Skip to main content
NCBI home page
BMC Plant Biology logoLink to BMC Plant Biology
. 2026 May 30;26:1293. doi: 10.1186/s12870-026-09151-z

Identification and transcriptome analysis of a major locus for eye depth in tetraploid potato Jinshu 16

Huijie Wang 1,#, Jin Pu 3,5,#, Rongyan Wang 3,#, Zhong Zhang 5, Wei Zhou 2, Xiaodong Bai 4, Liguang Huo 1, Hui Du 2,✉
PMCID: PMC13430739  PMID: 42218392

Abstract

Tuber eye depth is a key agronomic trait in potato, influencing both processing efficiency and market appearance. To investigate the genetic basis of deep-eye phenotype in the elite Chinese tetraploid cultivar Jinshu 16, we constructed an F1 segregating population by crossing Jinshu 16 (deep eye) with Atlantic (shallow eye). Phenotypic evaluation across two seasons revealed a 1:1 segregation ratio for eye depth (Eyd), indicative of control by a single dominant locus. By integrating whole-genome resequencing and bulk segregant analysis (BSA-seq) of extreme phenotypes, we fine-mapped the major locus to a 49.0–50.7 Mb interval on chromosome 10, co-localizing with the previously reported Eyd locus. Cytological examination showed increased cell density in the eye region of Jinshu 16 tubers. Transcriptome analysis identified specific upregulation of several lipid transfer protein and peroxidase genes within the mapped region in Jinshu 16 eye tissues, implicating lipid metabolism and cell wall modification in eye depth determination. Developed Insertion‑Deletion (InDel) markers from this region were significantly associated with the phenotype in the population. Our study defines the genetic locus responsible for the deep-eye trait in Jinshu 16, providing a foundation for gene functional studies and molecular marker-assisted breeding in this cultivar.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-09151-z.

Keywords: Potato, Eye depth, BSA-seq, Transcriptome, Molecular marker

Introduction

Potato (Solanum tuberosum L.) is an important tuber crop. The eyes on the tuber surface serve as a key morphological trait with dual biological and economic significance: they are involved in seed tuber sprouting and plant regeneration while also directly affecting the appearance quality and processing suitability of table potatoes. Notably, deep-eyed varieties generally exhibit high losses and low efficiency during processing, and their irregular tuber shape often fails to meet the consistency requirements of modern markets. As a result, breeding varieties with shallow eyes and regular tuber shapes has become an urgent industry need. To achieve this goal, systematically elucidating the genetic regulatory mechanisms of eye depth (Eyd) and accurately identifying key genes are of decisive importance for establishing a molecular design breeding system in potato and comprehensively enhancing industrial processing efficiency.

Early genetic studies on potato eye depth primarily relied on phenotypic assessments and classical linkage analysis, providing foundational but limited insights into its hereditary patterns [1, 2]. With the development of molecular marker technologies, particularly the application of linkage mapping and genome-wide association studies (GWAS), genetic research on this trait entered a new phase [3–8]. Studies have shown that eye depth is a quantitative trait controlled by major genes and modified by minor genes and environmental factors [6, 8–10]. Li et al. first mapped the major locus Eyd controlling eye depth to chromosome 10 in a diploid population, noting its close linkage with the tuber shape gene Ro, thereby establishing the genetic foundation of the major QTL for this trait [11]. Subsequent studies consistently confirmed the major role of this region across diploid and tetraploid populations with different genetic backgrounds [3, 12–14], revealing a monogenic or oligogenic control pattern still influenced by minor QTLs [9, 15]. In recent years, high-throughput sequencing-based approaches such as bulked segregant analysis (BSA-seq) and GWAS have further refined this locus to specific genomic intervals [16], laying important groundwork for candidate gene mining. However, research on the genetic mechanisms of eye depth in China’s main cultivated varieties, particularly tetraploid materials, remains relatively limited. The cloning and functional validation of key genes are still insufficient, which to some extent hinders the effective utilization of this trait in molecular breeding.

Jinshu 16 is a widely promoted mid-to-late maturing tetraploid potato variety in northern China, known for its outstanding advantages such as high and stable yield, strong stress resistance, and broad adaptability [17–19]. It is extensively cultivated in Shanxi and surrounding regions, making significant contributions to regional food supply and farmers’ income. However, the relatively deep-eye phenotype of this variety’s tubers somewhat affects its commercial appearance and processing potential, limiting its further promotion and application in high-end fresh markets and deep-processing industries. Therefore, conducting genetic analysis of the eye depth trait in Jinshu 16 and mapping its key regulatory loci hold practical significance for the targeted genetic improvement of this variety and the breeding of new lines that meet market demands.

Based on this background, this study used the deep-eyed variety Jinshu 16 as the female parent and crossed it with the shallow-eyed variety Atlantic to construct an F1 segregating population. Two-season precise phenotyping was conducted to analyze the inheritance pattern of eye depth. An integrated approach utilizing genome resequencing and bulked segregant analysis (BSA-seq) was employed for rapid gene mapping of extreme phenotypes. Furthermore, transcriptome sequencing of parental lines at key developmental stages of different tissues was performed to screen differentially expressed genes and predict candidate genes within the mapped interval. Finally, molecular markers were developed within the candidate region to validate their association with phenotypes. The aim is to preliminarily elucidate the genetic basis of the deep-eye trait in Jinshu 16, providing theoretical insights and technical support for subsequent functional validation of the gene and molecular marker-assisted selection breeding of this variety.

Materials and methods

Plant materials and population construction

The tetraploid, medium-deep eye potato cultivar Jinshu 16 (J16) was provided by the Institute of High Latitude Crops, Shanxi Agricultural University. The tetraploid, shallow eye depth cultivar Atlantic (Atl), developed by the USDA Agricultural Research Service in 1978 from the cross Wauseon × Lenape, is maintained by the Potato Breeding Laboratory, College of Agriculture, Shanxi Agricultural University. An F1 segregating population was developed by crossing Jinshu 16 as the female parent with Atlantic. Seedlings from the F1 population were first grown in Dehong, Yunnan Province, during the winter of 2023 to produce microtubers. These microtubers were subsequently planted in Taiyuan, Shanxi Province, in the spring (April) of 2024 and harvested in October 2024. The harvested tubers were then planted in Honghe, Yunnan Province, in the winter (December) of 2024 and harvested in April 2025 for subsequent phenotypic and molecular analyses.

