Introduction

Sunflower (Helianthus annuus L.) is an oilseed crop native to North America and one of the most important oil seed crops globally due to its high nutritional value and industrial applications1. It is a highly cross-pollinated crop with 2n = 34 chromosomes, exhibiting remarkable yield potential and adaptability to diverse environmental conditions2. According to reports from the United States Department of Agriculture (USDA), sunflower oil ranks third globally among edible oils derived from oilseeds, after soybean and rapeseed (canola). It is also rich in essential fatty acids such as linoleic acid (omega-6) and is important for human health3.

Sunflower hybrid varieties were first introduced in the early 1970s through the crossing of cytoplasmic male-sterile lines and fertility-restorer lines, marking a significant milestone in sunflower breeding4. Genetic diversity in plants, especially in sunflower, plays a key role in breeding and improving agronomic traits5. Genotypic selection requires diversity, and with increased genetic diversity in a population, the scope of both natural and artificial selection expands. In fact, genetic diversity is essential for farmers and breeders to develop new cultivars with high yield, adaptability to environmental stresses, and resistance to pests and pathogens6. A genetically diverse germplasm collection is essential for enhancing crop traits, as it promotes effective allele accumulation and streamlines the screening process7.

Since the morphological characterization of plant genetic resources has been limited to a few phenotypic traits, which are strongly influenced by environmental factors and show little variation, especially for highly heritable traits8. Therefore, the use of molecular markers as a precise and stable tool for evaluating genetic diversity and identifying important traits in plants is essential9.

A variety of molecular markers have been utilized to evaluate genetic diversity and identify potential heterotic groups in sunflower. These markers include restriction fragment length polymorphism (RFLP)10, amplified fragment length polymorphism (AFLP)11,12, random amplified polymorphic DNA (RAPD)13, simple sequence repeats (SSR)10,14, and single nucleotide polymorphism (SNP)15,16. The selection of each marker system depends on the specific application, the degree of polymorphism, available technical facilities and expertise, as well as financial limitations17.

Single nucleotide polymorphisms (SNPs) are the most common sequence variations in plant genomes and are widely used for analyzing genetic variation, population structure, marker-trait association, genomic selection, QTL mapping, and other breeding applications requiring extensive genome coverage18. SNP arrays based on hybridization and genotyping-by-sequencing (GBS) are among the most widely used platforms for genotyping and have been effectively applied in crop species. While GBS tends to have a high rate of missing data due to its low sequencing depth, SNP arrays provide higher marker density19.

Genotyping arrays serve as a practical alternative to whole-genome sequencing for acquiring genomic insights. Various arrays have been developed for SNP genotyping in sunflower, including 46 K20, 7 K and 5 K21, 1 K22, 25 K23.

Several studies have investigated the genetic diversity of sunflower. For example, a study developed an SNP array for sunflower, containing over 25 K markers. This array was successfully used for genotyping a panel of lines, hybrids, and mapping populations, accurately reflecting the genetic diversity of cultivated sunflower23. Filippi et al.24 used genotyping-by-sequencing (GBS) and generated a matrix of 11,834 common SNPs to assess the genetic diversity in sunflower germplasm collections from INTA, INRA, and USDA-UBC. The results showed that the genetic diversity was generally moderate. Population structure analyses and linkage disequilibrium (LD) patterns revealed structural differences between the collections, with the highest LD observed on chromosomes Chr10, Chr17, Chr5, and Chr2. The results of Mandel et al.25 using a 10 K Illumina SNP chip on the USDA-UBC accessions indicated that K = 3, supporting the existence of three genetically distinct clusters.

Given the importance of genetic diversity in sunflower germplasm and the limited application of GBS (Genotyping-by-Sequencing) in sunflower breeding programs, this study aimed to evaluate the utility of a 10 K SNP array for assessing genetic diversity, population structure, and identifying genomic regions under selection.

