Peonies of the genus Paeonia are among the most iconic ornamental plants in the world, and Iran sits within an important center of diversity for the genus. The Iranian flora includes taxa such as Paeonia wendelboi, Paeonia tomentosa, Paeonia wittmanniana, Paeonia archibaldii and field-assigned representatives of Paeonia mascula, many of which grow in fragmented mountain habitats and face pressure from habitat loss and uncontrolled collection. Understanding how genetic variation is distributed across these plants is not merely an academic exercise. It determines which populations conservationists should prioritize, which landraces and wild relatives breeders can draw upon, and whether the current taxonomic labels applied in the field actually correspond to coherent biological lineages.

The technical challenge the researchers faced is one that haunts much of polyploid plant genetics. Many peonies carry more than two copies of each chromosome, which means that at any given microsatellite locus a single individual may carry several different alleles, and conventional genotyping methods cannot always determine exactly how many copies of each allele are present. Rather than guessing at this so-called allele dosage, the team adopted a dosage-agnostic, polyploid-aware framework that treats alleles simply as present or absent in each genotype. This approach sacrifices some statistical power but avoids building entire analyses on unreliable dosage calls, making the results far more reproducible for organisms whose ploidy levels are uncertain or variable.

In total, seventy-two genotypes representing the five labeled taxa were screened at nine polymorphic simple sequence repeat loci. The analysis of molecular variance revealed a striking split: 31.35 percent of the detected SSR variation was attributable to differences among the labeled taxa, with a differentiation statistic of 0.3135 that was highly significant, while the remaining 68.65 percent resided within taxa. In other words, most of the genetic diversity of Iranian peonies is packed inside each taxon rather than separating one species cleanly from another. That pattern is a double-edged sword. It signals that individual populations harbor a wealth of unique variation worth protecting, but it also means that species boundaries drawn from morphology alone may not map neatly onto genetic reality.

Ordination analyses sharpened that message. The first two principal coordinate axes accounted for 38.93 percent of the reported variation and cleanly separated Paeonia wendelboi from the closely entangled Paeonia tomentosa and Paeonia wittmanniana complex. A discriminant analysis of principal components, a method that finds the axes best separating predefined groups, assigned 66 of the 72 genotypes, or 91.7 percent, to the label expected from their field identification. Crucially, every single discordant assignment involved Paeonia tomentosa or Paeonia wittmanniana, suggesting that these two taxa are genetically intertwined to a degree that challenges their current treatment as fully separate species. When the researchers used a threshold-based approach to define multilocus genotype groups, they recovered forty-one distinct groups across the sampled plants, underscoring how finely structured the variation is even within a modest number of loci.

The diversity estimates themselves told a conservation-relevant story. Paeonia wendelboi showed comparatively low genetic diversity, with a Simpson’s index of 0.785 and only eight multilocus genotype groups detected among thirty-two samples, whereas Paeonia tomentosa was far more variable, with a Simpson’s index of 0.941 and nineteen groups among just twenty-four samples. Low diversity in Paeonia wendelboi could reflect historical bottlenecks, restricted gene flow, or sampling of a genetically narrow species, and it flags this taxon as potentially vulnerable. Populations with reduced genetic diversity have less evolutionary flexibility in the face of climate change, disease, and habitat fragmentation, making Paeonia wendelboi a natural priority for conservation sampling and ex situ collection.

To connect genes with form, the team descriptively integrated their SSR data with a previously generated dataset of locality-level means for fifteen morphological traits. A principal component analysis of the morphology found that the first component alone explained 68.02 percent of the sample-weighted variation, indicating that a single dominant axis of shape and size differences, likely reflecting overall vegetative and floral architecture, distinguishes the sampled localities. An exploratory matrix correlation between the genetic and morphological datasets yielded a coefficient of 0.510, a moderate association suggesting that morphology does carry a genuine genetic signal in these peonies, but that a substantial fraction of the variation in traits is shaped by factors other than the neutral marker variation captured by the microsatellites. Mean Sørensen dissimilarity across three pairwise taxon comparisons was 0.535, with the turnover component of dissimilarity exceeding the nestedness-resultant component, meaning that populations differ mainly by replacing alleles and genotypes rather than by simply losing subsets of a shared pool.

Perhaps the most scientifically honest part of the study lies in its sensitivity analyses and stated limitations. When the authors repeated their analyses after globally excluding two loci, Loc6 and Loc9, the broad genetic separation of Paeonia wendelboi persisted robustly in the seven-locus dataset. The fine-scale separation between Paeonia tomentosa and Paeonia wittmanniana, however, proved sensitive to the marker set used, dissolving in some sensitivity runs. Meanwhile, inference for Paeonia archibaldii and the field-assigned Paeonia mascula accession was explicitly limited by small sample sizes. The authors are careful not to overclaim: their results provide a reproducible baseline, not a final taxonomic verdict, and they call for expanded cytogenetic work to pin down ploidy levels and genome-wide studies to resolve species boundaries with far greater resolution than nine microsatellite loci can deliver.