Phenotype evaluation and statistical analysis

Following physiological maturity, tubers of the F1 population (J16 × Atl) harvested from Taiyuan (spring–autumn 2024) and Honghe (winter 2024–spring 2025) were phenotyped for eye depth. To ensure consistency, visual assessment was conducted on 4–10 tubers per plant line. Eye depth was graded into five levels: smooth (level 1), shallow (level 2), slightly deep (level 3), deep (level 4), and very deep (level 5). For subsequent genetic analysis of the deep-eye phenotype in J16, levels 1–2 were classified as “shallow eye,” and levels 3–5 as “deep eye.”

Histological and cellular analysis

To explore the cellular mechanisms underlying eye depth differences between Jinshu 16 (J16, medium-deep eye) and Atlantic (shallow eye), histological comparisons were conducted on tuber eye tissues. Samples were fixed, dehydrated, cleared in xylene, embedded in paraffin, and sectioned at 4 μm. After dewaxing and hydration, sections were stained with 1% Fast Green, dehydrated, cleared, and mounted. Observations were made under a light microscope. For cellular density quantification, eight 1 mm² regions were randomly selected in both eye and adjacent non-eye parenchyma tissues of J16 and Atlantic.

Genome resequencing and genotyping

Genomic DNA was extracted from fresh leaves of the parental lines J16 and Atl using a modified cetyltrimethylammonium bromide (CTAB) method [20]. Whole‑genome resequencing of both parents was performed on the BGI MGISEQ‑T7 platform, with approximately 100 Gb of 150‑bp paired‑end reads generated for each parent. From the F1 segregating population, 20 plants with deep eye and 20 plants with shallow eye phenotypes were selected and individually sequenced, yielding approximately 20 Gb of clean data per plant. After quality control and adapter trimming using fastp [21], clean reads from both parents were aligned to the potato reference genome DMv6.01 [22, 23] using BWA‑MEM [24]. SNP calling was performed using GATK Haplotype Caller (with ploidy set to 4) [25] and BCFtools [26], followed by filtration based on quality (QUAL ≥ 30), mapping quality (MQ ≥ 35), and read depth (DP between 160 and 1500). SNPs were selected based on the expected simplex dominant (Aaaa) inheritance in J16 and homozygous recessive (aaaa) state in Atlantic. J16 SNPs were required to exhibit an allele frequency near 0.25 (0.20–0.30) or 0.75 (0.70–0.80), while Atlantic SNPs were required to be nearly homozygous (< 0.10 or > 0.90). These thresholds accommodate sequencing variability while reflecting the underlying genetic model.

BSA-seq analysis of the Eyd locus

To map the Eyd locus, sequence data from the 20 deep‑eye and 20 shallow‑eye individuals were separately pooled to form a “deep pool” and a “shallow pool.” For each pool, SNPs were called using GATK HaplotypeCaller with the same parameters as applied to the parents, requiring a minimum depth of 10× per allele and QUAL ≥ 30. Only SNPs that were polymorphic between the parental haplotypes were retained for downstream analysis. SNP‑index values for each pool were calculated as the ratio of reads supporting the alternative allele of the Atlantic haplotype to the total reads covering that position.

The delta SNP‑index was calculated as the absolute difference between the SNP‑index of the deep pool and that of the shallow pool for each window [27]. A Monte Carlo simulation tailored for the tetraploid simplex model (Aaaa × aaaa) was performed under the null hypothesis (expected allele frequency 0.25 for both pools); based on 10,000 iterations, the 95% and 99% confidence thresholds for Delta SNP‑index were determined as 0.185 and 0.225, respectively. The sliding-window approach (window size: 1 Mb; step: 100 kb) was then applied to average SNP‑index values across each chromosome. Candidate intervals associated with Eyd variation were defined as those where the delta SNP‑index consistently exceeded the 95% confidence threshold (0.185). Finally, candidate regions were visualized using R scripts (Supplementary Material 1) to generate Manhattan‑like plots of SNP‑index and delta SNP‑index profiles along the 12 potato chromosomes.

Transcriptome sequencing and analysis

Total RNA was extracted from eye and peel tissues of tubers at mid‑development stage (5 cm diameter) from parental lines J16 and Atlantic (Atl), with three biological replicates per tissue‑genotype combination. RNA integrity was assessed using an Agilent 2100 Bioanalyzer, and all samples had RIN values greater than 7. Transcriptome sequencing was performed by Annoroad Gene Technology (Beijing, China) on the BGI MGISEQ‑T7 platform, yielding approximately 4 Gb of raw data per sample. After quality control and adapter trimming using fastp [21], clean reads were aligned to the potato reference genome DMv6.1 with HISAT2 [28], and gene expression levels (FPKM and read counts) were quantified using StringTie [29].

Differential expression analysis was conducted using DESeq2 based on read counts for four comparisons: J16_Eye vs. J16_Peel, Atl_Eye vs. Atl_Peel, J16_Eye vs. Atl_Eye, and J16_Peel vs. Atl_Peel. Genes with |log₂ fold change| ≥ 1 and adjusted p‑value < 0.05 were considered differentially expressed. GO and KEGG enrichment analyses were performed on the differentially expressed gene sets from J16_Eye vs. J16_Peel and J16_Eye vs. Atl_Eye.

Candidate gene analysis in the Eyd locus

Based on the transcriptome data, the FPKM values of genes located within the fine‑mapped Eyd candidate region (DMv6.1 Chr10: 49–50.7 Mb, defined by windows with delta SNP‑index ≥ 0.2) were extracted. K‑means clustering (k = 10) was carried out using the Metware Cloud platform (https://cloud.metware.cn) to group candidate genes by expression pattern. Clusters 3, 4, and 6, which exhibited elevated expression in J16 eye tissues, were selected for further analysis. Genes with an average FPKM > 1 in at least one tissue group were retained, and expression heatmaps were generated using TBtools for visual evaluation [30].