Materials and methods

Plant materials and SNP genotyping

In this study, 94 sunflower genotypes (Helianthus annuus L.) obtained from various research centers in France, Iran, and the USA were examined (Table 1). The genotypes were chosen to include lines with known resistance to diseases (e.g., Phoma, Sclerotinia) and tolerance to environmental stresses (e.g., drought, salinity, phosphorus deficiency).

Table 1 Name and origin of sunflower lines.
Full size table

Genomic DNA was extracted from young leaves of 15-day-old seedlings following the method described by Dellaporta et al.26 with slight modifications. The SNP genotyping of genomic DNA was conducted by TraitGenetics, Gatersleben, Germany (http://www.traitgenetics.com), using the 10 K sunflower Genotyping Array27, which contained 9,250 SNP markers. The variants were selected from the 25 K Infinium array and identified using the reference sequence of sunflower16. SNP markers with heterozygosity greater than 10%, minor allele frequency (MAF) less than 5%, and more than 10% missing data were filtered and removed to obtain more accurate results. Finally, 7,909 SNP markers remained and were used for further analysis. Their genomic distribution and density across 94 sunflower lines were visualized using the CMplot online tool at Bioinformatics.com.cn (https://www.bioinformatics.com.cn/plot_basic_SNP_density_by_CMplot_107_en).

Linkage disequilibrium (LD)

Linkage disequilibrium (LD) among SNP marker pairs was estimated using TASSEL version 5 software (https://www.maizegenetics.net/tassel). In this analysis, LD within the sunflower genome was calculated as a squared allele frequency correlation (r2) for SNP pairs with P-values < 0.001. This threshold ensures that only significant SNP pairs are included in the LD analysis. Also, LD decay distance was determined by plotting r2 values against genetic distance (in bp) using quantile regression with the quantreg package in R version 4.4.1 (https://CRAN.R-project.org/package=quantreg9:13).

Evaluation of population structure

To analyze the genetic structure and determine the most probable number of genetically distinct groups (K), the dataset was analyzed using the Bayesian model-based clustering algorithm in Structure version 2.3.4 software (https://www.stucture-software.org)28. A total of 20 runs were performed for each K value (K = 1–10) with 100,000 Markov chain Monte Carlo (MCMC) iterations and a burn-in period of 100,000 runs, based on an admixture model and correlated allele frequency29. The optimal number of subpopulations was identified using the Structure Selector program (https://lmme.ac.cn/StructureSelector/) by determining the best K value. Additionally, the 94 sunflower accessions were grouped hierarchically using the weighted pair group method with arithmetic mean (WPGMA) based on the similarity matrix30 in DARwin version 6.010 software (https://darwin.cirad.fr)31. PCoA analyse was also performed in GenAlex version 6.501 (https://biology-assets.anu.edu.au/GenAlEx).

Analysis of genetic diversity

Analysis of molecular variance (AMOVA) and genetic diversity parameters, including the number of alleles (Na), effective number of alleles (Ne), Shannon’s index (I), observed heterozygosity (Ho), expected heterozygosity (He), and the percentage of polymorphic loci (PPL%), were calculated for the studied groups using the GenAlex version 6.5 software (https://biology-assets.anu.edu.au/GenAlEx)32.Genetic diversity indices, including expected heterozygosity (Hs), observed heterozygosity (Ho), corrected heterozygosity for genetic diversity (Ht), average heterozygosity within groups (Htp), genetic differentiation index (Dst), intergroup genetic differentiation coefficient (Fst), and inbreeding coefficient (Fis) were calculated using adegenet package in R version 4.4.1 software.

Identification of KEGG pathways

Fst is a measure of genetic difference between subpopulations. High Fst values (> 0.25) ​​indicate that at some loci (gene regions) natural or artificial selection pressure has caused genetic differences in different populations33. In this study, the sequences of SNP loci with significant genetic divergence (Fst > 0.35) were aligned to the Helianthus annuus reference genome using the EnsemblPlants database (https://plants.ensembl.org/Multi/Tools/Blast). Genes with the lowest expected value (E-val) and the highest identity percentage were selected for gene ontology enrichment analysis through the DAVID database (https://davidbioinformatics.nih.gov/tools.jsp). Furthermore, the DAVID database was used to identify the associated KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways (www.kegg.jp/kegg/kegg1.html)34,35.

