Intraspecific phylogeny and genomic resources development for an important medical plant Dioscorea nipponica, based on low-coverage whole genome sequencing data

Dioscorea nipponica Makino, a perennial twining herb with medicinal importance, has a disjunctive distribution in the Sino-Japanese Floristic Region. It has a long history in traditional Chinese medicine, with demonstrated efficacy against various health conditions. However, the limited genomic data and knowledge of genetic variation have hindered its comprehensive exploration, utilization and conservation. In this study, we undertook low-coverage whole genome sequencing of diverse D. nipponica accessions to develop both plastome (including whole plastome sequences, plastome-derived SSRs and plastome-divergent hotspots) and nuclear genomic resources (including polymorphic nuclear SSRs and single-copy nuclear genes), as well as elucidate the intraspecific phylogeny of this species. Our research revealed 639 plastome-derived SSRs and highlighted six key mutational hotspots (namely CDS ycf1, IGS trnL-rpl32, IGS trnE-trnT, IGS rps16-trnQ, Intron 1 of clpP, and Intron trnG) within these accessions. Besides, three IGS regions (i.e., ndhD-cssA, trnL-rpl32, trnD-trnY), and the intron rps16 were identified as potential markers for distinguishing D. nipponica from its closely related species. In parallel, we successfully developed 988 high-quality candidate polymorphic nuclear SSRs and identified 17 single-copy nuclear genes for D. nipponica, all of which empower us to conduct in-depth investigations into phylogenetics and population genetics of this species. Although our phylogenetic analyses, based on plastome sequences and single-copy nuclear genes revealed cytonuclear discordance within D. nipponica, both findings challenged the current subspecies classification. In summary, this study developed a wealth of genomic resources for D. nipponica and enhanced our understanding of the intraspecific phylogeny of this species, offering valuable insights that can be instrumental in the conservation and strategic utilization of this economically significant plant.


