordinate 1 and coordinate 2, respectively.It is evident from the PCoA plot that except two individuals as outliers and two small groups were formed, and all the other individuals formed a common group, however the individuals in all the groups were intermixed (Fig. 9).
This shows that the PCoA clustering was independent of the geographical origin of the isolates which is in accordance with other clustering analysis methodologies.The heatmap constructed with R software also divided the population of all the isolates into two major clusters which are further subdivided into sub-clusters.It is clear from the heatmap that the genetic variation is spread among the major clusters and sub-clusters irrespective of the geographical origin of the individual isolates (Fig. 10).

Discussion
In the present investigation, the phenotypic traits of the fungal isolates collected from apple black rot canker were in complete accord with those of the genus Dilpodia 43 .Depending on the color and septation of their conidia, the species of Diplodia may be roughly classified into two distinct groups based on morphological descriptions 44 .
Most Diplodia species, including D. malorum, D. mutila, D. corticola, D. africana, and D. rosulata, have aseptate, thick-walled conidia that may stay hyaline for a long time before becoming brown 43,[45][46][47][48] .Nevertheless, in other species, including D. alatafructa, D. pinea, D. scrobiculata, D. seriata, and D. pseudoseriata, the conidia become colored before discharge from the pycnidia and mostly stay non-septate 43,44 .All of the isolates morphological characteristics in this investigation were consistent with those previously described for D. seriata 1,13 .Additionally, no significant differences were identified across regional or genetic groups.Each of the isolates produced characteristic black rot canker symptoms on potted apple trees and the isolates were recovered from 98% of the inoculated apple twigs, hence proving the isolates to be potential pathogens of apple.Botryosphaeriaceae fungi being cosmopolitan in nature 13 have been reported to be common pathogens of a vast diversity of woody hosts including apple [49][50][51] .Pathogenicity of D. seriata has been evaluated on shoots, fruits, and leaves of apple in some countries 23,24 .The experimental data on the pathogenic nature of Diplodia spp. on woody plants is contentious.For instance, D. seriata is regarded as a secondary and weak pathogen of apples in England and New Zealand 27 , while it is reported as highly pathogenic on USA apples where it is responsible for apple cankers, leaf spots and fruit rot 14,28,29 .Similarly, D. seriata, being highly virulent on grapevines has been considered as a primary and virulent pathogen 31,33 but other workers have observed it to be saprophyte or weak pathogen on grapevines 6,36,43 .These contradictions in pathogenicity may be due to differences in virulence among the isolates.However, in the present investigation all the isolates of D. seriata were found to be highly pathogenic to apple as reflected by the length of lesions produced on the inoculated apple seedlings which were much larger than the lesions in control.
A significant variation was observed in isolates of D. seriata with respect to lesion length on artificially inoculated apple trees of cultivar Red Delicious.The variation in the length of canker lesions caused by different isolates under pathogenicity tests confirms the existence of virulence variability among the isolates of D. seriata.The isolates were categorized into 3 virulence groups depending on average canker length.The isolates showed no relevance between the virulence and geographical origin of isolates which was later supported by genetic grouping as well.Based on length of lesions produced in pathogenicity tests on grapevine canes, the D. seriata isolates were divided into four virulence groups 31 .Variation in D. seriata isolates with reference to wood colonization on grapevines has also been reported 6 .Similar findings on the variability in of D. seriata have been recorded by other workers as well 30,32 .While discussing the contrasting reports regarding the pathogenicity of D. seriata, Savocchia et al. 33 concluded that these differences may be because of the presence of intraspecific variations within this fungus.Our findings are well supported by the observations of Elena et al. 12 who reported "no relationship between the genetic groups and the virulence of isolates of D. seriata on grapevine".Similar conclusions were drawn by other workers e.g., Billones-Baaijens et al. 41 and Baskarathevan et al. 40 who observed no relationship between genotypic and pathogenic variability in the isolates Neofusicoccum luteum and N. parvum, respectively.Again, no correlation was found to exist between the radial mycelial growth and the virulence variability of isolates because the isolates which showed maximum radial mycelial growth did not always produce longest cankers as it should have been.This could be attributed to the reasons like production of phytotoxic secondary metabolites by D. seriata as reported by Martos et al. 52 , or presence of some other mechanism besides fungus mycelium growth inside the host which may be influencing the virulence of the pathogen 12 .