Molecular marker development and population validation

Based on resequencing data and variant calling results (VCF/BAM files) of J16 and Atlantic, eleven Insertion‑Deletion (InDel) markers (Table S1) were designed within the candidate region (Chr10: 49–50.7 Mb) [31], targeting sites where J16 was heterozygous (Indel index between 0.70 and 0.80) and Atlantic was homozygous (Indel index > 0.90).

PCR products were separated by 8% polyacrylamide gel electrophoresis (PAGE) and visualized by silver nitrate staining followed by development with NaOH and formaldehyde. Gels were photographed using a digital camera. The captured images were then converted to black‑and‑white and cropped for presentation; no brightness or contrast adjustments were applied. Screening identified two markers (J16_Indel06 and J16_Indel09) that showed clear banding patterns and conformed to expected genotypes. These markers were used to genotype the F1 population (J16 × Atl). Genotype‑phenotype associations were assessed by correlating marker data with eye depth scores (deep vs. shallow) collected over two growing seasons.

Results

Phenotypic characterization and inheritance analysis of tuber eye depth

Phenotypic observation revealed that the tubers of tetraploid cultivar Jinshu 16 (J16) exhibited significantly deeper eye depression compared to the Atlantic variety (Fig. 1A, B). To investigate the cytological basis of this difference, paraffin sectioning was performed. Microscopic comparison showed that cells in the eye region of J16 were more densely arranged, with a higher cell number per unit area than in adjacent non-eye tissues (Fig. 1C-I). In contrast, the cell density in the eye region of Atlantic was essentially consistent with that in other parts of the tuber. These findings suggest that the deep-eye phenotype may result from relatively delayed cell development and tighter tissue organization in the eye region.

Fig. 1.

Fig. 1

Open in a new tab

Eye phenotypes and population distribution in Jinshu 16 (J16) and Atlantic (Atl). A–B Representative tuber images of J16 and Atl. Scale bar, 10 cm. C–H Histological sections of eye tissues from J16 (C, E, G) and Atl (D, F, H). Scale bar, 1 mm (C, D); 200 μm (E–H). I Quantification of cell density (cells per 1 mm²) in peel and eye tissues of J16 and Atl. Data are shown as mean ± SD (n = 8). Asterisks indicate significant differences (***P < 0.001, ns, not significantly; Student’s t-test). J-N Eye-depth visual scale from level 1 (very shallow eye) to 5 (very deep eye). Scale bar, 2 cm. O Distribution of eye-depth grades (levels 1–5) in the F1 population (J16 × Atl) across two growing seasons (Taiyuan, Shanxi, spring–autumn 2024 and Honghe, Yunnan, winter 2024–spring 2025)

To explore the genetic mechanism, an F1 population was constructed by crossing J16 × Atlantic. After obtaining microtubers through winter propagation in Dehong, Yunnan, the population was planted for phenotypic investigation over two consecutive growing seasons: in Taiyuan, Shanxi (planted April 2024, harvested October 2024) and in Honghe, Yunnan (planted December 2024, harvested April 2025). Eye depth was classified into five grades based on phenotypic variation, with Atlantic assigned as grade 1 and J16 as grade 4 (Fig. 1J-N). The eye depth of 210 F1 individuals showed a continuous distribution from grade 1 to grade 5, with generally consistent trends across the two years (Fig. 1O, Table S2). In the 2024–2025 winter–spring population in Honghe, slightly more individuals were graded as 1, while grades 2 and 3 accounted for a higher proportion in both seasons.

For genetic analysis, grades 1–2 were categorized as shallow-eye and grades 3–5 as deep-eye phenotypes (Table S2). The segregation ratio in the F1 population was tested against a 1:1 ratio (Taiyuan 2024: 113 shallow: 97 deep, χ² = 1.2190, P = 0.2695; Honghe 2024–2025: 117 shallow: 93 deep, χ² = 2.7429, P = 0.0976). In both years, the chi-square tests showed no significant deviation from the expected 1:1 ratio (P > 0.05), suggesting that the eye depth trait is controlled by a single dominant major locus. Considering the tetraploid genetic background of J16, it is hypothesized that this locus may exist in a simple dominant (Aaaa) configuration in J16.

Genetic mapping of the eye-depth locus

To identify the Eyd locus in J16, 20 deep‑eye and 20 shallow‑eye individuals were selected from the F1 population (J16 × Atlantic). Whole‑genome resequencing (approximately 20 Gb per individual) was performed for each selected plant, and the resulting sequence data were merged to form a deep pool and a shallow pool, respectively. The resequencing data of both parental lines (J16 and Atlantic) were aligned to the potato reference genome DMv6.1.

SNPs were selected based on the expected simplex dominant (Aaaa) inheritance in J16 and homozygous recessive (aaaa) state in Atlantic. J16 SNPs were required to exhibit an allele frequency near 0.25 (0.20–0.30) or 0.75 (0.70–0.80), while Atlantic SNPs were required to be nearly homozygous (< 0.10 or > 0.90). These thresholds accommodate sequencing variability while reflecting the underlying genetic model. Sequence reads from the two bulk pools were similarly aligned, and only SNPs meeting the above parental haplotype criteria were extracted for each pool. The SNP‑index for each pool was then calculated as the ratio of reads supporting the alternative allele of the Atlantic haplotype to the total reads covering that position.