Results

Evaluation SNP data

The data on SNP markers are summarized in Table 2. A total of 7,909 SNPs were identified across the sunflower genome using SNP calling (Table 2, Fig. 1). These SNPs were physically mapped to 17 chromosomes. The highest number of SNPs was detected on chromosome 9 (622 SNPs), while the lowest was found in unmapped or unknown genomic regions (5 SNPs). The marker density varied across the sunflower genome, ranging from 0 to over 50 SNPs/Mb (Fig. 1). In general, SNPs were more abundant in the telomeric regions of the chromosome arms compared to the pericentromeric regions (exact centromere location unknown). Of the identified SNPs, 6,245 were transition types (A ↔ G and T ↔ C), and 1,664 were transversion types (A ↔ C, A ↔ T, T ↔ G, and C ↔ G) (Table 2). The ratio of transitions to transversions (Ts/Tv) ranged from 1.50 to 5.61 across chromosomes, with an average value of 3.75. This indicates that transitions were approximately 3.75 times more frequent than transversions.

Table 2 Summary of SNP data in the studied sunflower panel.
Full size table
Fig. 1
Fig. 1
Full size image

Genomic distribution of 7,909 SNPs across 94 diverse sunflower lines and the corresponding SNP density, plotted using the CMplot online tool at Bioinformatics.com.cn (https://www.bioinformatics.com.cn/plot_basic_SNP_density_by_CMplot_107_en).

The highest Ts/Tv ratio was observed on chromosome 7 (5.61), while the lowest was found on unmapped regions (1.50). The SNP density plot (Fig. 1) reveals a non-uniform distribution of SNPs across the genome, with certain chromosomes (e.g., chromosomes 6, 7, 8, 9, 10 and 17) showing higher SNP density on one arm than the other. MAF values showed moderate variation across chromosomes, ranging from 0.2586 (in chromosome 14) to 0.3350 (in chromosome 2), with a genome-wide average of 0.2850. PIC values also varied among chromosomes, with the lowest value of 0.2769 observed on chromosome 14 and the highest value of 0.3234 on chromosome 2, resulting in an overall mean of 0.2948. The value of r2, indicating linkage disequilibrium (LD) among SNP pairs, was used to calculate the relationships between SNP markers. A total of 358,995 marker pairs with an average r2 = 0.158 were identified across the whole genome, of which 103,265 pairs exhibited significant linkage at p < 0.001. The r2 values ranged from 0.1157 on chromosome 7 to 0.2387 on chromosome 2, suggesting stronger LD on chromosome 2. Additionally, Dist, representing the genetic distance between SNP pairs, varied from 6,433,051 base pairs on chromosome 8 to 12,983,604 base pairs on chromosome 4. The number of significant SNP pairs (NSSP) ranged from 2,220 on chromosome 7 to 9,305 on chromosome 10. These results indicate broad marker coverage, moderate linkage disequilibrium, and high marker resolution in this sunflower population (Table 3).

Table 3 Polymorphism information content (PIC) and minor-allele frequency (MAF) calculated for each chromosome by using SNPs and a summary of observed linkage disequilibrium (r2) among SNP pairs.
Full size table

The r2 values were plotted against the genetic distance across the genome to visualize the LD decay. A threshold of r2 = 0.1 was applied to define the cut-off for LD decay. The analysis showed that LD in the sunflower genome begins to decay at approximately 1.2 million base pairs (bp). Within this distance, the genetic relationship between between pairs of markers decreases significantly, as illustrated in Fig. 2.