The ISSR technique proved effective for ascertaining the genetic diversity between the isolates of D. seriata having diverse geographical origins.Five ISSR primers recorded a high degree of polymorphism in the isolates of D. seriata, hence proving that ISSR markers are very relevant in verifying the genetic diversity in this fungus.The ISSR markers have been found relevant for investigating the genetic variability in D. seriata and other fungal plant pathogens by various other workers 12,54,[56][57][58][59] .
For investigating the population structure of D. seriata isolates Bayesian clustering of ISSR markers proved very useful and provided the basis to investigate the genetic structure.The analyses performed with STRU CTU RE divided the isolates into two main clusters however; it was observed that some of the isolates did not constantly belong to these clusters.Further, the isolates originating from diverse geographical regions did not fall in different genetic groups, indicating that the isolates belong to a common ancestral background irrespective of their geographical origin.Elena et al. 12 , reported high genetic diversity in D. seriata on grapevine from Spain using ISSR markers and recorded that there was no relation between the clustering groups and the geographic origin of isolates.While performing ITS sequencing of D. seriata isolates Phillips et al. 13 concluded that no correlation existed between the clustering structure and the geographical origin of the isolates.
Clustering of D. seriata isolates at regional level revealed that most of the variability was found within groups than between the groups.This indicates that the genetic diversity between the geographic populations is less but within populations diversity is high.Padder et al. 56 reported high genetic diversity in Venturia inaequalis populations on apple in Kashmir valley using ISSR markers and found greater variability within populations than among populations.Dinolfo et al. 57 studied intraspecific diversity within the isolates Fusarium poae derived from Argentina and England but there was no definite relation between genetic diversity and geographic origin, and diversity was high within rather than between the Argentinean and English populations.Fan et al. 55 analyzed genetic variation in Chinese and Californian isolates of Monilinia fructicola using ISSR markers and found that the diversity was high within population rather than between populations.Many other researchers have also reported high genetic diversity in fungal plant pathogens using ISSR markers 58 .Although only two clusters were derived in the STRU CTU RE analysis of the data generated from the 48 isolates of D. seriata, the ISSR technique revealed a high level of polymorphism (96%).Similar results were recorded by Elena et al. 12 who reported 88% polymorphism in 83 D. seriata isolates from Spain.Employing AFLP markers Martin 38 recorded 86% polymorphism in the isolates of D. seriata derived from Spain.With UP-PCR technique Billones-Baaijens et al. 41 recorded 93% polymorphism in the isolates of N. luteum.Using the same technique, Baskarathevan et al. 40 also recorded 93% polymorphism in 60 isolates of N. parvum.Many factors such as gene flow, recombination (sexual or parasexual), mutation and adaptation of the pathogen to diverse host species may be responsible for the presence of high genetic diversity in D. seriata.Although Hesler recorded the formation of perfect state (Botryosphaira obtusa) for the smoky canker pathogen on apple twigs but other researchers from India and other countries failed to report any sexual state in field 22 .Thus, further investigations are required to be undertaken both under natural field conditions and at genetic levels to verify whether some hidden telomorphic state is responsible for the high genetic diversity in this fungal species.Estimates of pairwise F ST were low in geographical populations.This indicates that there is frequent gene flow between populations probably due to movement of infected planting material.

Conclusions
The morphological differences shown by the isolates in phenotypic characterization did not show any correlation with the genetic clusters.Moreover, the pathogenicity tests revealed D. seriata as highly pathogenic on apple and indicated variability in virulence, but the pathogenic variability was not correlated to either the genetic variability or the geographical origin of the isolates.To the best of our knowledge, this combined phenotypic, pathogenic, and genetic variability of D. seriata isolates causing black rot canker, dieback, and fruit rot of apple in India is studied for the first time and it will help in understanding the pathogenic nature of the D. seriata populations and development of more effective management strategies in future.

