Thanks to visit codestin.com
Credit goes to bmcbiol.biomedcentral.com

Skip to main content
BMC is moving to Springer Nature Link. Visit this journal in its new home.
Download PDF

Signatures in domesticated beet genomes pointing at genes under selection in a sucrose-storing root crop

BMC Biology volume 23, Article number: 299 (2025) Cite this article

Abstract

Background

The genus Beta encompasses important crops such as sugar, table, fodder, and leaf beets. All cultivated beets are believed to have originated from the wild sea beet, B. vulgaris subsp. maritima. Sugar beet, a recent crop dating back nearly 200 years, was selectively bred for enhanced root yield in combination with high sucrose content.

Results

We assembled a Beta diversity panel comprising wild and cultivated beet accessions. Whole-genome sequencing identified 10.3 million SNP markers. Four distinct genetic clusters were identified: table beet, sugar beet, Mediterranean sea beet, and Atlantic sea beet. A phylogenetic analysis revealed that cultivated beet accessions were genetically closer to Mediterranean than to Atlantic sea beet and that cultivated beets producing storage roots share a common ancestor. Cultivated beets exhibited genome regions with reduced nucleotide diversity compared to Mediterranean sea beets, indicating selection signatures. These regions contained putative candidate genes with potential roles in root development, suppression of lateral root formation, flowering time, and sucrose metabolism.

Conclusions

A yet unknown sucrose transporter on chromosome 6 showed reduced nucleotide diversity exclusively in sugar beet accessions compared to other Beta types with low sucrose content, suggesting its role in sucrose storage. Within a region of high nucleotide diversity between accessions with contrasting root phenotypes, we found two genes encoding auxin response factors, which play a crucial role in root development. We reason these genes to be significant root thickening regulators in root crops.

Background

The order Caryophyllales contributes important crops such as spinach, quinoa, amaranth, and buckwheat. Within this order, the genus Beta comprises cultivated and wild forms. Sugar beet, fodder beet (syn. mangelwurzel), leaf beet (syn. mangold, Swiss chard, chard, spinach beet), and table beet (syn. garden beet, red beet, beet root) belong to the species Beta vulgaris ssp. vulgaris. They are closely related to their wild relatives, B. vulgaris ssp. maritima, B. vulgaris ssp. adanensis, B. patula, and B. macrocarpa. The cultivated lineages exhibit pronounced phenotypic differences due to their use. It is believed that cultivated beets originated from the wild progenitors of B. vulgaris. ssp. maritima, also known as “sea beet” [1]. In a taxonomic update from 2006, the genus Beta was classified into two main Sects. [2]. The first section, Beta, encompasses all cultivated lineages, including B. vulgaris ssp. vulgaris, B. vulgaris ssp. maritima, B. vulgaris ssp. adanensis, B. patula, and B. macrocarpa. The section Corollinae consists of B. corolliflora, B. macrorhiza, B. lomatogona, and B. nana.

Leaves of the Mediterranean sea beet populations have been consumed as a vegetable for over 2000 years. The enlarged root types (historically termed “Runkelrüben”) have been speculated to originate in the Near East (Iraq, Iran, and Turkey) and then spread west [3]. However, the period of domestication is still uncertain.

Table and leaf beet are important vegetables cultivated in both temperate and tropical climates worldwide. The petioles of leaf beets exhibit a rich spectrum of colors, encompassing shades of white, yellow, pink, and red. Table beets are primarily grown for their succulent roots and hypocotyls, which exhibit a wide range of diversity in terms of shape, including round, globe-shaped, flattened, and cylindrical roots. The red color in their roots is attributed to the presence of betalain pigments, while yellow hues result from betaxanthin pigments. The accumulation of betalains in beets is regulated by two loci, namely R and Y, which have been identified and molecularly described. The BvCYP76AD1 gene at the R locus encodes a cytochrome P450 enzyme essential for betalain biosynthesis [4]. Conversely, the Y locus contains the anthocyanin MYB-like gene BvMYB1, determining the presence or absence of pigmentation (red or yellow) in the flesh of beet roots [5]. Notably, the dried red beet powder, known for its antioxidant properties, is employed as a natural food dye in the food industry [6, 7].

In 1747, the German chemist A.S. Marggraf found that “Runkelrüben” roots store sucrose (then known familiarly as “cane sugar”). This incentivized growing Runkelrüben for sucrose production by the end of the eighteenth century. The sucrose content of beet roots at that time was around 4% (fresh weight). Selection for root dry mass started after the first sugar factory was built in Silesia in 1801, then a part of Germany. Mass selection to improve open-pollinated populations was the major breeding method for the first hundred years before introducing progeny testing [8]. This resulted in an immense increase in root weight and sucrose content [9, 10]. Today, sugar beets are cultivated in more than 50 countries around Europe, Asia, the Americas, and parts of Africa (FAOSTAT, 2022: OECD/FAO (2022). Most modern beet varieties are F1 hybrids. In Europe, root yield ranges between 60 and 85 t/hectare, and sucrose content ranges between 17 and 19% fresh weight. In contrast, the sucrose concentration in fodder beet is lower, ranging between 4 and 10% [11].

Wild sea beets produce small, non-spherical sprangled roots characterized by numerous lateral/secondary branches emerging from a single taproot [6]. Remarkably, there are no discernible phenotypic differences in the roots of leaf beets compared to sea beets. However, table beets, fodder beets, and sugar beets exhibit significantly enlarged storage organs, predominantly formed by a non-branched primary taproot and, to some extent, by the hypocotyl.

Sea beet roots possess supernumerary root cambia, although they are not notably swollen compared to the roots of table, fodder, and sugar beets. It can be hypothesized that the presence of supernumerary root cambia might have facilitated the selection of roots with swollen characteristics. Nevertheless, there is considerable variation in cambium ring numbers in both wild and cultivated beets. Typically, the roots of fodder and table beets have 3–5 cambium rings, and sugar beets can produce up to 12 successive concentric rings of cambia [10]. Table and fodder beets produce an enlarged hypocotyl, constituting a substantial part of the storage organ, in contrast to sugar beets, where the root is the primary storage organ [1]. Cultivated beets with thick storage roots have been intensively selected against lateral or secondary root branching, resulting in a thick single taproot. The development of thick storage roots can be attributed to concurrent and synchronous increases in cell division and enlargement within these cambial rings [6, 12].

Cultivated beets are biennial species that form a large leaf and root mass in the first year under natural conditions; however, annual variants do exist, particularly among certain breeding lines, landraces, and leaf beet types. Conversely, sea beets are generally annuals, but biennial and even perennial forms also occur, especially along the Atlantic coasts and in northern European climates. The floral transition occurs after winter, initiating with shoot elongation termed bolting and the production of up to 10,000 florets [10]. Flowering must be avoided entirely during field production due to a drastic reduction in sucrose yield. Beets are an allogamous crop species due to a complex gametophytic self-incompatibility in wild and cultivated forms, preventing self-pollination. However, self-incompatibility is overcome by a self-fertility locus with a self-fertility allele SF, allowing inbred line production for hybrid breeding [1].

Sea beets are more abundant around the Mediterranean and at the coast of Western and Northern Europe; however, they can be also found is more distant regions like Asia. Although they belong to the same species, their phenological development and morphology are strikingly different from cultivated root types. They remain a vital genetic resource for various biotic stress resistances [13].

Cultivated beets and wild forms are diploid (2n = 2x = 18) with an estimated genome size ranging from 714 to 758 Mb [14]. The beet genome is highly complex and repetitive, with approximately ~ 63% of its content comprising repetitive elements [15]. The first reference genome assembly was derived from the doubled-haploid sugar beet “KWS2320” and comprised ~ 567 Mbp, with 85% of the sequences assigned to 9 chromosomes and predicting 27,421 protein-coding genes [16]. The inbred line “EL10” genome was recently published using long-read sequencing, BioNano optical mapping, DoveTail Hi-C scaffolding, and Illumina short-reads. The “EL10.2_2” genome assembly comprises ~ 569 Mb, with 99.2% of the sequences assigned to nine chromosomes and 24,186 predicted protein-coding genes [17]. Two more fully annotated de novo genome assemblies have been published for one B. vulgaris ssp. maritima (“WB42”) and one B. patula accession (“BETA548”) [18, 19]. The assembled genome sizes for B. vulgaris ssp. maritima and B. patula were 590 Mb and 624 Mb, respectively.

Whole-genome sequencing studies have evaluated genetic diversity within and between cultivated and wild beets. Galewski and McGrath (2020) analysed genetic diversity and phylogenetic relationships among 23 cultivated beet accessions, representing a broad range of phenotypic variation for foliar and root traits. Each accession was treated as a population due to the pooled sequencing of individual plants belonging to that particular accession. Table beets were differentiated as a distinct crop type compared to sugar beets. Fodder and chard accessions were intermediate between table and sugar beet [9]. Another study estimated phylogenetic relationships among 606 sugar beet and wild beet accessions using reference-free, distance-based approaches [20]. Sea beets clustered into Mediterranean and Atlantic accessions. Sea beets from Greece were closest to sugar beets, which led the authors to suggest that sugar beets might originate from this region. In a recent study, genomic data from 656 genomes belonging to section Beta of the genus Beta were compared to characterize genes undergoing selection during domestication and breeding between wild and cultivated accessions [21]. In total, 10 million variants were identified across the 656 beet accessions, encompassing wild and cultivated forms. Across the sugar beet genomes, various regions totaling 15 Mb were characterized under artificial selection due to low genetic variation. These variation-poor regions were enriched for genes involved in shoot system development, stress response, and carbohydrate metabolism.

The objectives of this study were to investigate the population structure and phylogenetic relationships among cultivated beet types and their wild sea beet relatives, and to determine whether different cultivated forms can be genetically distinguished using genome-wide data. We also aimed to analyze patterns of genetic diversity and differentiation across both wild and cultivated Beta vulgaris accessions. Furthermore, we sought to identify genomic regions and candidate genes associated with domestication and artificial selection, particularly those influenced by intensive breeding during the last 200 years. A worldwide collection of wild and cultivated accessions was grown in the field to assess agronomically important phenotypes. After whole-genome sequencing, the population structure of cultivated beet lineages and their wild relatives was assessed. We depicted genome regions with remarkably high sequence diversity between lineages with contrasting phenotypes such as root thickening and bolting. Low nucleotide diversity within cultivated lineages highlighted regions under intense selective pressure during the past 200 years. Candidate genes were localized within these regions with putative functions in storage root formation and sucrose storage. These genes are proposed to be key genes for root crop domestication and sucrose storage.