Given that the deep‑eye locus in J16 is presumed to be simplex dominant (Aaaa), deep‑eye individuals in the F1 population are expected to carry one copy of the dominant allele. Consequently, the SNP‑index in the deep pool was predicted to be approximately 0.25 (or 0.75 depending on haplotype phasing). In contrast, Atlantic is homozygous recessive (aaaa) at this locus; thus, the shallow pool was expected to show SNP‑index values close to 0 or 1. Sliding‑window analysis (window size: 1 Mb; step: 100 kb) was applied to compute the delta SNP‑index between the two pools (Fig. 2, Table S4). The Monte Carlo simulation established a 95% confidence threshold of 0.185 and a 99% threshold of 0.225. The analysis revealed a single significant association peak on chromosome 10, with delta SNP‑index values reaching a maximum of 0.212 (Fig. 2). This observed peak significantly exceeded the 95% confidence threshold (0.185) and approached the 99% threshold (0.225), representing approximately 85% of the theoretical maximum (0.25) for a simplex‑linked locus in a tetraploid. The interval spanning approximately 49.0–50.7 Mb (Table S4, S5) showed delta SNP‑index values consistently above 0.185. This result indicates that the major Eyd locus in J16 co-locates with the previously reported Eyd locus, confirming the genetic consistency of the deep‑eye phenotype in this cultivar.

Fig. 2.

Fig. 2

Open in a new tab

Bulked-segregant analysis sequencing (BSA-seq) of the Eyd locus in the F1 population (J16 × Atl). A SNP-index profiles for the deep and shallow pools aligned to the DM (v6.1) reference genome. B Delta SNP-index profile obtained by subtracting the shallow-pool SNP-index from the deep-pool SNP-index. The Eyd locus is indicated by a blue arrow and localized to the distal region of chromosome 10. The x-axis represents the 12 potato chromosomes, separated by vertical dotted lines; the horizontal dotted line denotes the 95% confidence threshold (0.185) derived from Monte Carlo simulation under the tetraploid simplex inheritance model. The observed peak delta SNP‑index (0.212) significantly exceeds this threshold and reaches approximately 85% of the theoretical maximum (0.25)

Transcriptome analysis of tuber eye tissues

To further identify candidate pathways and genes potentially regulating the development of the tuber eye in potato, transcriptome sequencing was performed on tuber eye and peel tissues at mid-development stage (5 cm diameter) from parental lines J16 and Atlantic (Atl). Clean reads from all samples were aligned to the potato reference genome DMv6.1, with an average total mapping rate of 91.7% (ranging from 90.4% to 92.6%; Table S6). Gene expression levels were normalized using Fragments Per Kilobase of transcript per Million mapped reads (FPKM). Principal component analysis (PCA) based on FPKM values confirmed strong reproducibility among biological replicates (Fig. 3A).

Fig. 3.

Fig. 3

Open in a new tab

Transcriptome sequencing and differential gene expression analysis of tuber eye and peel tissues from Jinshu 16 and Atlantic. A Principal component analysis (PCA) of transcriptomic data across replicates. B Number of differentially expressed genes (DEGs) in four pairwise comparisons. C Venn diagram illustrating the overlap of DEGs among the four comparisons. D–E Gene Ontology (GO) enrichment analysis of DEGs from J16_Eye vs. Atl_Eye (D) and J16_Eye vs. J16_Peel (E). F–G KEGG enrichment analysis of DEGs from J16_Eye vs. Atl_Eye (F) and J16_Eye vs. J16_Peel (G)

Differential expression analysis using DESeq2 with the criteria of |log₂ fold change| ≥ 1 and adjusted p‑value < 0.05 identified a total of 9,816 non‑redundant differentially expressed genes (DEGs) from the four pairwise comparisons: J16_Eye vs. J16_Peel, J16_Eye vs. Atl_Eye, Atl_Eye vs. Atl_Peel, and J16_Peel vs. Atl_Peel. Comparisons between genotypes (J16_Peel vs. Atl_Peel and J16_Eye vs. Atl_Eye) yielded a total of 9,139 DEGs, which was substantially higher than the total of 7,715 DEGs from within‑genotype comparisons (J16_Eye vs. J16_Peel and Atl_Eye vs. Atl_Peel) (Fig. 3B, C). These results demonstrate that genetic background (J16 vs. Atl) is the primary determinant of transcriptional variation, exerting a stronger influence on gene expression patterns than tissue‑type differences (eye vs. peel). While tissue‑specific regulation is evident, its contribution is secondary to the genotype effect in shaping the transcriptome during tuber eye development.

To elucidate the molecular mechanisms of the deep‑eye trait in J16, we analyzed two sets of differentially expressed genes (DEGs): between eye tissues of different genotypes (J16_Eye vs. Atl_Eye, reflecting genetic effects) and between eye and peel tissues within J16 (J16_Eye vs. J16_Peel, reflecting tissue‑specific regulation). GO and KEGG enrichment analyses revealed distinct functional profiles. DEGs from J16_Eye vs. Atl_Eye were enriched in oxidation‑reduction processes, oxidoreductase activity, and iron ion binding (Fig. 3D-F), with KEGG pathways including phenylpropanoid biosynthesis and phenylalanine metabolism, all associated with oxidative metabolism and cell wall modification (Fig. 3G). In contrast, DEGs from J16_Eye vs. J16_Peel were enriched in lipid metabolic processes, lipid binding activity, and cellular components such as the cell wall (Fig. 3H-J), with KEGG pathways covering sesquiterpenoid and triterpenoid biosynthesis and ABC transporters, indicative of active lipid metabolism and transport (Fig. 3K). These results suggest that genetic variation drives redox‑related pathways, while tissue‑specific differentiation involves coordinated regulation of lipid metabolism and cell wall organization, together contributing to tuber eye depth formation.

Expression characteristics of candidate genes in the Eyd locus

To further identify candidate genes within the fine‑mapped Eyd interval (Chr10: 49–50.7 Mb, Table S5), expression profiles of all annotated genes in this region were extracted across four tissue groups: J16_Eye, J16_Peel, Atl_Eye, and Atl_Peel. K‑means clustering of their expression patterns divided the genes into 10 distinct clusters (Fig. 4A). Clusters 3, 4, and 6 exhibited elevated expression specifically in J16 eye tissues, comprising 30 genes in total, including 11 lipid transfer protein genes and 4 peroxidase superfamily genes (Table S7). After filtering genes with an average FPKM > 1 in at least one tissue group, 21 high‑confidence candidates were retained, among which 9 encode lipid transfer proteins and 1 encodes a peroxidase family member (Fig. 4B). Notably, three of the lipid transfer protein genes (IDs: Soltu.DM.10G018870, Soltu.DM.10G018810, and Soltu.DM.10G018840) displayed higher expression levels in J16 eye tissue than in the other three samples, which may better align with the expected expression pattern of candidate genes for the deep‑eye trait. These findings suggest that genes involved in lipid metabolism and oxidative regulation within the Chr10 interval could contribute to tuber Eyd variation between J16 and Atlantic. The preferential expression of certain lipid transfer proteins and a peroxidase family gene in J16 eye tissues may indicate their possible involvement in processes such as cell wall modulation, redox balance, or localized signaling during eye development, potentially influencing the formation of the deep‑eye phenotype.