Materials and methods
Fungal isolates.The disease samples collected from different parts of the apple trees during the years 2016 and 2017 were cleaned of by using distilled water.The bark at the diseased portions was carefully removed and the area was surface sterilized with 70% ethanol.Wood sections (5 mm 2 ) were taken along the margins of canker lesions by using sterilized blade and the tissue bits were sterilized for 60 s in 1% sodium hypochlorite solution.
The bits were washed thrice with sterilized distilled water to eliminate the remains of chemical followed by drying on blotting paper and inoculated on acidified PDA.The inoculated plates were placed under incubation at 25 ± 1 °C in the darkness.All the isolates of D. seriata collected from different geographical areas were purified by single spore technique 60 .D. seriata was identified based on morphological characters as per the descriptions of fungus given by Phillips 13 .In all, 48 isolates (Table 10) of D. seriata having different geographical origins were maintained for further studies and the stock cultures were preserved under refrigeration at 4 °C inside the refrigerator.
Pathogenicity tests.Pathogenic variability was studied by performing pathogenicity tests of all the 48 isolates on healthy two-year-old potted apple seedlings (cv.Red Delicious grafted on Quince root stock) following the procedure adopted by Spiers 61 .The selected plants were first treated with copper oxychloride 50 WP (Blitox 50) @ 0.3% as a spray to eliminate the existence of any pathogens which may be present on the host surface.The pots were kept inside a polyhouse and timely irrigated to prevent the young apple seedlings from water stress and intermittently sprayed with distilled sterilized water to maintain high humidity inside the chambers.The inoculations were done on dormant apple twigs.Each isolate was inoculated on three twigs and the tests were performed as per the following procedure.
On the selected twigs, after surface sterilizing the designated area with 70% ethanol, a 5 mm horizontal cut was given cutting the bark of the twig.Starting from the mid-point of the first cut, another vertical cut of 5 mm was made.The bark of this T-shaped cut was loosened with the help of a sterilized scalpel.A mycelial disc (4 mm) was put beneath the bark of incision and a cotton swab moistened with sterilized water was placed on the inoculated cut followed by wrapping with adhesive paper tape.The control was inoculated with plain media (PDA) disc and the incisions were covered as described earlier.The pathogenicity tests were regularly tracked for canker formation.Observations in terms of symptom development and lesion size by different isolates were recorded.The pathogen was re-isolated from the inoculated twigs to fulfill Koch's postulates.For re-isolations 1 cm pieces of wood after removing bark were cut at the two ends of canker lesions and sterilized in ethanol (70%) for 4 min, blotted dry and inoculated on Petri plates containing acidified PDA.The plates were kept under incubation at 25 ± 1 °C and observed for morphological characters of D. seriata.