Results

Assembling a Beta mini-core collection

Based on the passport data, we selected 485 B. vulgaris L. accessions with diverse geographical origins, including Atlantic and Mediterranean sea beet accessions, to capture the extensive genetic diversity in wild sea and cultivated beets. Together, these accessions encompass the substantial phenotypic variation of the species.

First, an observation trial was performed in the field, and accessions displaying phenotypic heterogeneity or not corresponding to the seed bank’s passport data were discarded. Next, we curated the passport data, ensuring only phenotypically true-to-type accessions were included in a Beta mini-core collection assembly. The remaining heterogeneous accessions were discarded. As a result, a Beta mini-core collection was assembled, consisting of 290 highly homogeneous accessions encompassing both wild sea beets and cultivated beets. It comprises 91 B. vulgaris ssp. maritima (69 Mediterranean and 22 Atlantic sea beets), 81 sugar beet, 59 table beet, 30 fodder beet, and 29 leaf beet accessions from 40 countries worldwide. This collection represents the large phenotypic diversity of this genus regarding phenological development, storage root formation, sucrose storage, and leaf and root coloration (Fig. 1).

Fig. 1
figure 1

Phenotypic diversity displayed among wild and cultivated beets captured in the study. The upper panel displays, from left to right, a non-bolting Atlantic sea beet plant with white, fangy, and branched root; a bolting Mediterranean sea beet plant with white, fangy, and branched root; a leaf beet plant with broad light green wrinkled leaves, a white and depressed petiole, and sea beet-like fangy root; a fodder beet plant with yellow and cylindrical thickened storage root; a fodder beet plant with yellow and round thickened storage root; a fodder beet plant with orange-red external root color and presenting sugar beet-like conical thickened storage roots with a prominent groove; a fodder beet plant with yellow-orange and flat thickened storage root; and a sugar beet plant with a thickened white conical storage root, a source of sucrose. The lower panel, from left to right, illustrates a table beet plant with dark red leaves, petioles, and a round, thickened storage root with bulky shoulders; a table beet plant with dark red leaves, petioles, and a round, thickened storage root with tapered shoulders; a table beet plant with green leaf blades and red veins, red-pink petioles, and a round, thickened storage root with tapered shoulders; a table beet plant with dark red leaves, petioles, and a sugar beet-like conical thickened storage root with a prominent groove; a table beet plant with dark red leaves, petioles, and a sugar beet-like conical thickened storage root without a groove also known as “smooth root”; a table beet plant with dark red leaves, petioles, and a cylindrical thickened storage root; and top view of cross-sections of different table beet roots displaying a target-like appearance due to the accumulation of betalain pigments in the parenchyma and/or cambium storage root tissues

Full size image

Whole genome sequencing of a Beta mini-core collection reveals high genetic variation among different crop types

Whole genome sequencing of 290 accessions was performed, and about 4.2 Tb of whole-genome sequencing data was generated after sequencing a representative plant from each accession. In total, 28.1 billion raw reads were generated (Additional file 1: Table S1). After trimming the adapter content, 27.6 billion reads were retained (Additional file 2: Table S2). The accessions’ GC content (%) ranged between 35 and 38%. The coverage depth ranged between 15.4 × and 45.75 × per sample, with an average depth of ~ 19.86 × per sample (Additional file 3: Table S3).

Filtered reads from all accessions, including B. vulgaris ssp. maritima, were mapped to the long-range sugar beet genome assembly EL10.2_2. The average mapping rate of single reads to this genome was 98.61%, and the average mapping rate for both read pairs correctly mapped to the reference was 87.36%. In general, accessions belonging to cultivated lineages showed a greater number of mapped reads and properly paired mapped reads than wild accessions. Sugar beet accessions exhibited the highest percentage of total mapped reads, ranging from 97.73 to 98.94%, and both pairs of properly mapped reads, ranging from 85.65 to 89.45%. The B. vulgaris ssp. maritima total mapped reads ranged from 97.62 to 98.75%, and both-pairs-properly-mapped-reads ranged from 84.17 to 88.05%.

The mean coverage breadth achieved was 85.62% of the reference EL10.2_2 genome assembly, with sugar beet accessions exhibiting the highest average coverage fraction of 86.55% (83.72 to 88.72%). As expected, wild sea beet accessions showed a lower average coverage fraction of 84.91% (80.28 to 87.21%) (Table 1, Additional file 3: Table S3).

Table 1 Summary statistics of mapped reads from 290 beet accessions using the EL10 genome as a reference
Full size table

Approximately 85 million variant sites, including single-nucleotide polymorphisms (SNPs) and insertions/deletions (INDELs), were genotyped across 290 accessions with reference to the EL10.2_2 genome assembly. We filtered for biallelic variants with a maximum of 10% missing data, a minor allele frequency > 5%, and a mean depth > 5 across all genotyped accessions. After filtering, 11,669,554 high-confidence (HC) variants were identified across 290 accessions. Of these, 10,365,662 (88.8%) were SNPs, and 1,303,892 (11.2%) were small INDELs (Additional file 4: Table S4).

The variant density was high across the genome, averaging 20.5 variants/kb. Generally, the highest variant density (SNPs and INDELs) was observed at the chromosomal ends. The average density of SNPs across 290 accessions was 1822 SNPs/100 kb, ranging from 0 to 4659 SNPs/100 kb. The average number of INDELs across 290 accessions was 229.2 INDELs/100 kb, ranging from 0 to 601 INDELs/100 kb (Fig. 2). The number of variants across different chromosomes ranged from 1.1 to 1.5 million (Additional file 4: Table S4, Additional file 5: Fig. S1), with an average of 1.29 million/chromosome. The highest number of variants was observed on chromosome 5, regardless of chromosome length.

Fig. 2
figure 2

Visualization of 11 million variants (SNPs and INDELs) of the Beta mini-core collection across nine EL10 sugar beet chromosomes. The CIRCOS plot shows three tracks (outside to inside) representing (i) gene density, (ii) SNP density, and (iii) INDEL density across nine sugar beet chromosomes. Chromosome lengths are in Mb. Gene, SNP, and INDEL densities were calculated in 100 kb non-overlapping windows

Full size image

For SNP variant positions, 3,019,720 (29.2%) SNPs were located in upstream and downstream regions of a predicted gene, 1,865,832 (18.0%) SNPs were located within predicted intronic regions, and 4,845,240 (46.8%) SNPs were located within intergenic regions (Additional file 6: Table S5). Only 479,462 (4.6%) SNPs were within predicted open reading frames. Of these, only 211,493 SNPs (44.1%) were predicted to be within a coding region, with 267,969 (55.9%) being synonymous variants and 206,807 (43.1%) being missense variants. Only 3481 (0.73%) of the variants led to “stop gained” or “nonsense” mutations (Additional file 7: Table S6).

Of INDEL variants, 736,158 (56.5%) were deletions, and the remaining 566,978 (43.5%) were insertions concerning the reference genome. In total, 448,030 (34.4%) INDELs were located in upstream and downstream regions of a predicted gene, 304,470 (23.4%) were located within predicted introns, and 504,179 (38.7%) of INDELs were located in intergenic regions (Additional file 6: Table S5). Only 16,213 (1.2%) INDELs were within predicted open reading frames. For INDELs altering the predicted polypeptide sequence, 9252 (57.1%) were predicted as frameshift mutations (Additional file 7: Table S6).

Population structure of the beet diversity panel

Leaf beet accessions were grouped with the cluster of the Mediterranean Sea beet accessions. PC2, which accounted for 3.1% of the variance, primarily separated the different cultivated beet lineages (Fig. 3A, Additional file 8: Fig. S2). We excluded the sea beet accessions from a second PCA analysis to further discern the genetic sub-clusters within the cultivated beet lineages. In this case, PC1 explained 4.4% of the variance and effectively differentiated between table beet and sugar beet accessions, with each cultivated lineage forming a distinct cluster along PC1. Fodder beet and leaf beet accessions showed less pronounced divergence and appeared intermediate between sugar beet and table beet accessions. Additionally, we observed a distinct separation between annual and biennial leaf beet accessions (Additional file 9: Fig. S3).

Fig. 3
figure 3

Genetic diversity, population structure, and Neighbor-joining phylogenetic tree of 290 wild and cultivated beet accessions from the Beta mini-core collection. A Principal component analysis (PCA) of 290 beet accessions. PC1 and PC2 represent the first two analysis components, accounting for 5.87 and 3.08% of the total variation, respectively. The colors represent different genetic clusters according to their beet type, with the following accessions: wild accession from the Atlantic (purple), Mediterranean (blue), sugar beet (dark green), table beet (red), fodder beet (orange), and leaf beet (light green). B The population structure analysis was conducted on the Beta mini-core collection using model-based clustering in ADMIXTURE with different numbers of ancestral kinships (K = 2 and K = 10). K = 4 was determined as the optimal number of populations. The graphs illustrate the ancestry proportions at K = 2, 3, and 4. The Y-axis represents the share of genetic ancestry, while the X-axis represents different accessions. In the plot, blue represents the Mediterranean sea beet ancestry, green represents sugar beet ancestry, purple represents the Atlantic sea beet ancestry, and red represents table beet ancestry. C Neighbor-joining phylogenetic tree of the 290 accessions. Based on PCA and the breadth of coverage, the tree was rooted using an Atlantic sea beet accession as an outgroup. The color of the branches corresponds to the color codes used in PCA. A blue arrow indicates hybrid accessions from private breeding companies. An orange arrow indicates the fodder, table, and sugar beet accessions forming conical roots

Full size image

As a next step, we performed a population structure analysis with the ADMIXTURE software. We used cross-validation errors to estimate the most suitable number of populations. At K = 2, two distinct clusters of B. vulgaris ssp. maritima and B. vulgaris ssp. vulgaris were formed. However, the table beet and sugar beet accessions in this case shared a similar genetic ancestry. At K = 3, a distinct cluster of table beet accessions was evident. At K = 4, minimum cross-validation was observed (Additional file 10: Fig. S4), with four distinct genetic clusters (Fig. 3B) corresponding to table beet, sugar beet, Mediterranean sea beet, and Atlantic sea beet ancestries. Two distinct genetic clusters of wild sea beet accessions with different ancestry were evident, as well as two distinct genetic clusters within B. vulgaris ssp. maritima could easily be distinguished, which was consistent with their morphology. Table beet and sugar beet accessions appeared to have clear genetic clusters with less admixture from other genetic clusters, which was also consistent with phenotypic observations. Interestingly, table beet accessions with complete genetic ancestry produce round-red roots. However, the table beet accessions with varying admixture from sugar beet ancestry form conical-red roots with banding patterns in the storage cambium tissues (Fig. 1).