Fig. 4.

Fig. 4

Open in a new tab

Expression pattern analysis of candidate genes in the Eyd locus. A K‑means clustering of all genes located within the Eyd candidate interval across four tissue groups: J16_Eye, J16_Peel, Atl_Eye, and Atl_Peel. Clusters 3, 4, and 6 are highlighted. B Heatmap showing expression profiles (FPKM) of the 21 candidate genes retained from clusters 3, 4, and 6 after filtering (average FPKM > 1 in at least one tissue group). Values represent mean FPKM from three biological replicates

Development and phenotypic validation of InDel markers within the Eyd locus

To validate the accuracy of the mapped Eyd locus in J16 and to develop molecular markers for marker‑assisted selection in J16‑derived populations, eleven Insertion-Deletion (InDel) markers were designed within the candidate region (Chr10: 49–50.7 Mb) based on resequencing data of J16 and Atlantic (Atl). The selected sites were required to be heterozygous (Indel index between 0.70 and 0.80) in J16 and homozygous (Indel index > 0.90) in Atlantic. Screening by PAGE electrophoresis showed that J16_Indel06 and J16_Indel09 produced heterozygous banding patterns in J16 and homozygous patterns in Atl (Fig. 5A, B). In J16, the shared band with Atl was more intense than the alternative band, consistent with the allelic frequency estimated from resequencing data. This pattern suggests that both markers are linked to the simplex dominant Eyd allele in tetraploid J16 and could be used to trace this allele in J16’s progeny. Using these two markers, 210 individuals of the F1 population (J16 × Atl) were genotyped. The segregation of genotypes (Aa vs. aa) fitted a 1:1 ratio (96:103; χ² = 0.246, P = 0.620, Fig. 5C, D, Table S8). Correlation analysis between genotypes and eye‑depth phenotypes collected over two growing seasons (Taiyuan 2024 and Honghe 2024–2025) revealed a statistically significant association in both seasons (χ² test, P < 0.0001, Table S8). These results confirm that phenotypic variation is strongly linked to the marker loci and indicate that the Eyd locus in J16 has been accurately mapped. The developed markers are therefore effective for marker‑assisted selection of the deep‑eye phenotype in J16‑derived breeding populations.

Fig. 5.

Fig. 5

Open in a new tab

Correlation validation between molecular marker genotypes and eye‑depth phenotypes. A – B PAGE gel electrophoresis showing banding patterns of molecular markers J16_Indel06 (A) and J16_Indel09 (B) in the parental lines J16 and Atlantic (Atl), along with representative F1 individuals. The gel images shown in (A) and (B) are from Replicate 1 of three independent biological replicates. The original, uncropped full-length gel images for J16_Indel06 (Replicates 1–3) are provided in Fig. S1, and those for J16_Indel09 (Replicates 1–3) are provided in Fig. S2. (C – D) Number of F1 plants with deep‑ or shallow‑eye phenotypes for each genotype class of J16_Indel06 (C) and J16_Indel09 (D)

Discussion

This study successfully fine-mapped the major locus controlling the deep-eye trait in the tetraploid cultivated potato variety Jinshu 16 to a genomic interval of 49.0–50.7 Mb at the distal end of chromosome 10 using the BSA-seq method (Fig. 2). This result is highly consistent with the position of the Eyd locus first reported in a diploid population [11] and is consistent with subsequent studies that validated the major effect of this region across different ploidy levels and genetic backgrounds [3–5, 12, 13]. This consistency across ploidy and germplasm strongly indicates that the distal region of chromosome 10 plays a highly conserved, central role in regulating potato tuber eye depth. It is particularly noteworthy that our mapped interval overlaps with the 309.10 kb region of the qEyd10.1 locus identified in a recent fine-mapping study [16]. This not only confirms the reliability of the mapping results but also provides a precise genomic anchor for further candidate gene screening within the specific genetic background of Jinshu 16. We acknowledge the complexity of autotetraploid inheritance, but the observed 1:1 segregation ratio (P > 0.05) across two seasons strongly supports a simplex dominant (Aaaa) model for this locus. Under tetrasomic inheritance with random chromosome segregation and negligible double reduction (rare in potato), a simplex parent produces equal proportions of dominant and non-dominant gametes, yielding a 1:1 ratio when crossed with a recessive parent. The good fit to 1:1 indicates that double reduction or dosage effects are negligible here.

Transcriptome analysis further revealed high expression of a peroxidase superfamily gene cluster and a lipid transfer protein (LTP) gene cluster within the mapped interval (Fig. 4B). This finding is consistent with a recent study, which also identified a peroxidase gene cluster within the fine-mapped interval and found that several peroxidase genes were significantly upregulated in deep-eye materials during the initial tuber formation stage, suggesting their potential role in regulating eye depth through auxin (IAA) metabolism [16]. The unique contribution of this study lies in the discovery of the specific high expression of the LTP gene cluster in the eye tissue of Jinshu 16, a pattern significantly correlated with the deep-eye phenotype. Combined with cytological observations showing denser cell arrangement in the deep-eye region, we propose a potential dual gene cluster co-regulation model for this locus: the peroxidase gene cluster may inhibit cell expansion by catalyzing lignin polymerization and degrading IAA [32, 33], while the LTP gene cluster may influence the mechanical properties and developmental polarity of epidermal tissues by participating in cuticular lipid transport [34–37]. The close adjacency and co-expression patterns of these two gene clusters on the chromosome suggest they may function as a unit in regulating eye morphogenesis, though the possibility of independent action also exists. Given the complexity of the tetraploid genome and the preliminary nature of this study, we did not perform detailed sequence variation analysis for the candidate genes, as such analysis would not yield conclusive results at this stage. Functional validation of the causal variants is left for future research. Future research could employ multi-gene editing strategies, such as large-fragment knockout of one or both clusters, to systematically dissect their independent and interactive functions.