Phenotypic characterization.
Radial mycelial growth.All the 48 isolates of D. seriata from the four geographical locations were included in this study.Each Petri plate containing PDA was inoculated in the center with 5 mm culture bit of the actively growing isolate of D. seriata followed by incubation of the plates at 25 ± 1 °C.Each isolate was grown on four PDA plates.The observations on radial mycelial growth were recorded after 72 h by taking average of four radial measurements perpendicular to each other.The test was conducted thrice for the purpose of accuracy.
Conidial size and shape.Conidial characteristics such as shape, size, color and septation of all the 48 isolates were assessed on oatmeal agar medium.The cultures were kept on a bench in the laboratory for 4 weeks at about 20-25 °C for inducing pycnidia formation under diffused sunlight.The spore measurements of each isolate were recorded under light microscope [HF (bright field, filter 1); Carl Zeiss] at 100× magnification by preparing semipermanent slides in lactophenol under 10 random microscopic fields after calibrating the microscope following standard procedures 64 .For each isolate the conidial dimensions (length and width) of 50 conidia were recorded, and the length: width ratio was estimated.
Genomic DNA extraction and PCR amplifications.For raising enough mycelium for DNA extraction, all the isolates of D. seriata were grown on PDA broth in Erlenmeyer flasks.Mycelium was harvested by filtering the contents of the flask through a double layered filter paper, dried and stored at − 80 °C.Adopting modified CTAB (Cetyl trimethyl ammonium bromide) method 62 , total genomic DNA was extracted from all the isolates.The DNA was kept at 4° C for amplification and stored at − 20° C for future use.
The reaction mixture in PCR tubes were given short spin or vortexed in microfuge and placed in 96-well thermal cycler.PCR amplifications were conducted in thermal cycler programmed for initial step at 94 °C for five minutes followed by 35 denaturation cycles at 94 °C for one minute, annealing at appropriate temperature (Table 11) for one minute, elongation at 72 °C for two minutes and a final step of extension for 10 min.To 25 μl of the amplified PCR product, 5 μl of 6X loading dye was added to have final concentration of 1X.
The quality of PCR products was evaluated on 2% agarose gel.TAE (0.5X) was used to prepare the gel and ethidium bromide at 0.5 μg/μl was added to it.The gel was loaded with amplified products of all the 48 isolates of D. seriata and run at 5 V/cm and visualized under UV light.The photograph of resultant banding pattern ISSR characterization and analysis of genetic variability.The banding pattern of each primer was scored as 1 (band present) and 0 (band absent) and the pattern was converted into a binary matrix with the help of MS Excel.The binary data matrix was used to work out TNB (total number of bands), NPB (number of polymorphic bands), and PPB (percentage of polymorphic bands).The Jaccard's coefficient of similarity was calculated in the PAST (v 4.0) software by making pairwise comparisons between the individual isolates.Furthermore, a dendrogram was constructed with the help of same software by adopting the UPGMA method.The cophenetic correlation coefficient was worked out by the PAST program for measuring the goodness of fit between the cophenetic matrix and the distance matrix 69 .For each marker the alleles were calculated in all the isolates by numbering the alleles as 1, 2, 3 and so on, starting from the allele possessing maximum molecular weight.The bands were visually scored as 0 (absent) and 1 (present).The binary data generated by ISSR primers was analyzed by cluster analysis utilizing the program NTSYS-pc version 2.02 63 .Based on shared bands among the isolates, similarity was estimated from the data generated by all the 5 ISSR primers using Jaccard's coefficient.The NTSYS software was utilized for determining the Nei and Li 65 coefficients to ascertain the similarity between the isolates of D. seriata.The investigation of genetic diversity among the isolates of D. seriata was done by analysis of the band arrangements using a Bayesian clustering approach executed in STRU CTU RE v.2.2.3 67 considering a population admixture model wherein within population correlations of allele frequencies are worked out.The runs for each K value (1 to 10) were set for 20 independent replications having a burn-in period of 100,000 accompanied by 500,000 iterations of the Markov Chain Monte Carlo (MCMC).The posterior probability of the data (ln P(D)) was estimated utilizing the ad hoc ΔK statistic, based on rate of change between successive K values, and this estimate was later employed for inferring the uppermost level of structure in the dataset 67 .The highest values of Ln P(D), which represents the posterior probability of the data for a certain K, and ΔK were used to determine K (the rate of change in the log probability of data between successive values of K) 76 .The sample populations were divided into genetic groupings using bar graphs that represent individual assignment probability.By using Structure Harvester v6.0 to generate the K values, the best-fit population cluster number (K) was verified 76 .Employing ARLEQUIN v. 3.5 68 , AMOVA was used to find out the differences in assigning the isolates to different clusters which were obtained after Bayesian analyses.Further, the pairwise population differentiation (PhiPT) was also worked out by utilizing Genalex 6.503 package [69][70][71] .Microsatellite data set were analyzed for basic genetic diversity indices such as, N: Sample size; Ppl: percentage of polymorphic loci; Na: observed number of alleles; Ne: effective number of alleles; I: Shannon's information index; h: diversity; uh: Unbiased diversity using population genetic software GenALex 6.503 71 .
The principal coordinate analysis (PCoA) was conducted by employing Genalex software to generate a coordinate graph of the most significant axes to verify the clustering pattern.Distribution of the genetic variation into two components viz.within populations and among the populations was calculated by performing the analysis of molecular variance (AMOVA) in Genalex at 999 permutations.