Fig. 2
Fig. 2
Full size image

Linkage disequilibrium (r2) vs. physical distance (bp) for the sunflower genome. A cut-off line was plotted at r2 = 0.1.

Findings on population structure

Based on structural analysis using SNP markers and Bayesian model in STRUCTURE software, two genetic subpopulations (K = 2) were identified among 94 sunflower inbred lines (Fig. 3). The proportions of genotype membership were 75.53% for Subgroup 1 (red) and 6.38% for Subgroup 2 (green), while 18.09% of the inbred lines exhibited admixed genetic backgrounds. The WPGMA dendrogram based on shared-allele genetic distances divided the sunflower lines into two main clusters The first and second cluster consisted of cluster A (green) and cluster B (red) sunflower lines, respectively. Most lines in the green cluster, including lines 1, 10, 15, 20, 26, 46, and 74, originated from French breeding programs, suggesting a common genetic background. In contrast, the red cluster contained several lines from U.S. (USDA) breeding programs as well as some Iranian genotypes. The WPGMA clustering results were consistent with the STRUCTURE analysis, except for genotypes 49, 40, and 43, which were classified into different subgroups (Fig. 4).

Fig. 3
Fig. 3
Full size image

A panel of 94 sunflower inbred lines genomic structure declared with SNP markers in Structure software applying Bayesian method. At K = 2, the highest ΔK value was observed.

Fig. 4
Fig. 4
Full size image

Complete linkage clustering dendrogram generated using 7,909 SNPs and 94 sunflower inbred lines. Colors reflect groupings derived from structure analysis. The dendrogram was constructed using DARwin v6.0.10 (https://darwin.cirad.fr).

The PCoA plot also supported the observed population structure (Fig. 5). Subgroup 1 (represented in red), showed a wider spread in the plot, indicating higher genetic diversity, whereas Subgroup 2 (represented in green) appeared more tightly clustered, reflecting closer genetic similarity among its members. The first two principal coordinates explained 9.64% and 7.22% of the total genetic variation, respectively. This result is expected in a panel of highly inbred lines genotyped with 7,909 SNP markers. The high level of inbreeding leads to reduced genetic diversity, resulting in a lower total variation being captured by the Principal Coordinates.

Fig. 5
Fig. 5
Full size image

Principal coordinate analysis (PCoA) of 94 sunflower inbred lines based on 7909 SNP markers. Colors reflect groupings derived from structure analysis.

Genetic diversity

The analysis of molecular variance (AMOVA) showed that 16% of the total genetic variation was among populations, which was statistically significant (Fst = 0.156, P = 0.001). Additionally, 61% of the variation was attributed to differences among individuals within populations, and 24% was due to variation within individuals (Table 4). The analysis of genetic diversity between the two subgroups identified by structure analysis showed that Subgroup 1 exhibited higher genetic diversity with Na = 1.988, Ne = 1.647, and PPL = 98.81%, whereas Subgroup 2 showed lower genetic diversity with Na = 1.871, Ne = 1.360, and PPL = 87.06%. The overall mean for the number of observed alleles (Na = 1.929) and the number of effective alleles (Ne = 1.504) indicated a moderate level of genetic diversity in the population. Additionally, the high percentage of polymorphic loci (PPL = 92.94%) reflects a high potential for genetic improvement in this population (Table 5).

Table 4 Analysis of molecular variance (AMOVA) result from the diversity panel of 94 sunflower inbred lines.
Full size table
Table 5 Genetic variation among two subgroups identified by structure analysis on a diversity panel of 94 sunflower inbred lines.
Full size table