No distinct clusters of leaf and fodder beet ancestry were formed. Interestingly, leaf beets shared substantial genetic ancestry with the Mediterranean sea beet accessions (Fig. 3B). The biennial-leaf beet accessions had more sugar beet ancestry than annual-leaf beet accessions. Fodder beet shared substantial genetic ancestry with table, sugar, and Mediterranean sea beet ancestry. Accessions within the fodder beet cluster carrying different levels of sea beet ancestry may indicate varying degrees of breeding intensity. The fodder beet accessions with higher Mediterranean sea beet ancestry produce round beets. However, the accessions with higher sugar beet ancestry form sugar beet-like conical roots.

Based on genetic ancestry, highly admixed sugar beet, table beet, and Mediterranean sea beet accessions were removed, and 245 accessions were kept for further analysis.

Lineage-specific variation and phylogeny of the beet diversity panel

After removing the admixed accessions, genetic variation within each cultivated lineage and the wild sea beet populations was calculated. The number of SNPs was reduced to 10,052,949, and the INDELs to 1,260,345 across 245 samples. The Mediterranean sea beet accessions exhibited the highest number of SNPs and INDELs (9,751,509), followed by leaf beet (9,647,860), fodder beet (9,272,343), sugar beet (8,196,734), the Atlantic sea beet (7,456,809), and table beet accessions (7,425,689) (Additional file 11: Table S7).

The Mediterranean sea beet accessions shared the highest number of SNPs with leaf beet accessions (7,747,807 out of 9,520,600). In contrast, table beet accessions had the fewest SNPs in common with the Mediterranean sea beet accessions (5,879,955 out of 9,423,745). Interestingly, the table beet accessions also exhibited the lowest number of lineage-specific SNPs. Sugar beet accessions shared 6,258,918 SNPs with the Mediterranean sea beet accessions, but they also carried the highest number of lineage-specific SNPs (Additional file 12: Table S8).

The Atlantic sea beet sub-cluster shared the most SNPs with the Mediterranean sea beet accessions (5,845,172 out of 9,493,914). The fewest SNPs were shared with sugar beet (4,912,371) and table beet (4,364,533). Interestingly, all sub-clusters, except table beet, exhibited an abundance of lineage-specific SNPs. Additionally, table beet accessions shared the fewest common SNPs with the Atlantic sea beet (Additional file 13: Table S9).

Consistent with the PCA and admixture analysis results, the phylogenetic analysis revealed that cultivated beets are more closely related to Mediterranean sea beet than Atlantic sea beet. As expected, leaf beet accessions formed a sister clade with the Mediterranean sea beet accessions. All cultivated beets with storage roots formed one major clade. Within this “enlarged root” clade, the table beet accessions formed a sister clade with fodder and sugar beet accessions. The fodder beet, conical red table beet, and sugar beet accessions formed a single clade. Commercial hybrids formed a separate clade, sharing the most common ancestor with the sugar beet germplasm resources from public breeding programs (Fig. 3C).

Linkage disequilibrium and heterozygosity within the genetic sub-clusters of the beet diversity panel

We calculated linkage disequilibrium (LD) across the genome. LD decay between sequence variants averaged ~ 2.5 kb. A rapid LD decay was more apparent in Mediterranean sea beet accessions (~ 1.7 kb) than in cultivated beet lineages. Within the cultivated beet lineages, the table beet accessions had the largest LD blocks (~ 5.9 kb) compared to sugar beet (~ 4.6 kb), fodder beet (~ 4.5 kb), and leaf beet accessions (~ 3.0 kb) (Fig. 4A, Table 2).

Fig. 4
figure 4

Linkage disequilibrium (LD) decay, percentage of heterozygous sites, population differentiation (FST), and nucleotide diversity within different genetic clusters of the Beta mini-core collection. The analyses were performed using whole-genome SNP data from fodder beet (n = 30), sugar beet (n = 71), leaf beet (n = 29), table beet (n = 44), Atlantic sea beet (n = 18), and Mediterranean sea beet (n = 53) accessions. The color used in the analyses corresponds to the color codes used in PCA. A LD decay within each sub-cluster of the Beta mini-core collection. The Y-axis represents the pairwise correlation coefficient (r2) between two SNP markers, and the X-axis represents the physical distance (kb) between corresponding SNP markers. B Overview of heterozygosity within each genetic cluster. Each box plot represents the percentage of heterozygous SNP sites within each genetic cluster. The red dot represents the mean value, while the centerline inside each box corresponds to the median. The lower and upper hinges of the box represent the 25th and 75th percentiles, respectively. Any data points located outside the whiskers are considered outliers. C Nucleotide diversity within both cultivated beet lineages and the Mediterranean sea beet accessions, along with pairwise FST values denoting the differentiation of cultivated beet lineages from the Mediterranean sea beet accessions. Each circle corresponds to the nucleotide diversity value of its respective group, and the value along each connecting line indicates the FST value between the two groups. D Overview of nucleotide diversity within each genetic cluster. Each box plot represents the nucleotide diversity within each genetic cluster. The red dot represents the mean value, while the centerline inside each box corresponds to the median. The lower and upper hinges of the box represent the 25th and 75th percentiles, respectively. Any data points located outside the whiskers are considered outliers

Full size image
Table 2 Linkage disequilibrium (LD) analysis across cultivated and wild Beta accessions
Full size table

Among the cultivated lineages, table beet displayed large LD blocks across all chromosomes except chromosome 1 (Additional file 14: Fig. S5). The largest LD blocks were on chromosome 2. In contrast, sugar beet accessions showed the largest LD blocks on chromosome 1 (Additional file 15: Fig. S6). Leaf and fodder beet accessions displayed uniform LD decay across all nine chromosomes.

Cultivated and wild beets are outcrossing species due to an efficient self-incompatibility system. While commercial beet varieties are hybrids, inbred populations are commonly maintained by sib-mating. Therefore, we expected a considerable degree of heterozygosity. Across all the 290 accessions, 20.97% of the SNP sites were heterozygous (Additional file 16: Table S10). The highest average number of SNP variant sites was observed in the Mediterranean sea beet accessions (22.9%) followed by leaf beet (21.9%), fodder beet (21.1%), sugar beet (20.2%), table beet (19.8%), and Atlantic sea beet (19.3%) accessions (Additional file 17: Fig. S7). In general, leaf beet and Mediterranean sea beet plants displayed higher heterozygosity rates than the rest of the mini-core collection (Additional file 16: Table S10). The sugar beet accessions showed the highest variation regarding heterozygosity rates, ranging from 5.24 to 34.46% (Fig. 4B).

Patterns of genomic variation between cultivated and wild beet accessions

We conducted pairwise FST calculations between cultivated and wild accessions to measure population differentiation. First, the Mediterranean Sea beet accessions served as the reference population due to their highest heterozygosity, suggesting their potential ancestral position for Atlantic and crop types. The highest mean fixation index was observed between the Atlantic and the Mediterranean sea beet accessions (FST = 0.103). Comparing Mediterranean sea beet and cultivated lineages revealed lower FST values (0.099 for table beet, 0.094 for sugar beet, 0.054 for fodder beet, and 0.019 for leaf beet) (Fig. 4C, Table 3).

Table 3 Pairwise population differentiation (FST) comparisons between the Mediterranean sea beet accessions and the rest of the Beta mini-core collection
Full size table

Then, we compared the Atlantic sea beet with the cultivated lineages, which resulted in higher FST values than the comparison with Mediterranean sea beet (Additional file 18: Table S11). The highest FST value was observed between table beet and Atlantic sea beet (0.197), followed by sugar beet (FST = 0.178), fodder beet (0.148), and leaf beet (0.116). The lowest FST values were found between the Mediterranean and the Atlantic sea beets.

We also assessed the nucleotide diversity within cultivated and wild beets (Additional file 19: Table S12). Surprisingly, the leaf beet accessions had the highest nucleotide diversity (π = 5.17 × 10−4), followed by the Mediterranean sea beet (π = 5.09 × 10−4), fodder beet (π = 4.96 × 10−4), sugar beet (π = 4.43 × 10−4), and table beet (π = 4.17 × 10−4). The lowest nucleotide diversity was observed within the Atlantic sea beet accessions (π = 3.96 × 10−4) (Fig. 4C,D).

Signatures in domesticated beet genomes pointing at genes under selection in a sucrose-storing root crop

The domestication and improvement process often leads to a purifying selection of genomic regions associated with favorable phenotypic traits. We expected that selection for highly heritable traits like sucrose content, bolting resistance, and root weight would leave selection footprints in the sugar beet genome. Therefore, we first employed XP-CLR and FST analyses, focusing on the top 5% of regions overlapping both approaches.

First, we compared sugar beet with the Mediterranean sea beet. We identified selective sweep regions in the EL10 genome spanning approximately 44.47 Mb (using XP-CLR) (Additional file 20: Fig. S8) and 69.97 Mb (using FST) (Additional file 21: Fig. S9). In total, 30.27 Mb (5.32%) of the EL10 genomic region overlapped between both analyses, indicating putative selection footprints. The majority, approximately 54.16% (16.39 Mb), of the regions under selection were concentrated on chromosomes 1, 6 (Fig. 5A,B), and 2, where the longest and the highest number of sweeps were also located (Table 4). Within these genome regions, we identified 1741 potential domestication and improvement genes.

Fig. 5
figure 5

Selective sweeps in sugar beet and gene regions under selection in sugar, table, and fodder beets. A XP-CLR and FST across EL10 sugar beet chromosome 1 using 10-kb sliding window size and 1-kb step size while comparing sugar beet accessions with the Mediterranean sea beet accessions. The blue dots represent the XP-CLR and mean FST values in the upper and lower plots along the EL10 chromosome 1. The horizontal red line marks the empirical 95th percentile of XP-CLR and FST values. The red dashed box highlights the regions of interest carrying interesting candidate genes under strong selection. B XP-CLR and FST across the EL10 sugar beet chromosome 6. C Venn diagram of shared and lineage-specific genes under selection in sugar beet, table beet, and fodder beet accessions across nine EL10 chromosomes

Full size image
Table 4 Summary statistics of selective sweep regions using XP-CLR and FST across EL10 chromosomes within the cultivated beet lineages forming a storage root. The overlapping genomic regions from XP-CLR and FST were considered the selective sweep regions. The table shows the total length (bp), maximum length (bp), average length (bp), and the number of selective sweep regions in sugar, table, and fodder beets across nine EL10 chromosomes
Full size table

Then, we compared table beet with Mediterranean sea beet. We identified selective sweep regions spanning approximately 45.53 Mb using FST and 65.68 Mb (XP-CLR) of the EL10 genome. In total, 25.74 Mb (4.53%) of the EL10 genomic region overlapped between both analyses, indicating putative selective sweep regions in table beet accessions. Nearly 53.41% (13.75 Mb) of the table beet genome under selection was localized on chromosomes 6, 2, and 8. Furthermore, the maximum length of selective sweeps and the number of sweeps were predominantly located on chromosomes 2, 6, and 8 (Table 4). Within these regions, we identified 1856 potential candidate genes associated with selection or domestication in table beet accessions.