Introduction
Dioscorea nipponica Makino, a perennial twining herb of the genus Dioscorea belonging to the family Dioscoreaceae, is disjunctively distributed across the Sino-Japanese Floristic Region (Obidiegwu et al., 2020;Yang et al., 2022).This plant is characterized by its slender and cylindrical aerial stem, alternate simple palmate leaves with an anisometric triangular shallow, medium, or deep crack along the leaf edge, and unisexual yellowish-green flowers that droop like small bells (Ding and Gilbert, 2000;Ou-Yang et al., 2018).As it matures, it produces dry capsules that are yellow, obovate-elliptic, and prismatic in shape with winged edges (Ding and Gilbert, 2000).Dioscorea nipponica has a long history of use in traditional Chinese medicine, where it has proven effective against various conditions, including rheumatoid arthritis, Kashin-Beck disease, sprains, bruises, chronic bronchitis, and cough (Ou- Yang et al., 2018;Yang et al., 2022).Modern pharmacological research has also revealed the multifaceted attributes of D. nipponica, showcasing its wideranging benefits, such as its anti-inflammatory, anti-tumor, analgesic, antitussive, calming, and phlegm-dispelling properties (Wu et al., 2023).Notably, recent scientific studies have isolated both fat-soluble and water-soluble steroidal saponins from rhizomes of D. nipponica, attributing most of its pharmacological effects to saponins and sapogenins (Ou-Yang et al., 2018).Additionally, its aboveground parts contain over ten types of phenanthrene derivatives, further contributing to its medicinal value (Lu et al., 2010;Li et al., 2017).Despite its medicinal significance, a pivotal research gap lies in the limited availability of genomic information, impeding the exploration of new bioactive compounds and a comprehensive understanding of their synthesis.Furthermore, although demand is increasing with growing awareness, the development and utilization of D. nipponica resources have been slow (Ou-Yang et al., 2018).The species faces challenges meeting market demand due to the gradual depletion and unrestrained exploitation of wild resources (Chen et al., 2007;Ou-Yang et al., 2018).Therefore, there is an urgent need for additional molecular markers to enhance efforts related to conservation, utilization, and breeding of this economically significant species.
To date, the taxonomic classification and intraspecific phylogenetic relationship of D. nipponica have remained subjects of ongoing debate and controversy.Various studies have proposed different perspectives, leading to conflicting perspectives on the subspecies delineation.For instance, Ding and Gilbert (2000) proposed a division of D. nipponica into two distinct subspecies, drawing upon considerations such as chromosome count, cork layer characteristics, and geographical distribution.The original subspecies, D. nipponica subsp.nipponica, is predominantly located in the northern reaches of the Qinling Mountain range, characterized by a rhizome with an easily peelable cork layer and a chromosome count of 20.In contrast, D. nipponica subsp.rosthornii, the second subspecies, features a persistent cork layer and a chromosome count of 40, primarily inhabiting the southern region of the Qinling Mountains in central China.However, Gao et al. (2008) presented an alternative perspective, suggesting a closer relationship between D. nipponica subsp.rosthornii and D. althaeoides than with D. nipponica subsp.nipponica.Their proposition advocated for the elevation of D. nipponica subsp.rosthornii to the status of an independent species, a viewpoint further corroborated by the chemotaxonomic analysis conducted by Li et al. (2020).Our recent phylogenetic analysis of Stenophora species/subspecies based on complete plastome sequences provided strong support for the monophyly of D. nipponica subsp.nipponica and D. nipponica subsp.rosthornii (Hu et al., 2023).Nevertheless, it is crucial to acknowledge that all previous studies have been limited by their restricted taxon sampling, typically involving just one specimen each for D. nipponica subsp.nipponica and D. nipponica subsp.rosthornii (e.g., Gao et al., 2008;Hu et al., 2023), or relying on a limited set of genetic markers, such as matK, rbcL, and trnL-F (e.g., Gao et al., 2008).Therefore, obtaining a more extensive range of diverse samples from various geographic regions is crucial, with a specific focus on covering both the northern and southern areas of the Qinling Mountains and expanding the scope beyond China's borders.Additionally, employing stronger molecular markers, such as both plastomes and multiple nuclear loci, is vital for conducting intraspecific phylogenetic analysis of this species.
The advent of next-generation sequencing (NGS) technologies has ignited a profound revolution in the acquisition of genomescale data across a broad spectrum of plant species (Dodsworth, 2015;Lu et al., 2022).Within the realm of NGS methodologies, lowcoverage whole genome sequencing, commonly known as genome skimming, has emerged as a notably cost-effective and efficient approach (Straub et al., 2012;Twyford and Ness, 2017;Jin et al., 2020).This technique employs shallow sequencing of genomic DNA, strategically capturing a significant fraction of the genome, with a particular emphasis on high-copy elements such as ribosomal DNA, the plastome, and nuclear repeats like simple sequence repeats (SSRs) and transposable elements (Straub et al., 2012;Dodsworth, 2015;Nevill et al., 2020;Lu et al., 2022).Plastome sequences, among these molecular markers, have proven immensely valuable in plant species identification and phylogenetic studies, thanks to their distinctive characteristics: the absence of recombination, low rates of nucleotide substitutions, small effective population sizes, and typically uniparental inheritance (Birky et al., 1983;Lu et al., 2021).Moreover, recent studies have demonstrated the capacity to recover single-copy nuclear genes (SCNGs) by leveraging multiple assembled nuclear sequences derived from low-coverage whole genome sequencing data (e.g., Liu et al., 2021;Zhou et al., 2022), offering a promising resource for plant phylogenetic analyses.
Thus, to expand genomic resources available for D. nipponica and deepen our insights into its intraspecific phylogeny and taxonomy, we conducted low-coverage whole genome sequencing on diverse D. nipponica accessions originating from various regions, including both the northern and southern areas of the Qinling Mountains, as well as Japan.Our objectives encompassed the following key aspects: 1) identification of plastome-derived markers, including whole plastome sequences, plastome-derived SSRs and plastome-divergent hotspots; 2) development of genomewide nuclear markers, encompassing polymorphic nuclear SSRs (PolynSSRs) and SCNGs; and 3) reconstruction of the phylogenetic relationships among D. nipponica accessions based on both plastome and SCNG data.These findings will offer valuable support for the conservation and strategic utilization of this economically significant species.