Why should a study of Iranian peonies capture wider attention? Because it exemplifies a methodological shift sweeping through plant conservation genetics. Polyploid crops and wild relatives, from bananas and potatoes to many ornamentals, have long resisted standard population genetic toolkits, and dosage-agnostic frameworks like the one deployed here offer a way to extract reliable signal without costly whole-genome sequencing. The finding that nearly seventy percent of variation lies within taxa also carries a universal lesson for conservation planning: protecting a single reserve per species is not enough. Multiple genetically distinct populations must be safeguarded to capture the full breadth of variation, and the forty-one multilocus genotype groups identified here provide a concrete, data-driven shopping list for seed collectors and gene bank curators.

The study, supported by Tarbiat Modares University and the Iran National Science Foundation under project number 4021138, and conducted in collaboration with Spanish researchers at IHSM La Mayora, ultimately reads as both a milestone and an invitation. It establishes the first rigorous multilocus baseline for Iranian Paeonia, demonstrates that Paeonia wendelboi is a genuinely distinct and genetically depauperate lineage deserving urgent attention, and exposes the Paeonia tomentosa and Paeonia wittmanniana complex as a fascinating evolutionary puzzle in which morphology, genetics and taxonomy only partially align. As genome-wide tools become accessible for non-model ornamentals, the mountains of Iran may yet yield a fully resolved peony family tree, and this study has laid the essential first stones of that foundation.

Subject of Research: Genetic diversity and population structure of Iranian Paeonia taxa assessed with SSR markers and morphological traits

Article Title: Genetic diversity and population structure of five Iranian Paeonia taxa revealed by SSR markers and morphological traits

Article References: Fazli, M., Ahmadi, N., Hormaza, J. I., & Larranaga, N. (2026). Genetic diversity and population structure of five Iranian Paeonia taxa revealed by SSR markers and morphological traits. BMC Plant Biology. https://doi.org/10.1186/s12870-026-10055-1

Image Credits: AI Generated

DOI: 10.1186/s12870-026-10055-1

Keywords: Paeonia, genetic diversity, SSR markers, polyploidy, population structure, multilocus genotypes, Iran, plant conservation, microsatellites, morphological traits, DAPC, BMC Plant Biology

The study, led by Bhawanpreet Kaur and C. S. Mukhopadhyay of Guru Angad Dev Veterinary and Animal Sciences University in Ludhiana, together with H. M. Yathish of Karnataka Veterinary, Animal and Fisheries Sciences University, focused on six breeds: three indigenous ones, the Gaddi, the Tibetan Mastiff and the Mudhol Hound, and three exotic ones, the Labrador Retriever, the German Shepherd and the Pug. Fifty dogs were sampled from four states spanning northern and southern India: Punjab, Himachal Pradesh, Haryana and Karnataka. Peripheral blood samples were collected aseptically under ethical approval from the Institutional Animal Ethics Committee, and genomic DNA was extracted using the classical phenol-chloroform-isoamyl alcohol method before being quantified and quality-checked with spectrophotometry and gel electrophoresis.

The technical heart of the study is a method called double digest restriction-site associated DNA genotyping by sequencing, or ddRAD-GBS. In this technique, genomic DNA is cut with a pair of restriction enzymes, in this case SphI and MluI, derived from the bacteria Streptomyces phaeochromogenes and Micrococcus luteus. The resulting fragments, which represent a reproducible subset of the genome, are then sequenced on an Illumina platform using 150-base-pair paired-end reads. This reduced-representation strategy allows researchers to interrogate tens of thousands of genomic positions across many individuals at a fraction of the cost of whole-genome sequencing, making it a popular choice for population genetics in non-model organisms. The sequencing and genotyping work was carried out at AgriGenome Laboratories in Kerala and at the Animal Genomics Laboratory at GADVASU.

The raw sequencing output was subjected to rigorous quality control using the FASTQC tool, with the researchers examining per-base quality scores, sequence length distributions, GC content and duplication levels. The Q30 score, a measure of the proportion of bases sequenced with high confidence, ranged from 91.272 to 95.262 percent across samples, while GC content fell between 49.5 and 55 percent, close to the theoretical expectation for canine DNA. After filtering out low-quality genotypes, multi-nucleotide polymorphisms and likely paralogous variants, the team identified a staggering 356,461 SNP loci across the samples, of which 75,811 high-quality SNPs were retained for downstream population genetic analyses. The computational pipeline relied on R programming and Linux bash scripting, with the data converted into a genlight object for analysis using packages such as vcfR, poppr, ape and adegenet.