Next, we compared fodder beet accessions with the Mediterranean sea beet. Putative selective sweep regions spanned approximately 45.66 and 72.77 Mb of the EL10 genome using FST and XP-CLR, respectively. In total, 29.46 Mb (5.18%) overlapped between the two analyses. Nearly 50.43% (14.86 Mb) of the fodder beet genome under selection was situated on chromosomes 2, 6, and 4, where the maximum number of sweeps was also located (Table 4). Within these regions, we identified 1815 genes.

We reasoned that intensive selection for high root weight and against bolting over the past 200 years has left its footprint in the cultivated root types (sugar beet, fodder beet, and table beet). Consequently, genes or genome regions involved in beet domestication and improvement traits should be conserved across these lineages. Among the genes within the candidate selection signature regions of these lineages, 381 genes were common (Fig. 5C). As expected, sugar beet and fodder beet had the highest number of shared genes (745), followed by fodder beet/table beet (679), and sugar beet/table beet (546). Notably, table beet accessions exhibited the highest number of lineage-specific genes under selection (1012), followed by sugar beet (831) and fodder beet (772).

Next, we searched for genes putatively involved in agronomic traits typical for a sucrose-storing biennial root crop. The vast number of DNA polymorphisms provided an opportunity to investigate nucleotide diversity between cultivated lineages and the Mediterranean sea beet at the gene level. First, we focussed on putative sucrose transporters. Sugar beet, table beet, and fodder beet accessions exhibited reduced nucleotide diversity within a homolog of the Arabidopsis SUCROSE TRANSPORTER 4 (SUT4) on chromosome 1 when compared to leaf beets and Atlantic and Mediterranean sea beet (Fig. 6A). Another sucrose transporter, SUT2, on chromosome 6 showed reduced nucleotide diversity exclusively in sugar beet accessions compared to all other sub-clusters (Fig. 6B). However, no signatures of selective sweeps were detected in the sucrose transporter BvTST2.1, which is responsible for vacuolar sucrose uptake in sugar beet taproots [22], and BvSUT1, a member of the disaccharide transporter (DST) superfamily involved in the loading of sucrose into the phloem of sugar beet leaves [23].

Fig. 6
figure 6

Nucleotide diversity within fodder beet, leaf beet, sugar beet, Atlantic sea beet, and Mediterranean sea beet accessions in and around the genomic regions housing genes involved in sucrose metabolism and auxin response. The color of the line corresponds to the color codes of the sub-clusters used in PCA. The Y-axis depicts the nucleotide diversity value, while the X-axis denotes the location along the chromosome. The vertical red lines represent the gene boundaries. A Nucleotide diversity values between different sub-clusters in the selective sweep region spanning the SUT4 gene on chromosome 1. B Nucleotide diversity values between different sub-clusters in the selective sweep region spanning the SUT2 gene on chromosome 6. C Nucleotide diversity values between different sub-clusters in the selective sweep region spanning IAA8 gene on chromosome 1. D Nucleotide diversity values between different sub-clusters in the selective sweep region spanning ARF5 gene on chromosome 1

Full size image

At the end of chromosome 1, homologs of two auxin-responsive genes from Arabidopsis are located within a region of reduced nucleotide diversity, INDOLEACETIC ACID-INDUCED PROTEIN 8 (IAA8) (Fig. 6C) and AUXIN RESPONSE FACTOR 5 (ARF5) (Fig. 6D). In Arabidopsis, it has been shown that auxin response factors play a crucial role in root development, lateral root formation, and the pattern formation of the root central vascular cylinder and vascular tissue. We hypothesize that cultivated beets were domesticated for their ability to form thickened roots. During the beet improvement process, breeders selectively increased the root size while simultaneously reducing root branching, also known as fangy roots, a typical characteristic of sea beet roots.

In cultivated beet accessions producing thickened storage roots, a complete loss of genetic diversity was observed within and around IAA8 and ARF5, when compared to leaf beets, and Atlantic and Mediterranean sea beets. In Arabidopsis, it has been shown that IAA8 is expressed in the developmental vasculature of the shoot apex, hypocotyl, and root tip, acting as a transcriptional repressor of auxin response and controlling lateral root formation [24]. Interestingly, ARF5, widely known as MONOPTEROS (MP), encodes a transcription factor, IAA24, and plays a role in mediating embryo axis formation, vascular development, and the polar transport of auxins [25, 26].

Inositol transporters and raffinose synthase genes have been functionally characterized in beet. Therefore, we selected INOSITOL TRANSPORTER 1 (INT1) [21] on chromosome 3 and RAFFINOSE SYNTHASE 5 (RS5) [27] on chromosome 1. It is known that these genes are involved in post-harvest freezing tolerance, an important trait targeted during beet breeding. Raffinose, a trisaccharide, acts as a cryoprotectant in the pith of the storage taproot, and inositol is crucial for synthesizing raffinose. Interestingly, we found a reduced nucleotide diversity within INT1 among sugar beet and table beet accessions compared to other sub-clusters (Additional file 22: Fig. S10A). Additionally, reduced nucleotide diversity was observed within RS5 only in sugar beets compared to all other sub-clusters (Additional file 22: Fig. S10B).

Surprisingly, we found no evidence of selective sweeps within the ORFs of previously described flowering time genes such as BvBTC1, BvFT1, BvFT2, and BvBBX19 [28,29,30,31,32]. However, we did observe a reduced nucleotide diversity in the putative FLC (FLOWERING LOCUS C) ortholog on chromosome 6 [33] and within a homolog of SUPPRESSOR OF OVEREXPRESSION OF CO 1 (SOC1), an important floral integrator and a downstream target of FLC in Arabidopsis (Additional file 22: Fig. S10C-D).

Table beets are known for their distinctive earthy taste in their roots, which is believed to have been intentionally selected for enhanced flavor. A QTL study in table beets revealed a strong association between high geosmin concentration and a locus on chromosome 8, with terpene synthesis genes located within this QTL region. Terpene synthase genes are known to be involved in geosmin biosynthesis [34]. Additionally, table beet accessions displayed a reduced nucleotide diversity within terpene synthase genes on chromosomes 7 and 8 (Additional file 23: Fig. S11A-B).

In addition to the identified candidate genes, we detected a notable reduction in nucleotide diversity within a WUSCHEL-RELATED HOMEOBOX 4 (WOX4) gene region on chromosome 1. This reduction was particularly evident in sugar, table, and fodder beets, as compared to leaf beets and wild sea beets. This localized decline in and around the WOX4 gene suggests a potential signature of selection acting on the gene itself or its regulatory neighborhood during beet domestication (Additional file 23: Fig. S11C).

Demographic reconstruction based on coalescent-inferred effective population size (Ne) revealed distinct historical trajectories among six Beta vulgaris groups: Mediterranean sea beet, Atlantic sea beet, table beet, sugar beet, leaf beet, and fodder beet (Additional file 24: Fig. S12). All cultivated groups exhibited clear signatures of historical bottlenecks, whereas wild populations retained higher ancestral diversity.

Sugar beet experienced the most pronounced and recent bottleneck, with an effective population size (Ne) dropping sharply before recovering to a stable plateau. Table beet exhibited a persistently low Ne (~ 104) across the entire timescale, indicating a long-term bottleneck likely shaped by intense selection with limited post-domestication recovery and limited breeding pools. Fodder beet followed a similar trajectory as table beet, showing a moderate bottleneck followed by demographic stabilization in most recent times. Notably, leaf beet showed a modest increase in Ne within the most recent times. This recent expansion may reflect relaxed selection, broader cultivation, or recent introgression. Among the wild populations, the Mediterranean sea beet retained a high ancestral Ne (~ 105–10⁶) followed by a recent increase, while the Atlantic sea beet exhibited a modest decline before stabilizing. These results support a scenario of strong domestication bottlenecks in sugar beet and other cultivated forms, in contrast to the more stable demographic histories of their wild progenitors.

Discussion

We assembled a diversity panel of the genus Beta consisting of cultivated beet lineages and their wild progenitor B. vulgaris ssp. maritima (sea beet). This mini-core collection represents this species’ diverse geographical origin and phenotypic variation. Whole genome sequencing unraveled the genotypic diversity within and between cultivated and wild lineages, the gene flow between subspecies, and the footprints of selection during domestication and breeding. Within the genomes of cultivated beets, putative selective sweeps of extremely low genetic diversity were identified. These regions harbor candidate genes with a putative role in storage root formation and sucrose storage.

We applied PCA and admixture analyses to analyse the relatedness between sea beet accessions. Consistent with prior studies, sea beets formed two distinct clusters (Mediterranean and Atlantic) [20, 21, 35]. This divergence is likely a result of the geographical separation between these regions, environmental heterogeneity, and local adaptation leading to limited genetic exchange or gene flow among the wild accessions. The two sea beet populations are separated by the Iberian Peninsula and the Pyrenees mountains, which might have historically restricted gene flow. Additionally, the Atlantic coast is characterized by a cooler, wetter climate, and stronger maritime influence, in contrast to the warmer, drier Mediterranean basin. These climatic differences impose distinct selective pressures on traits such as flowering time, vernalization response, salt tolerance, and life cycle duration (annual vs. biennial/perennial). Such environmental divergence likely facilitated ecological speciation processes, reinforcing genetic differentiation even in the absence of strong reproductive barriers. These findings are consistent with earlier studies that report morphological, physiological, and genetic divergence between these coastal ecotypes [36, 37].

Consistent with other studies, the Atlantic sea beets, especially those from Ireland, were most distantly related to the sugar beet line EL10, the donor of the reference genome sequence [21]. PCA and phylogenetic analyses revealed remarkable proximity between the Mediterranean sea beet and cultivated beets, suggesting that the Mediterranean sea beet is the immediate ancestor of cultivated beets. Moreover, a combination of PCA, admixture, shared SNPs, and FST data revealed for the first time that leaf beets, among the cultivated lineages, are genetically closest to the Mediterranean sea beets. Remarkably, this finding aligns with historical data, indicating that ancient Greek and Roman civilizations used the leaves of the vegetable beet in their culinary practices and medicinal remedies [1, 6]. Interestingly, all cultivated beets with enlarged thickened roots lie in one clade, which suggests that the beets producing thickened roots share a more recent common ancestor different from leaf beets. Noteworthy, all modern hybrids cluster closely in one subclade within the sugar beet clade, consistent with previous findings [20].