Plant materials, DNA extraction and genome sequencing
A sample of eight D. nipponica accessions originating from diverse regions, including Beijing (BJ), Gansu (GS), Henan (HeN), Hunan (HuN), Jining (JN) and Zhejiang (ZJ) Provinces of China, as well as Fukushima (FD) and Nagano (NI) Prefectures of Japan was used in this study (Table 1).The sampling strategy was designed to encompass the primary geographical distribution of this species, including both the northern and southern areas of the Qinling Mountains, as well as Japan, and was carried out under the special permission granted by the Institute of Botany, Jiangsu Province and Chinese Academy of Sciences.For each accession, fresh, healthy leaf samples were collected from a wild mature individual, and dried with silica-gel.The voucher specimens were deposited at Herbarium of Institute of Botany, Jiangsu Province and Chinese Academy of Sciences (NAS).Genomic DNA was extracted from approximately 50 mg of silica-dried leaf samples using DNAsecure Plant Kit (Tiangen Biotech, Beijing, China), in accordance with the manufacturer's prescribed protocol.DNA quantity and purity were checked by spectrophotometry and agarose gel electrophoresis.
For each accession, a barcoded paired-end library with an insert size of 350 bp was constructed using NEBNext Ultra DNA Library Prep Kit for Illumina.Following the ligation of indexed adapters, these indexed DNA libraries were pooled and subjected to pairedend sequencing in a single lane of HiSeq X Ten.Subsequently, Trimmomatic v.0.36 (Bolger et al., 2014) was employed to eliminate adaptor sequences, contamination, and low-quality reads from the raw data.This filtering process yielded approximately 4.5 Gb of clean data for each accession.The entire sequence library preparation, genome sequencing, and raw data filtering procedures were conducted by Novogene Bioinformatics Technology Co., Ltd., located in Beijing, China.

Characterization of plastomederived SSRs
The MISA-web application (Beier et al., 2017) was utilized to identify simple sequence repeats (SSRs) within the eight newly assembled plastome sequences of D. nipponica.SSR identification criteria were as follows: a minimum threshold of 10 repeat units for mononucleotide SSRs, 5 for dinucleotide SSRs, 4 for trinucleotide SSRs, and 3 for tetra-, penta-, and hexanucleotide SSRs, respectively.Data visualization was performed by the OmicStudio tools at https://www.omicstudio.cn/tool(Lyu et al., 2023).
To further develop potential molecular markers for distinguishing D. nipponica from its closely related species, i.e., D. collettii, D. gracillima, D. villosa, and D. zingiberensis (Hu et al., 2023), a total of six plastome sequences, including two for D. nipponica (BJ and FD, representing Chinese and Japanese accession, respectively), and one each for D. collettii (OQ525992), D. gracillima (OQ525995), D. villosa (KY085893), and D. zingiberensis (OQ526000) were analyzed.Nucleotide diversity of CDS, introns and IGS regions was computed using the same method described above.

Development of polymorphic nuclear SSRs
Low-coverage whole genome sequencing data from each D. nipponica accession were firstly mapped onto the genome sequences of D. alata (Bredeson et al., 2022) and D. zingiberensis (Li et al., 2022) to exclude mitochondria and chloroplast reads, using BWA-MEM v.0.7.17 (Li, 2013).The resulting Binary Alignment/Map (BAM) data, which exclusively contained nuclear reads, were de novo assembled into scaffolds using the SOAPdenovo2 program (Luo et al., 2012), a de Bruijn graphbased assembly program.Subsequent to nuclear scaffolds generation, the identification of candidate polymorphic nuclear SSRs (PolynSSRs) were conducted using CandiSSR pipeline (Xia et al., 2016), with default parameters.The data visualization was also accomplished using the OmicStudio tools (Lyu et al., 2023).