The genetic diversity parameters of the studied SNPs, summarized for each chromosome, are presented in Table 6. The level of variation in Ho, Hs, Ht, Fst, Fis, and Dest differed among the sunflower chromosomes. The highest values for Ho and Hs were observed for chromosomes 17 and 12 (0.10721 and 0.36733, respectively), which may be attributed to the presence of genes involved in maintaining genetic diversity in these regions. The Ht index was relatively uniform among chromosomes, with the lowest value (0.33085) recorded for chromosomes 14 and 13. The Fst values ranged from 0.03394 (chromosome 12) to 0.16564 (chromosome 4), indicating moderate to high genetic divergence among the studied genotypes across different chromosomes. Additionally, the Fis values, representing the level of inbreeding within populations, ranged from 0.61864 (chromosome 4) to 0.75483 (chromosome 12). These high Fis values indicate a strong excess of homozygosity, which is fully expected given that the materials used in this study are inbred lines. The Dest values, reflecting the true genetic differentiation among populations, ranged from 0.04364 (chromosome 12) to 0.20642 (chromosome 4). The high Dest value for chromosome 4 indicates that this chromosome may contain genomic regions with high genetic differentiation among populations.

Table 6 Diversity indices of 94 sunflower inbred lines using 7909 SNPs.
Full size table

Functional analysis of selected genomic regions

The process of selection can lead to changes in allele frequency within a population, resulting in genetic differentiation between populations in genomic regions under selection. Therefore, significant differences in allele frequency between populations can be considered as evidence of positive selection at specific genetic loci. Following selection, in addition to the allele associated with the target gene, neutral loci adjacent to the target allele are also retained in the genome, leading to an increase in the frequency of specific haplotypes in the selected region and a consequent reduction in haplotype diversity in that region. This shift in allele frequency and the frequency of specific haplotypes within a genomic region indicates the effect of natural selection36. To identify the regions under selection, Fst values between different subgroups were calculated (Fig. 6). In this study, we identified 285 genomic regions showing clear signatures of selection across the sunflower germplasm. These regions are associated with a total of 283 candidate genes (Table S1). Functional annotation of these genes using DAVID revealed three KEGG pathways, among which proteasome (P = 0.015) (Fig. 7) and pyruvate metabolism34,35 (P = 0.020) (Fig. 8) were statistically significant based on p-values (< 0.05), suggesting their potential biological relevance in the context of selection.

Fig. 6
Fig. 6
Full size image

Manhattan plot of distribution of genetic differentiation (Fst) values between identified subgroups.

Fig. 7
Fig. 7
Full size image

KEGG pathway of proteasome.

Fig. 8
Fig. 8
Full size image

KEGG pathway of pyruvate metabolism.

Discussion

Sunflower is an agronomically important oilseed crop and the application of molecular markers will improve the efficiency and accuracy of selection in sunflower breeding20. Among DNA markers, SNP are widely distributed across the genome and can be found in coding, non-coding, and intronic regions of genes37. They are effective tools for assessing genetic variation and improving breeding efficiency through marker-assisted selection (MAS) and genomic selection18. A few GBS approaches were reported recently for sunflower20,22,38,39. The use of GBS-SNP markers provides high-resolution insights into population structure and genetic diversity, enhancing breeding strategies and accelerating the development of improved sunflower cultivars. Expanding genetic diversity studies using SNP platforms can significantly improve sunflower breeding outcomes40.

The results of this study provide valuable insights into the genetic diversity and population structure of sunflower germplasm based on SNP markers. The identification of 7909 SNPs across the sunflower genome demonstrates the resolution and depth of the SNP calling method used in this study. The highest number of SNPs on chromosome 9 (622 SNPs) and the lowest in unmapped regions (5 SNPs) indicate differences in marker density and recombination rate across the genome. The higher abundance of SNPs in telomeric regions compared to pericentromeric regions is consistent with previous findings in other plant species19,41,42, which have attributed this pattern to increased recombination rate and gene density in telomeric regions.