Then, we analysed the relatedness among cultivated lineages. Sugar and table beet accessions exhibited distinct genetic clusters, with only a few accessions having genetic admixtures from other ancestries. This result reflects strong selection, intense breeding for diverse purposes, and genetic drift resulting from reproductive isolation within these cultivated lineages. The divergence between these two lineages corresponds to previous studies [9, 38, 39]. The effective population size further indicated these patterns, and a recent rapid bottleneck was identified in the past 500 years within both sugar and table beet cultivated lineages (Additional file 24: Fig. S12). Documentary evidence also supports this finding, indicating that beet roots were not used as livestock feed or as a vegetable before the sixteenth century [6, 40].

A small subset of sugar beet accessions exhibited highly admixed ancestry, primarily from Mediterranean sea beet, corresponding to their diverging phenotypic characteristics like root fanginess. These accessions originated from Turkey and Greece, which are recognized as the primary centers of beet domestication [6, 9, 20]. This could mean that beets from these regions were first domesticated from the wild Mediterranean sea beet, and later, they have not undergone further admixture or intense selection. Alternatively, this finding may indicate inaccurate passport data.

Most table beets originating from Greece, the Soviet Union, and Turkey displayed a distinct, clean ancestry, which indicates extensive selection and breeding efforts. The conical roots and an alternating red and white pattern in the parenchyma and vascular tissue, observed in certain table beet accessions due to betalain accumulation, could be attributed to the introgression of genomic regions governing the conical root shape and sucrose content from sugar beet ancestry, as revealed by admixture analysis.

Based on our PCA analysis, fodder beet and leaf beets were considered intermediate between table and sugar beets, which was also supported by the admixture analysis, where no distinct genetic clusters were found for these cultivated lineages. Notably, among fodder beets, those accessions with a higher percentage of sugar beet ancestry, characterized by their sugar beet-like conical roots, originate from Europe. It is widely acknowledged that the “White Silesian” sugar beet progenitor was derived from sucrose-storing “Runkelrüben” lineages. Thus, our findings align with the existing knowledge regarding the genetic relationships between fodder and sugar beets [3, 9, 41]. Furthermore, the fodder beet accessions with round roots originating from Syria, Iran, Turkey, and China exhibited a notably higher percentage of Mediterranean sea and table beet ancestry than sugar beet (Fig. 4B). This observation leads to the tempting speculation that fodder beets with both round and flat roots emerged in the Mediterranean region, and the development of such fodder beets could be attributed to the spontaneous or deliberate hybridization events between table beets and Mediterranean sea or leaf beet accessions.

Cultivated beets, which form a storage root, displayed reduced genetic diversity compared to the Mediterranean sea beet, likely due to the bottleneck effect resulting from domestication and artificial selection. The fixation index (FST), a measure of population differentiation, ranged from 0.019 to 0.099 in leaf beet and table beet accessions. This range is comparable to the FST difference between Spinacia oleracea and S. turkestanica (0.030) [42] but considerably lower than the difference observed between lettuce species (0.305) [43] and cultivated and wild rice (0.30) [44]. The lower FST values could be attributed to the high number of shared SNPs between leaf and fodder beets and Mediterranean sea beets.

We examined the impact of domestication and artificial selection at the gene level, identifying genome regions with reduced genetic diversity in cultivated beets. According to FST and XP-CLR analyses, shared regions were considered areas under strong selection. Domestication and intensive breeding likely contributed to the decreased genetic diversity in these regions, with the implicated genes influencing these processes. We found 30.27 Mb of genome region with very low diversity or complete fixation in sugar beet but with high diversity in the Mediterranean sea beet accessions. They can be interpreted as selective sweeps due to strong selection over the past 200 years. Notably, all agronomically important characters can be selected in root-type forms before flowering, which may explain the high success rate of mass selection.

In the beet lineages characterized by enlarged taproots, 381 shared genes were identified as being under selection. Homologs of two auxin-responsive genes from Arabidopsis (IAA8, ARF5) are potential candidate genes involved in the formation of storage roots. Auxin plays a pivotal regulatory role in diverse developmental processes, encompassing root and stem growth, vascular differentiation, embryo patterning, and lateral branching patterns in the root [45]. In Arabidopsis, IAA8, encoding a transcriptional repressor of the auxin response, interacts with the TIR1 auxin receptor and ARF (auxin response factors) transcription factors in the nucleus, thereby governing lateral root formation. Transgenic lines overexpressing IAA8 exhibited a significantly lower number of lateral roots than the wild type, whereas loss-of-function mutant lines of IAA8 displayed a significantly higher number of lateral roots [24]. A study identified potential interactions of the IAA8 protein with ten ARFs, including ARF5 [46].

We hypothesize that the interaction between Aux/IAA transcriptional repressors and auxin response factor genes plays a crucial role in reducing lateral branching and increasing root size. In beets, it has been demonstrated that the transition from primary growth to secondary growth involves transient changes in the levels of auxin (IAA), active cytokinins (CKs), and abscisic acid (ABA). This suggested their involvement in regulating the initiation of secondary growth from the cambial rings [47].

In addition to these genes, we observed reduced genetic diversity in a WUSCHEL-RELATED HOMEOBOX 4 (WOX4) gene on chromosome 1, which is under artificial selection in sugar, table, and fodder beets. In Arabidopsis, cambium cell-specific transcript profiling demonstrated that WOX4 is one of the two key master regulators of cambium activity. It plays a role in xylem cell expansion and stem cell regulation and promotes cambial cell proliferation in Arabidopsis roots [48].

We were also interested in genes regulating sucrose storage. We identified two sucrose transporter genes on chromosomes 1 and 6. The consistent pattern of significantly reduced genetic diversity in both genes suggests their potential involvement in high sucrose storage in sugar beet roots.

Conclusions

Our study sheds light on the genetic diversity and composition of cultivated and wild beets, revealing the impact of domestication and intensive breeding on the beet genome. Future research will provide deeper insights, including haplotype analyses, and building and functional analyses of candidate genes. The Beta mini-core collection and the genome sequences are important resources that will allow genome-wide association studies and allele mining to introduce new allelic variation to the gene pool of cultivated beets.

Methods

Plant materials and growth conditions

First, we selected 485 wild and cultivated beet accessions from the Plant Breeding Genebank at CAU, Kiel, representing diverse geographical origins. Plants were grown in row plots in the field in Kiel (Germany) between May and October 2020 to assess the phenological development, phenotypic characters, and homogeneity within accessions. Based on these data, 290 phenotypically homogenous wild and cultivated beet accessions were chosen as a Beta mini-core collection for further analyses (Additional file 25: Table S13) (Fig. 1). It incorporates germplasm from various sources, including the USDA (Fort Collins and East Lansing, USA), the IPK Genebank (Gatersleben, Germany), and collections from the Plant Breeding Institute, Kiel, as well as commercial hybrids from Germany. This collection includes both annual and biennial accessions from Beta vulgaris ssp. vulgaris and B. vulgaris ssp. maritima.

The mini-core collection was grown in a field near Göttingen (Germany) in 2021 and 2022. At the end of April, seeds were hand-sown as single-row plots with seven plants/row in a randomized complete block design with three replications. Each replication block consisted of 290 row plots, each containing a single accession. Each row plot contained seven plants per accession. Each replication block was divided into two sub-blocks. Each sub-block was 13.5 m × 8 m in size. The distances between rows and between plants were set to 45 and 20 cm, respectively. In total, 870 row plots were sown in all three replication blocks.

DNA extraction and whole-genome sequencing

Leaf samples were collected from young leaves of 6-week-old single plants grown in the field. Leaf samples were transported on dry ice and were immediately lyophilized. For DNA extraction, one representative plant was selected from each accession at the end of the cultivation season 2021. DNA was isolated using the QIAGEN DNeasy Plant Mini Kit following their standard protocol. The purity and quality of DNA were assessed by agarose gel electrophoresis, and the concentration was determined using a NanoDrop2000 spectrophotometer (ThermoFisher Scientific, Waltham, United States).

Novogene (United Kingdom) performed whole-genome sequencing using short reads on the Illumina NovaSeq 6000 sequencing platform. We aimed to achieve an average of 12 Gb of paired-end (PE) 2 × 150 bp reads with a quality Q > 30 Phred score per sample. This coverage is approximately equivalent to approximately 15 times the haploid sugar beet genome (~ 758 Mb).

Raw reads quality check and alignment

Raw sequence reads were obtained in the zipped FASTQ format. Quality check for the raw reads was performed using the FastQC v0.11.9 [49] and MultiQC v1.13 [50] algorithms. Raw reads were trimmed, and library bar-code adapters were removed with trimmomatic-v0.39 [51] using the following criteria: ILLUMINACLIP:adapters.fa-:2:30:10, LEADING:3; TRAILING:3; SLIDINGWINDOW:4:15; MINLEN:36. Filtered reads were used for further downstream analysis. A long-read genome assembly of a sugar beet inbred line EL10 was used as a reference genome [17]. Filtered reads were aligned to the EL10.2_2 reference genome assembly (Phytozome genome ID: 782), using BWA-MEM (v-0.7.17) [52] with default parameters. The sequence alignment files (SAM) from each accession were sorted and indexed individually using SAMtools [53]. Read grouping was assigned using AddOrReplaceReadGroups, and PCR duplicates were marked using MarkDuplicates on each bam file using Picard tools (http://broadinstitute.github.io/picard/).

GATK best practices were followed to call and filter the variants [54]. Firstly, variants were identified with GATK (v4.2.5.0) [55] using HaplotypeCaller function on each bam file in a genomic variant call format (GVCF) mode. Then, individual g.vcf files from 290 samples were consolidated into a single GVCF file using CombineGVCFs function of GATK. Finally, SNPs and small InDels were identified with a joint calling approach using the GenotypeGVCFs function of GATK. To obtain high-confidence variants, SNPs were hard-filtered using parameters “QD < 2.0 | | FS > 60.0 | | MQ < 40.0 | | SOR > 3.0 | | MQRankSum < − 12.5 | | ReadPosRankSum < − 8.0,” and indels with “QD < 2.0 | | FS > 200.0 | | SOR > 10.0 | | MQRankSum < − 12.5 | | ReadPosRankSum < − 8.0.” The hard-filtered variants with genotype calls with a depth < 2 and > 50 were excluded. Finally, only bi-allelic variants were kept, and variants with a missing rate of > 10% or a minor allele frequency (MAF) of < 0.05 were removed, which resulted in a high confidence (HC) set of ~ 11.7 million variants, including SNPs and small InDels. The complete bioinformatics pipeline is shown as a workflow (Additional file 26: Fig. S13).