Based on the fine-mapping results, this study developed InDel molecular markers tightly linked to the deep-eye trait (Fig. 5), providing a direct tool for marker-assisted selection (MAS) in Jinshu 16 and its derived materials. It should be noted, however, that the development of these markers was based on genomic variations specific to the parental cross, and their applicability in other genetic backgrounds requires systematic validation. Therefore, future research needs to assess the effectiveness of these markers in a broader range of Jinshu 16-derived hybrid populations and in diverse tetraploid and diploid germplasm resources, and to develop more broadly applicable functional markers based on different haplotype characteristics. Furthermore, the complex genetic relationships between eye depth and other important agronomic traits must be considered, especially given that this region on chromosome 10 harbors not only the Eyd locus but also closely linked genes regulating tuber shape (e.g., the Ro locus) and anthocyanin biosynthesis, which may exhibit pleiotropic effects or linkage drag [38–40]. Thus, in practical breeding applications, it is recommended to combine the markers developed in this study with molecular markers for other related traits to construct a multi-trait MAS system, or to adopt genomic selection strategies to achieve synergistic improvement of multiple important traits through genome-wide breeding value prediction. With the advancement of potato functional genomics, these markers can be integrated with high-throughput genotyping technologies in the future to develop comprehensive breeding chips. Coupled with cutting-edge technologies like gene editing, this will ultimately establish an efficient and precise molecular design breeding system for potato appearance quality.

Conclusion

This study elucidated the genetic basis of the deep-eye trait in potato cultivar Jinshu 16, confirming that the trait is regulated by a major locus on chromosome 10 that corresponds to the known Eyd locus. Transcriptome analysis indicated the involvement of multiple pathways, including lipid metabolism, in its regulation, while molecular markers developed based on the findings enable effective trait selection. These results provide important insights into the genetic mechanisms of potato eye development and support targeted improvement of this trait.

Supplementary Information

12870_2026_9151_MOESM1_ESM.docx (20KB, docx)

Supplementary Material 1. R scripts for generating Manhattan‑like plots.

12870_2026_9151_MOESM2_ESM.docx (934.1KB, docx)

Supplementary Material 2. Fig. S1 Full-length, unprocessed gel images of J16_Indel06 (Replicates 1–3). Fig. S2 Full-length, unprocessed gel images of J16_Indel09 (Replicates 1–3).

12870_2026_9151_MOESM3_ESM.xlsx (362.1KB, xlsx)

Supplementary Material 3. Table S1 Primers designed for InDel markers within the eye-depth candidate region on chromosome 10. Table S2 Phenotypic scores and marker genotypes of the F1 population (J16 × Atlantic). Table S3 Distribution of eye-depth grades in the F1 population across two growing seasons. Table S4 Sliding-window analysis of SNP-index and Δ(SNP-index) values across chromosome 01 in deep-eye and shallow-eye bulk pools. Table S5 Expression profiles (FPKM) of candidate genes within the eye-depth locus on chromosome 10. Table S6 Summary of RNA-seq read alignment statistics to the potato reference genome (DMv6.1). Table S7 Expression profiles of candidate genes from clusters 3, 4, and 6 within the eye-depth locus. Table S8 Association between InDel marker genotypes and eye-depth phenotypes in the F1 population across two growing seasons.

Acknowledgements

We are grateful to Dr. Chunzhi Zhang from the Agricultural Genomics Institute, Chinese Academy of Agricultural Sciences, for her critical comments on the manuscript. We also thank Dr. Guangtao Zhu from the School of Life Sciences, Yunnan Normal University, for his valuable assistance during the experiments.

Abbreviations

BSA-seq

Bulked segregant analysis sequencing

CTAB

Cetyltrimethylammonium bromide

DEG

Differentially expressed gene

Eyd

Eye depth

FPKM

Fragments per kilobase of transcript per million mapped reads

GO

Gene Ontology

GWAS

Genome-wide association study

InDel

Insertion-Deletion

KEGG

Kyoto Encyclopedia of Genes and Genomes

LTP

Lipid transfer protein

MAS

Marker-assisted selection

PAGE

Polyacrylamide gel electrophoresis

PCA

Principal component analysis

POD

Peroxidase

QTL

Quantitative trait locus

SNP

Single nucleotide polymorphism

Authors’ contributions

HD and HJW designed the experiments. JP, RYW, and LGH performed the experiments. JP, RYW, and ZZ conducted the bioinformatics and data analysis. XDB and LGH contributed to resources and supervision. HD, HJW, and WZ revised and improved the manuscript. All authors reviewed and approved the final version of the manuscript.

Funding

This work was supported by the Project of the Shanxi Province Agricultural Key Core Technologies Research Program (Grant No. NYGG30-02) and the Shanxi Modern Agricultural Potato Industry Technology System (Grant No. 2026CYJSTX06-08).

Data availability

The raw sequence data reported in this paper have been deposited in publicly accessible repositories. The whole-genome resequencing data of the parental line Jinshu 16 have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1189869 (BioSample: SAMN45021058; SRA: SRS23483756). All other raw sequencing data, including the resequencing data of the parental line Atlantic and the F₁ individuals, as well as the transcriptome data of both parents, have been deposited in the Genome Sequence Archive (GSA) in the National Genomics Data Center, China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences, and are publicly accessible at https://ngdc.cncb.ac.cn/gsa under BioProject accession number PRJCA058635.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

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

Huijie Wang, Jin Pu and Rongyan Wang contributed equally to this work.