The prevalence of transition mutations over transversion mutations, with a Ts/Tv ratio of 3.75, reflects the expected bias toward transitions in plant genomes and suggesting that the observed SNPs were not a product of sequencing and mapping errors43. Moreover, the Ts/Tv ratio observed in our study (3.75) was considerably higher compared to other plant species such as wheat (1.75;18), maize (2.05;19), and tall fescue aligned to wheat chromosomes (1.73;44). Notably, the high Ts/Tv ratio on chromosome 7 (5.61) indicates increased stability or selection pressure in this region. The PIC values in the present study ranged from 0.2769 to 0.3234, with a genome-wide average of 0.2948, indicating a moderate level of polymorphism across the sunflower genome. These values are in line with previous sunflower studies. For example, Filippi et al.45 reported an estimated PIC value of 0.232 using SNP markers. Similarly, Zeinalzadeh-Tabrizi et al.46 observed similar levels of diversity (0.10 to 0.58) using SSR markers. This consistency supports the reliability of the genetic diversity detected in the current germplasm panel. The MAF values observed in the present study (0.2850) are consistent with those reported by, Filippi et al.45, who showed MAF value greater than 0.1. MAF is a key indicator for identifying high-quality markers, with values greater than 0.5% indicating high polymorphism in the genome47. Higher MAF values enhance genetic variation analysis by increasing the detection of minor alleles and genetic diversity48. This emphasizes the importance of considering MAF in genome-wide mapping studies. It is worth mentioning that, the highest MAF (0.3350) and PIC (0.3234) on chromosome 2 suggest the presence of loci with high polymorphism and gene diversity, which may be useful for future breeding efforts.

In this study, the population structure analysis revealed two distinct subgroups (K = 2). Similarly, Rasoulzadeh Aghdam et al.49 identified two possible subpopulations (K = 2) in the studied population panel. In contrast, Mandel et al. (2013) using a 10 K SNP chip on the USDA-UBC accessions indicated that K = 3, supporting the existence of three genetically distinct clusters. The clustering pattern observed in both PCoA and WPGMA trees supports the presence of genetic differentiation between these subgroups, which may be attributed to differences in breeding origin and selection history, reinforced by inbreeding and limited gene flow among sunflower breeding programs. In several studies, PCoA has been used to assess population structure and confirm genetic structure in sunflower7,12,23,46. This method can usually be used in conjunction with other population structure analysis methods to better understand and examine the results.

Although, the use of SNP markers is common in genetic research, studies that specifically investigate the genetic diversity of sunflower using these markers are still limited. Filippi et al.24 analyzed 11,834 SNP markers using AMOVA and found that 4.58% of the total genetic variance was due to differences among breeding collections, indicating a significant but relatively low level of population differentiation. In contrast, Kholghi et al.14, using SSR markers, reported that 86% of the total genetic variation was attributed to within-population diversity among Iranian confectionery sunflower landraces, reflecting an open and heterogeneous genetic structure. Similarly, Basirnia et al.8, employing REMAP and IRAP markers, found that 94% of the genetic variation existed within groups, with only 6% attributed to between-group differences, further supporting the high intra-population diversity in sunflower germplasm. Additionally, in a study by Sahari et al.50 using 30 SSR markers and sequencing of 7 genes, a core collection of 8 populations was identified, capturing 88% of the genetic diversity present in Tunisian sunflower landraces.

In the current investigation, AMOVA was employed to evaluate genetic differentiation among groups. The analysis revealed statistically significant differences between the examined groups (p < 0.001), providing strong evidence for genetic divergence among the studied populations. Specifically, 16% of the total genetic variation was attributed to differences among populations, with an Fst value of 0.156, indicating a moderate to high level of genetic differentiation. while the remaining 84% was distributed within and among individuals, which may have resulted from selective breeding and environmental adaptation. Although high within-group diversity was observed, it does not rule out the existence of significant genetic structure across populations, as evidenced by AMOVA and Fst results. This pattern is further supported by the STRUCTURE and PCoA results, which are consistent with the differentiation identified by AMOVA.