Variant annotation

We annotated the HC variant set using the Ensemble Variant Effect Predictor (VEP) [56] based on the EL10.2_2 sugar beet genome and annotation (Phytozome genome ID:782). We used VEP to categorize variants in coding regions based on their features: synonymous, missense, splice acceptor, splice donor, splice region, start lost, start gained, stop lost, and stop retained. Gene content, SNP density, and INDEL density within 100 kb non-overlapping windows were plotted using Circos [57]. Then, we used the HC variant set for phylogeny, population stratification, demography, and selective sweep analyses.

Phylogeny and population structure analyses

For clustering analysis, the variants were pruned for linkage disequilibrium (LD) to reduce the bias and redundancy. The independent variants, i.e., LD pruned variants, were used for PCA (principal component analysis) and admixture analysis. Therefore, the HC variant set was pruned for LD with PLINK (version 1.9) [58] using a window size of 10 kb with a step size of one SNP and an r2 threshold of 0.5, resulting in 1.4 million pruned variants. A PCA was performed on LD pruned variants using the SNPrelate package in R [59] (Additional file 27: Note S1). We estimated the top 10 principal components. The first (PC1) and second (PC2) principal components were plotted using custom R scripts. A neighbor-joining (NJ) tree was constructed using PHYLIP with 100 bootstraps. The p-distance matrix was generated from the variant call format (VCF) file using VCF2Dis (v1.47), fneighbor, and fconsense (Version: EMBOSS:6.6.0.0 PHYLIPNEW:3.69.650). An NJ tree was constructed with 100 bootstraps using PHYLIP (version 3.696), a consensus was generated, and the tree layout was generated using FigTree. The accession with the lowest breadth of coverage and furthest placement on PC1 from the rest of the accessions was used to root the tree.

A population structure analysis was performed using ADMIXTURE (Version: 1.3.0) [60]. We analyzed population structure using cluster K, ranging from 1 to 10, with a default fivefold cross-validation (–cv = 5). Each K was run with ten replicates. Obtained Q matrices were aligned using Pong [61]. Highly admixed sugar beet, table beet, and Mediterranean sea beet accessions were discounted, and only accessions with > 60% genetic ancestry threshold at K = 4 were chosen for further analyses.

Linkage disequilibrium analysis

The LD was calculated for each subpopulation or genetic cluster using the HC variants set. Then, LD was calculated for the whole population, including wild accessions. The LD was calculated on SNP pairs within a 500-kb window using default parameters in PopLDdecay (v 3.31) [62]. The LD decay was plotted using custom R scripts based on the ggplot2 package. The LD decay was measured as the distance at which Pearson’s correlation efficiency (r2) dropped to half of the maximum.

Population genomics analysis

Pairwise population differentiation (FST) and nucleotide diversity (π) were calculated. Individuals were divided into subpopulations based on the results of the clustering analysis. Nucleotide diversity (π) for subpopulations and genetic differentiation (FST) between different subpopulations were performed within a non-overlapping 10-kb sliding window and 1-kb steps using VCFtools (version 0.1.13) [63]. FST values were calculated between wild and cultivated beet populations using the 10 kb non-overlapping window approach. The top 5% ratios of mean FST values between wild and cultivated population regions were considered candidate regions for population divergence.

The inbreeding coefficient (F) was calculated using VCFtools (version 0.1.13) [63] to evaluate heterozygosity at variant loci in individual accessions. The inbreeding coefficient across different cultivated lineages was plotted using custom R scripts.

Identification of domestication sweeps

Selective sweeps in the sugar beet reference genome were identified using three different approaches, calculating XP-CLR [64] scores with the parameters “-w 1 0.005 200 1000 -p0 0.95,” FST, and nucleotide diversity (π) ratios, using VCFtools (v0.1.13) [63]. All three approaches involve comparing allele frequencies within and between subpopulations. We extracted the top 5% of scores/values from XP-CLR and FST analyses. The overlapping regions between the two analyses were considered potential regions under selection. To consolidate these overlapping regions, we used the bedtools merge option. We then identified the genes located within these regions by using bedtools intersect with the.GFF file.

We applied nucleotide diversity (π) as a criterion to further refine our candidate regions of selective sweeps. Specifically, we focused on regions with exceptionally low nucleotide diversity within cultivated lineages but with high nucleotide diversity within B. vulgaris ssp. maritima accessions. Within these regions exhibiting low genetic diversity, we gave special attention to genes related to sucrose transport, inositol transport, and auxin metabolism.

Demography history of beet evolution

MSMC2 (version 2.1.4) [65] was used to estimate the effective population size. Briefly, VCFtools was used to generate population-wise VCF files without any missing data. Using default parameters, the SNPs within each sub-population were phased using BEAGLE (version 5.4) (Browning and Browning 2007). Eight randomly chosen samples from each subpopulation were selected from the population VCF file. Then, a masked file was generated using bamCaller.py for each sample. Then, the “generate_multihetsep.py” script within MSMC tools was used to create input files for MSMC2 for each chromosome separately. MSMC2 was then run on the phased SNPs with a generation time 1 and the mutation rate per generation per site as 4 × 10−8.

Data availability

All data generated or analysed during this study are included in this published article, its information files and publicly available repositories. The raw sequencing data generated for the Beta Mini-Core Collection (BMCC) are available in the NCBI Sequence Read Archive under BioProject accession number PRJNA1311133 (https://dataview.ncbi.nlm.nih.gov/object/PRJNA1311133; https://www.ncbi.nlm.nih.gov/Traces/study/?acc=PRJNA1311133). SNP and INDEL variant files in VCF format, derived from whole-genome resequencing of 290 beet accessions, are available on Figshare at https://doi.org/10.6084/m9.figshare.30015358.v1.

Abbreviations

%:

Percent

ABA:

Abscisic acid

ARF :

AUXIN RESPONSE FACTOR

BTC1 :

BOLTING TIME CONTROL 1

BWA:

Burrows-Wheeler Aligner

DNA:

Deoxyribonucleic acid

IAA :

INDOLE-3-ACETIC ACID

INDEL:

Insertion or Deletion

Kbp:

Kilo base pairs

Mbp:

Mega base pairs

ORF:

Open Reading Frame

PCR:

Polymerase chain reaction

ROS:

Reactive oxygen species

SNP:

Single-nucleotide polymorphism

SUT/SUC:

SUCROSE TRANSPORTER

References

  1. Biancardi E, Panella LW, McGrath JM. Beta maritima: The Origin of Beets. Cham: Springer International Publishing. 2020;2(2):XXVI, 284.

  2. Kadereit G, Hohmann S, Kadereit JW. A synopsis of Chenopodiaceae subfam. Betoideae and notes on the taxonomy of Beta. Willdenowia. 2006:9-19.

  3. Scott R, Cooke D. The sugar beet crop: Chapman & Hall; 1993.

  4. Hatlestad GJ, Sunnadeniya RM, Akhavan NA, Gonzalez A, Goldman IL, McGrath JM, et al. The beet r locus encodes a new cytochrome P450 required for red betalain production. Nat Genet. 2012;44(7):816–20.

    Article  CAS  PubMed  Google Scholar 

  5. Hatlestad GJ, Akhavan NA, Sunnadeniya RM, Elam L, Cargile S, Hembd A, et al. The beet Y locus encodes an anthocyanin MYB-like protein that activates the betalain red pigment pathway. Nat Genet. 2015;47(1):92–6.

    Article  CAS  PubMed  Google Scholar 

  6. Goldman IL, Janick J. Evolution of root morphology in table beet: historical and iconographic. Front Plant Sci. 2021;12:689926.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Goldman IL, Austin D. Linkage among the R, Y and BI loci in table beet. Theor Appl Genet. 2000;100(3):337–43.

    Article  Google Scholar 

  8. Gayon J, Zallen DT. The role of the Vilmorin Company in the promotion and diffusion of the experimental science of heredity in France, 1840–1920. J Hist Biol. 1998;31(2):241–62.

    Article  CAS  PubMed  Google Scholar 

  9. Galewski P, McGrath JM. Genetic diversity among cultivated beets (Beta vulgaris) assessed via population-based whole genome sequences. BMC Genomics. 2020;21(1):189.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Melzer S, Müller AE, Jung C. Genetics and Genomics of Flowering Time Regulation in Sugar Beet. In: Tuberosa R, Graner A, Frison E, editors. Genomics of Plant Genetic Resources: Volume 2 Crop productivity, food security and nutritional quality. Dordrecht: Springer, Netherlands; 2014. p. 3–26.

    Chapter  Google Scholar 

  11. Melzer S, Müller AE, Jung C. Genetics and genomics of flowering time regulation in sugar beet. Genomics of Plant Genetic Resources: Volume 2 Crop productivity, food security and nutritional quality: Springer; 2013. p. 3–26.

  12. Artschwager E. Anatomy of the vegetative organs of the sugar beet. J agric Res. 1926;33(2):143–76.

    Google Scholar 

  13. Frese L, Desprez B, Ziegler D. Potential of genetic resources and breeding strategies for base-broadening in Beta. 2000.

  14. Arumuganathan K, Earle ED. Estimation of nuclear DNA content of plants by flow cytometry. Plant Mol Biol Rep. 1991;9(4):415–415.

    Google Scholar 

  15. Flavell RB, Bennett MD, Smith JB, Smith DB. Genome size and the proportion of repeated nucleotide sequence DNA in plants. Biochem Genet. 1974;12(4):257–69.

    Article  CAS  PubMed  Google Scholar 

  16. Dohm JC, Minoche AE, Holtgräwe D, Capella-Gutiérrez S, Zakrzewski F, Tafer H, et al. The genome of the recently domesticated crop plant sugar beet (Beta vulgaris). Nature. 2014;505(7484):546–9.

    Article  CAS  PubMed  Google Scholar 

  17. McGrath JM, Funk A, Galewski P, Ou S, Townsend B, Davenport K, et al. A contiguous de novo genome assembly of sugar beet EL10 (Beta vulgaris L.). DNA Res. 2023;30(1):dsac033.