References

  • 1.Evans NE, Foulger D, Farrer L, Bright SWJ. Somaclonal variation in explant-derived potato clones over three tuber generations. Euphytica. 1986;35(2):353–61. [Google Scholar]
  • 2.Love SL, Werner BK, Pavek JJ. Selection for individual traits in the early generations of a potato breeding program dedicated to producing cultivars with tubers having long shape and russet skin. Am Potato J. 1997;74(3):199–213. [Google Scholar]
  • 3.Prashar A, Hornyik C, Young V, McLean K, Sharma SK, Dale MFB, Bryan GJ: Construction of a dense SNP map of a highly heterozygous diploid potato population and QTL analysis of tuber shape and eye depth. Theor appl genet. 2014;127(10):2159–2171. [DOI] [PubMed]
  • 4.Rosyara UR, De Jong WS, Douches DS, Endelman JB. Software for Genome-Wide Association Studies in Autopolyploids and Its Application to Potato. Plant Genome-US. 2016;9(2):plantgenome2015–8. [DOI] [PubMed] [Google Scholar]
  • 5.Sharma SK, MacKenzie K, McLean K, Dale F, Daniels S, Bryan GJ. Linkage Disequilibrium and Evaluation of Genome-Wide Association Mapping Models in Tetraploid Potato. G3-Genes Genom Genet. 2018;8(10):3185–202. [DOI] [PMC free article] [PubMed]
  • 6.Sharma SK, McLean K, Hedley PE, Dale F, Daniels S, Bryan GJ: Genotyping-by-sequencing targets genic regions and improves resolution of genome-wide association studies in autotetraploid potato. Theor Appl Genet. 2024;137(8):180. [DOI] [PMC free article] [PubMed]
  • 7.Totsky IV, Rozanova IV, Safonova AD, Batov AS, Gureeva YA, Kochetov AV, Khlestkina EK. Genomic regions of Solanum tuberosum L. associated with the tuber eye depth. Vavilov J Genet Breed. 2020;24(5):465–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Vexler L, de la Leyva-Perez O, Konkolewska M, Clot A, Byrne CR, Griffin S, Ruttink D, Hutten T, Engelen RCB, Visser C. RGF : QTL discovery for agronomic and quality traits in diploid potato clones using Potato MASH amplicon sequencing. G3-Genes Genom Genet. 2024;14(10):jkae164. [DOI] [PMC free article] [PubMed]
  • 9.Hara-Skrzypiec A, Śliwka J, Jakuczun H, Zimnoch-Guzowska E. QTL for tuber morphology traits in diploid potato. J Appl Genet. 2018;59(2):123–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Zhao L, Zou M, Deng K, Xia C, Jiang S, Zhang C, Ma Y, Dong X, He M, Na T, et al. Insights into the genetic determination of tuber shape and eye depth in potato natural population based on autotetraploid potato genome. Front Plant Sci. 2023;14:1080666. [DOI] [PMC free article] [PubMed]
  • 11.Li X, De Jong H, De Jong DM, De Jong WS. Inheritance and genetic mapping of tuber eye depth in cultivated diploid potatoes. Theor Appl Genet. 2005;110(6):1068–73. [DOI] [PubMed] [Google Scholar]
  • 12.Lindqvist-Kreuze H, Khan A, Salas E, Meiyalaghan S, Thomson S, Gomez R, Bonierbale M. Tuber shape and eye depth variation in a diploid family of Andean potatoes. BMC Genet. 2015;16(1):57. [DOI] [PMC free article] [PubMed]
  • 13.Pandey J, Scheuring DC, Koym JW, Vales MI. Genomic regions associated with tuber traits in tetraploid potatoes and identification of superior clones for breeding purposes. Front Plant Sci. 2022;13:952263. [DOI] [PMC free article] [PubMed]
  • 14.Sliwka J, Wasilewicz-Flis I, Jakuczun H, Gebhardt C. Tagging quantitative trait loci for dormancy, tuber shape, regularity of tuber shape, eye depth and flesh colour in diploid potato originated from six Solanum species. Plant Breeding. 2008;127(1):49–55. [Google Scholar]
  • 15.Śliwka J, Wasilewicz-Flis I, Jakuczun H, Janiszewska M, Smyda-Dajmund P, McLean K, Zimnoch-Guzowska E, Bryan GJ, Sharma SK. Historical data provide new insights into inheritance of traits important for diploid potato breeding. Planta. 2025;261(4):69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Fan G, Duan S, Yang Y, Duan Y, Jian Y, Hu J, Liu Z, Guo Y, Jin L, Xu J, et al. Fine-mapping and candidate gene analysis of tuber eye depth in potato. Hortic Plant J. 2024;11(3):1248–59. [Google Scholar]
  • 17.Chen Y, Bai X, Yue X, Fan X, Wang C, Guo F, Shuai Y, Wang J. A new potato variety Jinshu 16. Tillage Cultivation. 2022;42(01):83–5. [Google Scholar]
  • 18.Ding P, Chen D, Feng H, Li J, Cao H, Tang M, Li J, Hao X, Han P, Meng Z, et al. Prevalence and Strain Composition of Potato virus Y Circulating in Potato Fields in China’s North-Central Province of Shanxi. Plant Dis. 2022;106(5):1434–45. [DOI] [PubMed] [Google Scholar]
  • 19.Liu J, Mei C, Wang H, Song Q, Feng R. Effects of Salt Stress on the Growth and Physiological Characteristics of Four Virus-free Potato Seedlings. J Shanxi Agricultural Sci. 2021;49(05):550–3. [Google Scholar]
  • 20.Allen GC, Flores-Vergara MA, Krasynanski S, Kumar S, Thompson WF. A modified protocol for rapid DNA isolation from plant tissues using cetyltrimethylammonium bromide. Nat Protoc. 2006;1(5):2320–5. [DOI] [PubMed] [Google Scholar]
  • 21.Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Pham GM, Hamilton JP, Wood JC, Burke JT, Zhao H, Vaillancourt B, Ou S, Jiang J, Buell CR. Construction of a chromosome-scale long-read reference genome assembly for potato. Gigascience. 2020;9(9):giaa100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Consortium TPGS. Genome sequence and analysis of the tuber crop potato. Nature. 2011;475(7355):189–95. [DOI] [PubMed] [Google Scholar]
  • 24.Li H, Durbin R. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics. 2009;25(14):1754–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, Garimella K, Altshuler D, Gabriel S, Daly M, et al. The Genome Analysis Toolkit: A MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Li H. A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data. Bioinformatics. 2011;27(21):2987–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Takagi H, Abe A, Yoshida K, Kosugi S, Natsume S, Mitsuoka C, Uemura A, Utsushi H, Tamiru M, Takuno S, et al. QTL -seq: rapid mapping of quantitative trait loci in rice by whole genome resequencing of DNA from two bulked populations. Plant J. 2013;74(1):174–83. [DOI] [PubMed] [Google Scholar]
  • 28.Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat Methods. 2015;12(4):357–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Pertea M, Pertea GM, Antonescu CM, Chang T, Mendell JT, Salzberg SL. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat Biotechnol. 2015;33(3):290–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Chen C, Wu Y, Li J, Wang X, Zeng Z, Xu J, Liu Y, Feng J, Chen H, He Y, et al. TBtools-II: A one for all, all for one bioinformatics platform for biological big-data mining. Mol Plant. 2023;16(11):1733–42. [DOI] [PubMed] [Google Scholar]
  • 31.Rozen S, Skaletsky H. Primer3 on the WWW for General Users and for Biologist Programmers. In. 2003;132:365–86. [DOI] [PubMed] [Google Scholar]
  • 32.Shigeto J, Tsutsumi Y. Diverse functions and reactions of class III peroxidases. New Phytol. 2015;209(4):1395–402. [DOI] [PubMed] [Google Scholar]
  • 33.Francoz E, Ranocha P, Nguyen-Kim H, Jamet E, Burlat V, Dunand C. Roles of cell wall peroxidases in plant development. Phytochemistry. 2015;112:15–21. [DOI] [PubMed] [Google Scholar]
  • 34.Missaoui K, Gonzalez-Klein Z, Pazos-Castro D, Hernandez-Ramirez G, Garrido-Arandia M, Brini F, Diaz-Perales A, Tome-Amat J. Plant non-specific lipid transfer proteins: An overview. Plant Physiol Bioch. 2022;171:115–27. [DOI] [PubMed] [Google Scholar]
  • 35.Jacq A, Pernot C, Martinez Y, Domergue F, Payré B, Jamet E, Burlat V, Pacquit VB. The Arabidopsis Lipid Transfer Protein 2 (AtLTP2) Is Involved in Cuticle-Cell Wall Interface Integrity and in Etiolated Hypocotyl Permeability. Front Plant Sci. 2017;8:263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Deeken R, Saupe S, Klinkenberg J, Riedel M, Leide J, Hedrich R, Mueller TD. The Nonspecific Lipid Transfer Protein AtLtpI-4 Is Involved in Suberin Formation of Arabidopsis thaliana Crown Galls. Plant Physiol. 2016;172(3):1911–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Nieuwland J, Feron R, Huisman BAH, Fasolino A, Hilbers CW, Derksen J, Mariani C. Lipid Transfer Proteins Enhance Cell Wall Extension in Tobacco. Plant Cell. 2005;17(7):2009–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.AI J, Wang Y, Yan Y, Li C, Luo W, Shang Mal, Gao Y. StOFP20 regulates tuber shape and interacts with TONNEAU1 Recruiting Motif proteins in potato. J Integr Agr. 2023;22(3):752–61. [Google Scholar]
  • 39.Du H, Zhai Z, Pu J, Liang J, Wang R, Zhang Z, Wang P, Zhu Y, Huang L, Li D, et al. Two tandem R2R3 MYB transcription factor genes cooperatively regulate anthocyanin accumulation in potato tuber flesh. Plant Biotechnol J. 2025;23(5):1521–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Fan G, Wang Q, Xu J, Chen N, Zhu W, Duan S, Yang X, De Jong WS, Guo Y, Jin L, et al. Fine Mapping and Candidate Gene Prediction of Tuber Shape Controlling Ro Locus Based on Integrating Genetic and Transcriptomic Analyses in Potato. Int J Mol Sci. 2022;23(3):1470. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