An analysis of genetic diversity parameters in the sunflower panel studied showed that observed heterozygosity (Ho), expected heterozygosity (Hs), and fixation index (Fst) differed in different chromosomes, indicating a non-uniform distribution of genetic diversity across the genome. Notably, the Fst values ranged from 0.03394 to 0.16564, highlighting moderate to high genetic differentiation among genotypes depending on chromosomal regions. The Fst values reported in the present study were generally higher than those observed by Mandel et al.25,51, who estimated Fst values of approximately 0.049 (SNP markers) and 0.047 (SSR markers) in panel of sunflower lines. This difference may be due to the broader genomic coverage and greater genetic diversity present in the population analyzed here, as well as to differences in marker density, population structure, and germplasm composition used in the two studies. Also, the Ho value observed in the present study (0.093) was higher than those reported by Filippi et al.24, where observed heterozygosity ranged from 0.007 to 0.030 across different breeding populations. In contrast, the Hs values obtained here (0.312) are lower than those reported by Filippi et al.24 using a 11 K SNP markers on three breeding populations (0.454). The high Fis value observed across the whole genome (Fis = 0.695) indicates a substantial deficit of heterozygotes relative to Hardy–Weinberg expectations. This pattern is expected in sunflower inbred lines, which undergo multiple generations of selfing to produce genetically uniform and homozygous materials. Therefore, the elevated Fis reflects the intentional inbreeding used in line development instead of biological inbreeding, limited gene flow, or population structure typically discussed in natural populations.

Using the top 0.35 Fst values as a threshold19,52,53, multiple genomic regions showing signs of selection were identified across the studied subpopulations. One of the most important advantages of identifying selection signs is that it can be done solely by relying on molecular data, even when phenotypic information is not available, to identify genes involved in the differentiation process19. This represents the first known application of the Fst method for selection signature analysis in sunflower, specifically focusing on identifying genetic regions and KEGG pathways involved in population differentiation. Functional analysis of the identified candidate genes revealed that the genes under selection were mainly involved in the regulation of cell growth, protein degradation via the proteasome pathway, and energy production and management via pyruvate metabolism.

In sunflower, central metabolic pathways such as pyruvate metabolism are significantly enriched during seed development and oil accumulation, indicating their key role in energy processes and carbon flux underlying phenotypic diversity54. Similarly, components of the proteolytic system, particularly proteasome activity, are involved in the response to oxidative stress in sunflower leaves, suggesting their relevance to stress adaptation55.

Although genes involved in the proteasome and pyruvate metabolism pathways have been previously identified and functionally characterized in sunflower—mainly in the context of growth and stress response—the results of the present study indicate that these pathways are also targets of selection and have contributed to genetic differentiation among sunflower inbred lines. Interestingly, similar functional categories were also reported in maize by Arzhang et al.19, suggesting that conserved biological processes may underlie genetic differentiation across different crop species.

The present study provides a detailed analysis of genetic diversity in 94 sunflower inbred lines. Although increasing the number of samples in future studies could enhance the accuracy of the results, the current data still provide a good representation of the genetic diversity of sunflower inbred lines. Additionally, the use of a 10 K SNP array in this study has offered a sufficient coverage of the sunflower genome and has successfully identified selected genomic regions. However, the use of arrays with a higher number of markers could lead to more accurate assessments of the entire genome.

Conclusions

Genetic diversity is considered the main basis for selection and breeding in plant breeding programs. In this study, 94 sunflower inbred lines were genotyped using a 10 K SNP marker array to perform SNP calling and assess genetic diversity. The findings indicated the existence of significant genetic diversity among population (Fst = 0.156), as well as a heterogeneous distribution of genetic diversity across the genome. The use of Fst statistic and a threshold of 0.35, several genomic regions under selection were identified. Moreover, functional analysis of the candidate genes revealed that these genes are involved in critical biological processes such as cell growth regulation, degradation via the proteasome pathway, and pyruvate metabolism. These results provide valuable insights into the application of SNP markers for identifying genetic diversity and the underlying biological pathways of differentiation in sunflower. Additionally, they could be useful for future sunflower studies, which supported by marker-assisted selection and genomic selection.