    Article  PubMed  Google Scholar 

  18. Capistrano-Gossmann GG, Ries D, Holtgräwe D, Minoche A, Kraft T, Frerichmann SLM, et al. Crop wild relative populations of Beta vulgaris allow direct mapping of agronomically important genes. Nat Commun. 2017;8(1):15708.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Rodríguez Del Río Á, Minoche AE, Zwickl NF, Friedrich A, Liedtke S, Schmidt T, et al. Genomes of the wild beets Beta patula and Beta vulgaris ssp. maritima. Plant J. 2019;99(6):1242–53.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Sandell FL, Stralis-Pavese N, McGrath JM, Schulz B, Himmelbauer H, Dohm JC. Genomic distances reveal relationships of wild and cultivated beets. Nat Commun. 2022;13(1):2021.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Felkel S, Dohm JC, Himmelbauer H. Genomic variation in the genus Beta based on 656 sequenced beet genomes. Sci Rep. 2023;13(1):8654.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Jung B, Ludewig F, Schulz A, Meißner G, Wöstefeld N, Flügge U-I, et al. Identification of the transporter responsible for sucrose accumulation in sugar beet taproots. Nat Plants. 2015;1(1):14001.

    Article  CAS  PubMed  Google Scholar 

  23. Nieberl P, Ehrl C, Pommerrenig B, Graus D, Marten I, Jung B, et al. Functional characterisation and cell specificity of BvSUT1, the transporter that loads sucrose into the phloem of sugar beet (Beta vulgaris L.) source leaves. Plant Biol. 2017;19(3):315–26.

    Article  CAS  PubMed  Google Scholar 

  24. Arase F, Nishitani H, Egusa M, Nishimoto N, Sakurai S, Sakamoto N, et al. IAA8 involved in lateral root formation interacts with the TIR1 auxin receptor and ARF transcription factors in Arabidopsis. PLoS ONE. 2012;7(8):e43414.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Berleth T, Jürgens G. The role of the monopteros gene in organising the basal body region of the Arabidopsis embryo. Development. 1993;118(2):575–87.

    Article  Google Scholar 

  26. Przemeck GKH, Mattsson J, Hardtke CS, Sung ZR, Berleth T. Studies on the role of the Arabidopsis gene MONOPTEROS in vascular development and plant cell axialization. Planta. 1996;200(2):229–37.

    Article  CAS  PubMed  Google Scholar 

  27. Keller I, Mudsam C, Rodrigues CM, Kischka D, Zierer W, Sonnewald U, et al. Cold-triggered induction of ROS- and raffinose metabolism in freezing-sensitive taproot tissue of sugar beet. Front Plant Sci. 2021;12:715767.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Dally N, Eckel M, Batschauer A, Höft N, Jung C. Two CONSTANS-LIKE genes jointly control flowering time in beet. Sci Rep. 2018;8(1):16120.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Höft N, Dally N, Hasler M, Jung C. Haplotype variation of flowering time genes of sugar beet and its wild relatives and the impact on life cycle regimes. Front Plant Sci. 2018;8:2211.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Pin PA, Benlloch R, Bonnet D, Wremerth-Weich E, Kraft T, Gielen JJL, et al. An antagonistic pair of FT homologs mediates the control of flowering time in Sugar beet. Science. 2010;330(6009):1397–400.

    Article  CAS  PubMed  Google Scholar 

  31. Pin PA, Zhang W, Vogt SH, Dally N, Büttner B, Schulze-Buxloh G, et al. The role of a pseudo-response regulator gene in life cycle adaptation and domestication of beet. Curr Biol. 2012;22(12):1095–101.

    Article  CAS  PubMed  Google Scholar 

  32. Dally N, Xiao K, Holtgräwe D, Jung C. The B2 flowering time locus of beet encodes a zinc finger transcription factor. Proc Natl Acad Sci. 2014;111(28):10365–70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Reeves PA, He Y, Schmitz RJ, Amasino RM, Panella LW, Richards CM. Evolutionary Conservation of the FLOWERING LOCUS C-Mediated Vernalization Response: Evidence From the Sugar Beet (Beta vulgaris). Genetics. 2007;176(1):295-307.

  34. Hanson SJ, Dawson JC, Goldman IL. Beta vulgaris ssp. vulgaris chromosome 8 shows significant association with geosmin concentration in table beet. G3 Genes|Genomes|Genetics. 2021;11(12):jkab344.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Richards CM, Reeves PA, Fenwick AL, Panella L. Genetic structure and gene flow in Beta vulgaris subspecies maritima along the Atlantic coast of France. Genetic Res Crop Evol. 2014;61(3):651-62.

  36. Boudry P, Mccombie H, Van Dijk H. Vernalization requirement of wild beet Beta vulgaris ssp. maritima: among population variation and its adaptive significance. J Ecol. 2002;1:693–703.

    Article  Google Scholar 

  37. Bartsch D, Lehnen M, Clegg J, Pohl-Orf M, Schuphan I, Ellstrand NC. Impact of gene flow from cultivated beet on genetic diversity of wild sea beet populations. Mol Ecol. 1999;8(10):1733–41.

    Article  CAS  PubMed  Google Scholar 

  38. Mangin B, Sandron F, Henry K, Devaux B, Willems G, Devaux P, et al. Breeding patterns and cultivated beets origins by genetic diversity and linkage disequilibrium analyses. Theor Appl Genet. 2015;128(11):2255–71.

    Article  PubMed  Google Scholar 

  39. Andrello M, Henry K, Devaux P, Desprez B, Manel S. Taxonomic, spatial and adaptive genetic variation of Beta section Beta. Theor Appl Genet. 2016;129:257–71.

    Article  CAS  PubMed  Google Scholar 

  40. Nottingham S. Beetroot. Self published. 2004.

  41. Fischer HE. Origin of the ‘Weisse Schlesische Rübe’(white Silesian beet) and resynthesis of sugar beet. Euphytica. 1989;41(1–2):75–80.

    Article  Google Scholar 

  42. Cai X, Sun X, Xu C, Sun H, Wang X, Ge C, et al. Genomic analyses provide insights into spinach domestication and the genetic basis of agronomic traits. Nat Commun. 2021;12(1):7246.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Wei T, van Treuren R, Liu X, Zhang Z, Chen J, Liu Y, et al. Whole-genome resequencing of 445 Lactuca accessions reveals the domestication history of cultivated lettuce. Nat Genet. 2021;53(5):752–60.

    Article  CAS  PubMed  Google Scholar 

  44. Huang X, Kurata N, Wei X, Wang Z-X, Wang A, Zhao Q, et al. A map of rice genome variation reveals the origin of cultivated rice. Nature. 2012;490(7421):497–501.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Du Y, Scheres B. Lateral root formation and the multiple roles of auxin. J Exp Bot. 2018;69(2):155–67.

    Article  CAS  PubMed  Google Scholar 

  46. Wang J, Yan D-W, Yuan T-T, Gao X, Lu Y-T. A gain-of-function mutation in IAA8 alters Arabidopsis floral organ development by change of jasmonic acid level. Plant Mol Biol. 2013;82(1):71–83.

    Article  CAS  PubMed  Google Scholar 

  47. Jammer A, Albacete A, Schulz B, Koch W, Weltmeier F, van der Graaff E, et al. Early-stage sugar beet taproot development is characterized by three distinct physiological phases. Plant Direct. 2020;4(7):e00221.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Zhang J, Eswaran G, Alonso-Serra J, Kucukoglu M, Xiang J, Yang W, et al. Transcriptional regulatory framework for vascular cambium development in Arabidopsis roots. Nat Plants. 2019;5(10):1033–42.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Andrews S. FastQC: a quality control tool for high throughput sequence data. Cambridge, UK: Babraham Bioinformatics, Babraham Institute 2010.

  50. Ewels P, Magnusson M, Lundin S, Käller M. MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics. 2016;32(19):3047–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Li H, Durbin R. Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics. 2010;26(5):589–95.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25(16):2078–9.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Van der Auwera GA, Carneiro MO, Hartl C, Poplin R, Del Angel G, Levy-Moonshine A, et al. From FastQ data to high confidence variant calls: the genome analysis toolkit best practices pipeline. Curr Protoc Bioinform. 2013;43:11 0 1-0 33.

    Google Scholar 

  55. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–303.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. McLaren W, Gil L, Hunt SE, Riat HS, Ritchie GRS, Thormann A, et al. The ensembl variant effect predictor. Genome Biol. 2016;17(1):122.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, et al. Circos: an information aesthetic for comparative genomics. Genome Res. 2009;19(9):1639–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  58. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Zheng X, Levine D, Shen J, Gogarten SM, Laurie C, Weir BS. A high-performance computing toolset for relatedness and principal component analysis of SNP data. Bioinformatics. 2012;28(24):3326–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Alexander DH, Lange K. Enhancements to the ADMIXTURE algorithm for individual ancestry estimation. BMC Bioinformatics. 2011;12(1):246.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Behr AA, Liu KZ, Liu-Fang G, Nakka P, Ramachandran S. Pong: fast analysis and visualization of latent clusters in population genetic data. Bioinformatics. 2016;32(18):2817–23.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Zhang C, Dong S-S, Xu J-Y, He W-M, Yang T-L. PopLDdecay: a fast and effective tool for linkage disequilibrium decay analysis based on variant call format files. Bioinformatics. 2019;35(10):1786–8.

    Article  CAS  PubMed  Google Scholar 

  63. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and VCFtools. Bioinformatics. 2011;27(15):2156–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  64. Chen H, Patterson N, Reich D. Population differentiation as a test for selective sweeps. Genome Res. 2010;20(3):393–402.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Schiffels S, Wang K. MSMC and MSMC2: the multiple sequentially markovian coalescent. Methods Mol Biol. 2020;2090:147–66.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We thank Paul Galewski (Department of Plant, Soil, and Microbial Science, Michigan State University), for providing the EL10.2_2 genome assembly. We thank Brigitte Neidhardt-Olf, Bettina Rohardt, Ines Schütt, and Monika Bruisch for their technical assistance.

Funding

Open Access funding enabled and organized by Projekt DEAL. We gratefully acknowledge financial support from the German Science Foundation (Deutsche Forschungsgemeinschaft, DFG) – Project Number 400993799 (Project 2 within the Research Training Group 2501 Translational Evolutionary Research, https://gepris.dfg.de/gepris/projekt/400993799). The informatics infrastructure was supported by the BMBF-funded de.NBI Cloud, part of the German Network for Bioinformatics (de.NBI).