Identification of single-copy nuclear genes
To retrieve Angiosperms353 target genes (Johnson et al., 2019) within the genome of D. nipponica, low-coverage whole genome sequencing data from each D. nipponica accession were independently subjected to HybPiper v. 2.1.6(Johnson et al., 2016) for assembling sequences for each gene, with all settings at default.Briefly, paired-end clean reads from each accession were mapped to target genes with bwa v. 0.7.17 (Li and Durbin, 2009).Subsequently, mapped reads were organized into distinct directories and assembled into contigs using SPAdes v.3.13.1 (Bankevich et al., 2012).These assembly contigs were then aligned to their associated target sequences using Exonerate v.2.2 (Slater and Birney, 2005).Finally, the recovered gene sequences were extracted using the HybPiper script called retrieve_sequences.py.Additionally, recovery statistics were generated using the two Python scripts, namely get_seq_lengths.pyand hybpiper_stats.py,which are included in the HybPiper pipeline (Johnson et al., 2016).The resulting gene sequences shared across all eight D. nipponica accessions were imported into Geneious Prime® 2022.0.1, and aligned individually with the multiple alignment plugin MAFFT v.7 (Katoh and Standley, 2013).After excluding those alignments with pairwise identity below 90%, the remaining alignments were concatenated into a supermatrix for nuclear phylogenetic analyses.

Phylogenetic analyses within D. nipponica
For plastome phylogenetic analyses, maximum likelihood (ML) and Bayesian inference (BI) analyses were conducted based on two distinct datasets: complete plastome sequences and a set of 80 shared protein coding regions present in all eight accessions examined in this study (Table 1), with D. zingiberensis (OQ526000) as an outgroup.Both complete plastome sequences and protein coding sequences were aligned using MAFFT v.7 plugin (Katoh and Standley, 2013) in Geneious Prime® 2022.0.1.The bestfitting substitution model, GTR + I + G, for each dataset was determined based on the Akaike Information Criterion (AIC) as computed by jModelTest v.2.1.4(Darriba et al., 2012).The ML analyses were performed using RAxML v.8.2.12 (Stamatakis, 2014) in the CIPRES Science Gateway v.3.3 (http://www.phylo.org/portal2/), with 1000 bootstrap replications.The BI analyses were conducted using MrBayes v.3.2.7 (Ronquist et al., 2012), comprising two independent runs of 1 × 10 6 generations.Each run employed four independent Markov chain Monte Carlo (MCMC) chains, consisting of one cold chain and three heated chains, with a sampling frequency of 1000 trees.The first 1000 trees were discarded as 'burn-in', and the remaining trees were used to construct a majority-rule consensus tree and estimate the posterior probabilities (PPs).

Plastome features of D. nipponica
Illumina paired-end (150 bp) sequencing produced 23,737,170-33,758,516 clean reads for these eight D. nipponica accessions (Table 1).The coverage depths resulting from mapping the Illumina reads to the plastome sequences ranged from 451× (JN) to 2631× (BJ) (Table 1).Additionally, both GetOrganelle and NOVOPlasty generated identical plastome assemblies.Taken together, these results indicated a high-quality and accuracy of our plastome assemblies.The whole plastome sequences of the eight D. nipponica accessions exhibited a narrow range in size, spanning from 153,917 bp (BJ and HeN) to 154,076 bp (FD) (Figure 1; Table 1).The plastome of D. nipponica maintained the typical circular quadripartite structure, consisting of a pair of inverted The plastome maps of D. nipponica accessions.Genes shown on the outside of the circle are transcribed clockwise, and those inside counterclockwise.Genes associated with different functional categories are color coded.The darker grey in the inner ring corresponds to the GC content and the lighter grey to the AT content.The D. nipponica plant was displayed within the inner circle.1).Notably, the lengths of the IR regions in six Chinese accessions were consistent at 25,508 bp, representing a 22 bp reduction when compared to the IR regions in the two Japanese accessions measuring 25,530 bp each (Table 1).
The GC content of whole plastome sequences (37.20%),LSC (35.00%) and IR (43.00%) regions were identical across the eight D. nipponica accessions, while in the SSC region, the GC content in six Chinese accessions (31.20%) was slightly lower than that in two Japanese accessions (31.30%) (Table 1).