Statistical analysis.
To estimate the population genetic structure and examine the distribution of genetic diversity, the isolates of D. seriata were primarily grouped into populations based on geographical origin.The pathogenicity experiments were conducted following randomized complete block design under poly house conditions.Groups based on geographic origin and those generated by the STRU CTU RE analysis were considered as independent variables.ANOVA (One-way) and posterior Tukey's Honestly Significant Difference tests were performed on dependent variables to find out the differences between the mean values.The molecular data generated during the study was analyzed with NTSYS-2.02e;Arlequin v. 3.5; GenAlex 6.503; PAST 4.0 and STRU CTU RE v. 2.2.3 as described earlier.The remaining data was analyzed by SPSS (ver.18) and SAS (ver.9.2) software.

Figure 1 .
Figure 1.Relative radial mycelial growth of the isolates of D. seriata on PDA.

Figure 4 .
Figure 4. UPGMA based dendrogram showing genetic diversity among D. seriata isolates based on Jaccard's coefficient of similarity using ISSR primers.

Figure 8 .
Figure 8.Estimated genetic structure for K = 2 groups of 48 Diplodia seriata isolates inferred by Bayesian clustering of ISSR data in 20 independent runs using the software STRU CTU RE v. 2.2.3.Each isolate is represented by a bicolored column according to its posterior probability.

Figure 9 .
Figure 9. Principal coordinate analysis of four populations of D. seriata based on ISSR molecular marker analysis.

Figure 10 .
Figure 10.Heat Map of isolates.The 48 fungal isolates representing different geographical regions were used to construct a heatmap.Heatmap shows that samples cluster irrespective of the geographical origin.Clustering demonstrated that all the isolates grouped into two main clusters which is consistent with the clustering by STRU CTU RE. https://doi.org/10.1038/s41598-023-42534-ywww.nature.com/scientificreports/

Table 3 .
Grouping of isolates based on incubation period.*Mean ± SD.

Table 4 .
Grouping based on virulence.

Table 5 .
Annealing temperatures and output results for five ISSR primers used in the genetic diversity study of 48 isolates of D. seriata.PBP (%) = polymorphic bands percentage.

Table 6 .
Genetic diversity of four populations of D. seriata based on ISSR profiling.N Sample size, Ppl percentage of polymorphic loci, Na observed number of alleles, Ne effective number of alleles, I Shannon's information index, h diversity, uh Unbiased diversity.*The values after ± denote the standard errors.

Table 7 .
Nei's genetic identity and genetic distance of five populations based on ISSR.Genetic distance (below diagonal) and Nei's genetic identity (above diagonal).

Table 8 .
Analysis of molecular variance (AMOVA) for testing homogeneity of genotype distribution in the D. seriata populations.df degree of freedom, SS sum of squared observations, MS mean of squared observations, PhiPT proportion of the total genetic variance that is due to the variance among individuals within a variant.

Table 9 .
Pairwise population PhiPT values based on 999 permutations from AMOVA analysis among four populations.

Table 10 .
Isolate No., geographical location, genetic grouping (structure) and other details of D. seriata isolates obtained from the host Malus × domestica.Numbers with an asterisk are indicative of isolates with < 80% assignment probability to different groups.Other studies: 1 represents the isolates selected for verifying the reproducibility of ISSR technique.NS not selected.was taken using Alfa Imager gel documentation system.The size of ISSR alleles was determined by their relative positions with respect to DNA ladder (100 base pair).The amplification reactions of all the primers were repeated thrice for a single DNA extraction of each of the isolates.Further, for band scoring only clear and reproducible bands were taken into consideration.