Author information

Authors and Affiliations

  1. Plant Breeding Institute, Christian-Albrechts-University of Kiel, Olshausenstr. 40, 24098, Kiel, Germany

    Amar Singh Dhiman, Nazgol Emrani & Christian Jung

  2. Max Planck Institute for Evolutionary Biology, August-Thienemann-Str. 2, 24306, Plön, Germany

    Demetris Taliadoros & Eva H. Stukenbrock

  3. Botanical Institute, Christian-Albrechts-University of Kiel, Am Botanischen Garten 1-9, 24118, Kiel, Germany

    Eva H. Stukenbrock

  4. USDA-ARS, Sugarbeet and Bean Research Unit, East Lansing, MI, USA

    J. Mitchell McGrath

  5. Present Address: KWS SAAT SE & Co. KGaA, Grimsehlstr. 31, 37574, Einbeck, Germany

    Amar Singh Dhiman

  6. Department of Medical Biochemistry and Microbiology, University of Uppsala, Uppsala, Sweden

    Demetris Taliadoros

Authors
  1. Amar Singh Dhiman
    View author publications

    Search author on:PubMed Google Scholar

  2. Demetris Taliadoros
    View author publications

    Search author on:PubMed Google Scholar

  3. Eva H. Stukenbrock
    View author publications

    Search author on:PubMed Google Scholar

  4. J. Mitchell McGrath
    View author publications

    Search author on:PubMed Google Scholar

  5. Nazgol Emrani
    View author publications

    Search author on:PubMed Google Scholar

  6. Christian Jung
    View author publications

    Search author on:PubMed Google Scholar

Contributions

ASD designed the experiments, produced the data, wrote and revised the manuscript. DT helped conduct the bioinformatics analysis and was involved in writing and revision of the manuscript. EHS, MMG, and NE were involved in writing and revision of the manuscript. CJ acquired the funding, supervised the design of the study, and was involved in writing and revision of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Christian Jung.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All the authors of the manuscript consent to its publication as a research article in BMC Biology.

Competing interests

The authors declare that they have 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

12915_2025_2422_MOESM1_ESM.zip

Additional file 1: Table S1: Summary statistics of the paired-end Illumina raw reads without trimming from 290 accessions. Table S2: Summary statistics of the paired-end Illumina trimmed and cleaned reads from 290 accessions. Table S3: Summary statistics of the sample information, total mapped reads (%), sequencing depth and genome coverage (%) across 290 accessions. “NA” : Not Available. Table S4: Summary statistics of sequence variants (SNPs and INDELs) in 290 beet accessions compared to the EL10 reference genome. EL10.2_2 genome assembly comprises 568.8 Mb and is assembled into 18 pseudomolecules. In total, 564.2 Mb (99.2%) of the total genome assembly is organized into 9 main chromosomes (pseudomolecules), and the remaining 4.6 Mb of the genome is placed on 9 small unscaffolded pseudomolecules. Fig. S1: A bar plot illustrating the number of variants, including SNPs and INDELs, across all nine EL10 chromosomes in the Beta mini-core collection of 290 accessions. The Y-axis represents the total number of variants, and the X-axis corresponds to the nine EL10 chromosomes. Chromosome 9 exhibits the minimum number of variants, approximately 1.1 million, while Chromosome 5 displays the maximum number, around 1.5 million variants. On average, each chromosome harbors approximately 1.3 million variants. Table S5: Summary statistics of the number of SNPs and INDELs with their predicted effect using VEP across 290 accessions. Table S6: Summary statistics of the number of SNPs and INDELs within coding sequences with their predicted effect using VEP across 290 accessions. Fig. S2: Principal component analysis (PCA) of 290 beet accessions comparing second and third principal components. PC2 and PC3 represent the second and third components, accounting for 3.08% and 2.50% of the total variation, respectively. The colors represent different genetic clusters according to their beet type, with the following accessions: wild accession from the Atlantic (purple), Mediterranean (blue), sugar beet (dark green), table beet (red), fodder beet (orange), and leaf beet (light green). Fig. S3: Principal component analysis (PCA) of 199 cultivated beet accessions, including sugar, table, fodder, and leaf beet accessions. PC1 and PC2 denote the first and second components, explaining 4.39% and 3.25% of the total variation, respectively. Leaf beet accessions are grouped into GP1 (biennial leaf beet) and GP2 (annual leaf beet) based on phenotypic classification. Although the grouping is phenotype-based, the two groups also exhibited differing levels of sugar beet and Mediterranean sea beet ancestry at the admixture level. Fig. S4: Cross-validation errors for each K in 10 replicates tested using model-based clustering in ADMIXTURE, incorporating varying numbers of ancestral kinships (K=1 and K=10). The Y-axis represents the cross-validation error value, while the X-axis corresponds to the Kth ancestry. The plot represents the cross-validation errors from 10 replicates at each K. Notably, the minimum cross-validation error and the variation within the 10 replicates were observed at K = 4. Table S7: Summary statistics of sequence variants (SNPs and INDELs) within each sub-cluster compared to the EL10 reference genome. Table S8: Pairwise shared variants between the Mediterranean sea beet accessions and the other sub-clusters. Table S9: Pairwise shared variants between the Atlantic sea beet accessions and the other sub-clusters. Fig. S5: Linkage disequilibrium (LD) decay among distinct genetic clusters of the Beta mini-core collection across all nine chromosomes. Accessions with low admixture are depicted in black. LD is plotted across nine chromosomes (1 to 9). Within each plot, the Y-axis represents the pairwise correlation coefficient (r2) between two SNP markers, and the X-axis represents the physical distance (kb) between corresponding SNP markers. Fig. S6: Linkage disequilibrium (LD) decay comparisons among nine chromosomes within each Beta mini-core collection genetic cluster. LD decay comparisons are depicted within sugar beet, table beet, fodder beet, leaf beet, Mediterranean sea beet, Atlantic sea beet, and 245 accessions with low admixture from other genetic ancestries. LD is collectively plotted across all nine chromosomes (1 to 9) within each distinct genetic cluster. In each plot, the Y-axis represents the pairwise correlation coefficient (r2) between two SNP markers, and the X-axis represents the physical distance (kb) between corresponding SNP markers. The colors marked in the legend represent nine chromosomes. Table S10: Summary statistics of the percentage of heterozygous SNPs in each accession. Fig. S7: The frequency distribution plots of the percentage of heterozygous SNPs within distinct genetic clusters. The frequency distribution of heterozygous SNPs within each genetic cluster, such as Mediterranean sea beet, leaf beet, fodder beet, sugar beet, table beet, and Atlantic sea beet, is plotted. Six classes were calculated, ranging from 5-10%, 10-15%, 15-20%, 20-25%, 25-30% and 30-35%. In each plot, the Y-axis represents the frequency of individual accessions belonging to the representative classes, plotted on the X-axis. The colors distinguish between the six different classes. Table S11: Summary statistics of pairwise comparisons of population differentiation (FST) between the Atlantic sea beet accessions and the other sub-clusters. Table S12: Summary statistics of nucleotide diversity within each genetic sub-cluster. Fig. S8: XP-CLR revealed selection sweeps in sugar beets across all nine EL10 chromosomes. XP-CLR across the nine EL10 chromosomes using 10 kb sliding window size and 1 kb step size while comparing sugar beet accessions with the Mediterranean sea beet accessions. The blue dots indicate XP-CLR values along the nine EL10 chromosomes, while the horizontal red line denotes the empirical 95th percentile threshold. According to XP-CLR, genomic regions above this threshold are considered putative regions under selection. The Y-axis represents the XP-CLR values ranging from 0 to 1000, and the X-axis illustrates the chromosome length in Mb. Fig. S9: Population differentiation using fixation index (FST) between sugar beet and Mediterranean sea beet accessions across all nine EL10 chromosomes. The FST calculations employed 10 kb sliding window size and 1 kb step size while comparing sugar beet accessions with the Mediterranean sea beet accessions. The blue dots on the graphs indicate FST values along the nine EL10 chromosomes, while the horizontal red line represents the empirical 95th percentile threshold. Genomic regions above this threshold are considered highly differentiated regions between sugar beet and Mediterranean sea beet, as indicated by FST. The Y-axis represents FST values ranging from 0 to 1, where 0 signifies no genetic differentiation and 1 indicates complete genetic differentiation between the two populations. The X-axis illustrates the chromosome length in Mb. Fig. S10: Nucleotide diversity values within various genetic clusters, including leaf and fodder beets, in the regions of interest encompassing selective sweeps, with a specific focus on genes involved in cold tolerance during post-harvest storage and the vernalization pathway in flowering time control. The colors represent different genetic clusters based on beet type, representing the wild accession from the Atlantic (purple) and the Mediterranean (blue), sugar beet (dark green), table beet (red), fodder beet (orange), and leaf beet (light green) accessions. The Y-axis illustrates the nucleotide diversity value, while the X-axis denotes the location along the chromosome. The vertical red lines represent the gene boundaries of the predicted gene models. The nucleotide diversity values across different clusters are graphically represented for (A) inositol transporter 1 on chromosome 2, (B) raffinose synthase 5 on chromosome 1, (C) FLOWERING LOCUS C (FLC) on chromosome 6, and (D) SUPPRESSOR OF OVEREXPRESSION OF CO 1 (SOC1) on chromosome 1. Fig. S11: Nucleotide diversity values within various genetic clusters, including leaf and fodder beets, in the regions of interest encompassing selective sweeps, with a specific focus on genes associated with earthy flavor and plant immunity. The colors represent different genetic clusters based on beet type, representing the wild accession from the Atlantic (purple) and the Mediterranean (blue), sugar beet (dark green), table beet (red), fodder beet (orange), and leaf beet (light green) accessions. The Y-axis illustrates the nucleotide diversity value, while the X-axis denotes the location along the chromosome. The vertical red lines represent the gene boundaries of the predicted gene models. The nucleotide diversity values across different clusters are graphically represented for (A) terpene synthase on chromosome 7, (B) terpene synthase on chromosome 8, (C) WUSCHEL-RELATED HOMEOBOX 4 (WOX4) gene on chromosome 1. Fig. S12: Estimated historical effective population size (Ne) of genetic clusters in the Beta mini-core collection using the program MSMC2. The X-axis represents the years on a logarithmic scale, while the Y-axis represents the effective population size. A mutation rate of 1.25e-8 was used. The colors represent different genetic clusters according to their beet type: wild accession from the Atlantic (purple), wild accession from the Mediterranean (blue), sugar beet (dark green), table beet (red), fodder beet (orange), and leaf beet (light green) accessions. Table S13: Description of sea beet and cultivated beet accessions used in this study. “NA” indicates data that is Not Available. Fig. S13: Overview of the bioinformatics workflow used in the study. The Illumina short-reads were mapped to the EL10.2_2 genome assembly using BWA-MEM. Note S1: Custom R script used for calculating and plotting PCA and heterozygosity. The script includes data processing steps, calculation of heterozygosity percentages, and the complete ggplot2 code used to visualize the results.

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/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Dhiman, A., Taliadoros, D., Stukenbrock, E.H. et al. Signatures in domesticated beet genomes pointing at genes under selection in a sucrose-storing root crop. BMC Biol 23, 299 (2025). https://doi.org/10.1186/s12915-025-02422-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12915-025-02422-5

Keywords

BMC Biology

ISSN: 1741-7007

Contact us