Plastome-derived SSRs
The MISA analysis detected a total of 639 simple sequence repeats (SSRs) derived from the plastomes of the eight different D. nipponica accessions.The number of plastome-derived SSRs for each accession ranged from 78 (for ZJ) to 83 (for NI).It appeared that the two Japanese accessions (with 82-83 SSRs) contained more SSRs compared to the six Chinese accessions (with 78-80 SSRs) (Figure 2; Table S2).Among these plastome-derived SSRs, mononucleotide repeats were the most prevalent, varying from 41 (for HuN) to 46 (for NI), followed by dinucleotide repeats (15-16 for each accession) and tetranucleotide repeats (10 for each accession).In contrast, trinucleotide repeats (4 for each accession), pentanucleotide repeats (ranging from 4 to 6 for each accession), and hexanucleotide repeats (3 for each accession) were relatively less common in the D. nipponica accessions (Figure 2; Table S2).The most frequent motifs observed were A/ T and AT/TA for mono-and dinucleotide repeats, constituting 50.63%-53.01%and 15.00%-15.85% of the total plastome-derived SSRs across all eight D. nipponica accessions, respectively.Additionally, a set of at least four plastome-derived SSRs, namely (A/T) 16 , (A/T) 18 , (C/G) 14 , and (AATAT/ATATT) 3 , could effectively distinguish between the Chinese and Japanese groups among the eight D. nipponica accessions (Figure 2; Table S2).

Single-copy nuclear genes
Clean reads from these eight accessions were collectively mapped to 351-352 genes, yielding 37 to 117 sequences for each respective accession (Table 2).Notably, among these sequences, 18 were found to be shared across all eight accessions.Following the exclusion of one alignment with pairwise identity below 90%, a supermatrix with a total length of 2154 bp was formed by concatenating the remaining 17 gene sequences.This supermatrix was subsequently utilized for nuclear phylogenetic analyses.However, it's worth emphasizing that only a handful of sequences (6 for BJ, 19 for GS, 3 for HeN, 2 for HuN, 25 for JN, 27 for ZJ, 28 for FD, and 18 for NI, respectively), each with a length comprising at least 50% of the target, were successfully recovered (Table 2).

Intraspecific phylogenetic relationship of D. nipponica
Both the ML and BI analyses, conducted using complete plastome sequences and 80 shared protein coding regions, provided robust support for the division of D. nipponica into distinct Chinese and Japanese groups (Figure 6A), with high bootstrap values (BS values = 100%) and posterior probabilities  Hu et al. 10.3389/fpls.2023.1320473Frontiers in Plant Science frontiersin.org(PPs = 1.0).Notably, no evident genetic distinction was observed within the Chinese group 6A).However, when phylogenetic analyses were conducted based on 17 single-copy nuclear genes, they indicated that the two Japanese accessions formed a monophyletic group with two Chinese accessions (HuN and ZJ) from Central and South China.These, in turn, constituted a sister group with the other four Chinese accessions (BJ, GS, JN, and HeN) from North China (Figure 6B).