12870_2026_9151_MOESM1_ESM.docx (20KB, docx)

Supplementary Material 1. R scripts for generating Manhattan‑like plots.

12870_2026_9151_MOESM2_ESM.docx (934.1KB, docx)

Supplementary Material 2. Fig. S1 Full-length, unprocessed gel images of J16_Indel06 (Replicates 1–3). Fig. S2 Full-length, unprocessed gel images of J16_Indel09 (Replicates 1–3).

12870_2026_9151_MOESM3_ESM.xlsx (362.1KB, xlsx)

Supplementary Material 3. Table S1 Primers designed for InDel markers within the eye-depth candidate region on chromosome 10. Table S2 Phenotypic scores and marker genotypes of the F1 population (J16 × Atlantic). Table S3 Distribution of eye-depth grades in the F1 population across two growing seasons. Table S4 Sliding-window analysis of SNP-index and Δ(SNP-index) values across chromosome 01 in deep-eye and shallow-eye bulk pools. Table S5 Expression profiles (FPKM) of candidate genes within the eye-depth locus on chromosome 10. Table S6 Summary of RNA-seq read alignment statistics to the potato reference genome (DMv6.1). Table S7 Expression profiles of candidate genes from clusters 3, 4, and 6 within the eye-depth locus. Table S8 Association between InDel marker genotypes and eye-depth phenotypes in the F1 population across two growing seasons.

Data Availability Statement

The raw sequence data reported in this paper have been deposited in publicly accessible repositories. The whole-genome resequencing data of the parental line Jinshu 16 have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1189869 (BioSample: SAMN45021058; SRA: SRS23483756). All other raw sequencing data, including the resequencing data of the parental line Atlantic and the F₁ individuals, as well as the transcriptome data of both parents, have been deposited in the Genome Sequence Archive (GSA) in the National Genomics Data Center, China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences, and are publicly accessible at https://ngdc.cncb.ac.cn/gsa under BioProject accession number PRJCA058635.

Articles from BMC Plant Biology are provided here courtesy of BMC

RESOURCES