To untangle the relationships among the dogs, the researchers deployed a battery of complementary analyses. They constructed a phylogenetic distance tree from pairwise genetic distances, supported by 100 bootstrap replicates, and built minimum spanning networks in which each dog is a node connected to its closest genetic relatives. They also performed principal component analysis, which compresses genome-wide SNP variation into a few independent axes, and discriminant analysis of principal components, a multivariate method that maximizes separation between predefined groups without the strong assumptions of Bayesian clustering approaches. Finally, membership probability plots estimated the likelihood that each individual belongs to a particular genetic cluster, providing a visual summary of population structure across the four states.

The results paint a coherent phylogeographic picture. Samples from Punjab and Himachal Pradesh showed the closest genetic similarity to one another, followed by Karnataka and then Haryana, indicating that geographic proximity and possibly historical movement of dogs have left detectable signatures in the genome. On the distance tree, the Labrador Retriever clustered most closely with the German Shepherd, while the Pug grouped with the Indian breeds, the Gaddi, Tibetan Mastiff and Mudhol Hound. In the minimum spanning network, a Tibetan Mastiff from Himachal Pradesh occupied a central position, sharing many SNPs with surrounding samples, while a German Shepherd acted as a hub connecting several exotic-breed branches. The researchers caution that central nodes in such networks represent shared ancestral-like SNP profiles rather than actual ancestors of the other samples.

Perhaps the most striking conclusion is that the indigenous breeds carry greater genetic diversity than their exotic counterparts. The authors attribute this to long-term adaptation to local environments and varied ecological niches, from the high Himalayan pastures patrolled by Gaddi dogs to the Deccan plains where Mudhol Hounds have been bred for centuries as coursing hounds. Higher diversity can translate into greater resistance to disease and environmental change, and it bolsters the overall health of India’s canine gene pool. By contrast, exotic pedigree breeds, shaped by closed studbooks and intensive selection, typically exhibit narrower genetic variation. The contrast also highlights the threats posed by unregulated crossbreeding and urbanization, which could gradually erode the distinctive genetic makeup of native populations.

The study is the first to characterize the Gaddi and Mudhol Hound at the molecular level, filling a conspicuous gap in canine genomics. Dogs were the first animals domesticated from wild species roughly 15,000 years ago, and more than 400 breeds exist today, yet the genomic resources underpinning canine research, from the original 2005 Boxer reference genome to the newer German Shepherd-derived assembly and the Dog10K consortium’s sequencing of 2,000 canids, have focused overwhelmingly on Western breeds. By anchoring Indian dogs in that framework, using the ROS_Cfam_1.0 genome as the alignment reference, the new work extends the reach of population genomics to a region whose canine heritage has been underexplored.

Beyond its evolutionary insights, the research has practical implications for breeding and conservation. The dense SNP catalog provides a foundation for future genome-wide association studies aimed at identifying alleles underlying traits of interest, and for marker-assisted selection programs that could enhance genetic improvement while preserving diversity. The dataset has been deposited in the Sequence Read Archive under BioProject accession PRJNA843534, making it available to other researchers. The work was funded by the Department of Biotechnology, Government of India, through a collaborative project on parentage determination and cytogenetic profiling in dogs.

The broader message is one of urgency tempered by opportunity. The authors note that the creation of dog breeds is a recent event arising from mixed ancestral stock, and that localized breeding practices and historical forces have sculpted the genetic makeup of each population. The clear genetic clusters revealed by the DAPC and membership probability analyses confirm that breed boundaries are real and detectable in genomic data, even in a modest sample of fifty dogs. As India urbanizes and imported breeds grow in popularity, the window for documenting and conserving native canine genetic resources may be narrowing. This study demonstrates that the tools to do so, from restriction-enzyme-based sequencing to open-source population genetics software, are now firmly within reach, and that the dogs guarding Himalayan flocks and coursing hares across the Deccan carry a genomic legacy worth preserving.

Subject of Research: Genome-wide SNP-based phylogeographic and genetic diversity analysis of indigenous and exotic dog breeds in India

Article Title: Phylogeographic and genetic diversity analysis through genome-wide SNPs in indigenous and exotic canine breeds owned in India

Article References: Kaur, B., Yathish, H. M., Kashyap, N., & Mukhopadhyay, C. S. (2025). Phylogeographic and genetic diversity analysis through genome-wide SNPs in indigenous and exotic canine breeds owned in India. Discover Biotechnology, 2(1), Article 5. https://doi.org/10.1007/s44340-025-00012-3

Image Credits: AI Generated

DOI: 10.1007/s44340-025-00012-3

Keywords: dog genetics, genome-wide SNPs, ddRAD-GBS, Gaddi dog, Mudhol Hound, Tibetan Mastiff, phylogeography, genetic diversity, population structure, India, DAPC, conservation genomics