Discussion
4.1 Plastome characteristics and plastomederived markers of D. nipponica Plastomes have become a cornerstone in the realm of plant phylogenetics and evolutionary studies, primarily due to their unwavering preservation of gene order and the striking absence of heteroplasmy and recombination (Birky et al., 1983;Daniell et al., 2016).Notably, plastomes exhibit uniparental inheritance, typically maternal in angiosperms and paternal in gymnosperms, offering a distinctive avenue for unraveling the respective contributions of seed and pollen dispersal to the genetic makeup of natural populations-a perspective enriched when contrasted with nuclear markers (Birky, 1995;Mohammad-Panah et al., 2017).In this study, the investigation of plastome features in eight D. nipponica accessions unveiled a striking uniformity in terms of gene count, content, and arrangement (Figure 1; Table 1), which suggested an enduring evolutionary stability within D. nipponica.
The differences in the sizes of inverted repeats (IRs) and the variations observed at four distinct junctions (IRa/SSC, IRa/LSC, IRb/SSC, and IRb/LSC) often significantly influence the overall size of the plastomes and the count of genes, across angiosperm species (Sun et al., 2016).Remarkably, within a single species, the lengths of IR regions were always consistently identical among different individuals or accessions (e.g., Zizania latifolia, Lu et al., 2022;Dioscorea alata, Lu et al., 2023), playing a crucial role in stabilizing plastomes (Blazier et al., 2016).Our study, in contrast to earlier findings indicating uniform IR length within a specific species, revealed discrepancies in the length of IR regions between Chinese and Japanese accessions (Table 1).This discovery may signify historical divergence or unique evolutionary history of D. nipponica accessions in these distinct geographic regions (Jansen and Ruhlman, 2012).
Recent studies have underscored the role of simple sequence repeats (SSRs) in adaptation, survival, and evolution of species (Labbéet al., 2011;Yuan et al., 2021).The differences observed in plastome-derived SSRs between Chinese and Japanese accessions (Figure 2), hold significance in understanding the genetic variations and evolutionary history of D. nipponica.It is plausible that geographic isolation, differing environmental factors, and historical events have led to genetic differences between these distinct geographical accessions (Qiu et al., 2011).Further exploration of these variations and their correlation with historical events and/or environmental factors, particularly within the context of a population genetic framework, could provide valuable insights into the evolutionary and population dynamics of D. nipponica.
Due to their smaller effective population sizes compared to nuclear genomes, as well as the limited gene dispersal via seeds as opposed to pollen-mediated gene flow, plastome-derived markers have the potential to serve as effective indicators for historical bottlenecks, founder effects, and genetic drift (Mohammad-Panah et al., 2017).Regrettably, when this study was initiated, plastomederived markers tailored for D. nipponica were very limited.The present study successfully identified a significant array of plastomederived SSRs, ranging from 78 for ZJ to 83 for NI (Figure 2; Table S2), as well as six mutational hotspots (CDS ycf1, trnL-rpl32, trnE-trnT, rps16-trnQ, Intron 1 of clpP, and Intron trnG) (Figure 3).These newly discovered plastome-derived markers hold substantial promise as valuable tools for population genetics investigations and phylogenetic analyses of D. nipponica.

Nuclear genomic resources for D. nipponica
Nuclear SSR markers (nSSRs) have proven to be invaluable tools various aspects, such as population genetic analyses, identification of germplasm resources, and marker-assisted breeding programs, due to their co-dominant inheritance, adherence to Mendelian inheritance principles, wide genomic distribution, high polymorphism, and verifiable neutrality (Kaldate et al., 2017;Lu et al., 2022).In this study, we made a noteworthy discovery of 988 high-quality candidate polymorphic nSSRs (Figure 5; Table S4), which hold immense potential in shedding light on the genetic diversity and population structure of D. nipponica.More importantly, conducting comparative analyses of nSSRs and plastome-derived SSRs may provide complementary and sometimes contrasting perspectives on the genetic structure, differentiation, and gene flow (pollen-and seedmediated) among D. nipponica populations (Mohammad-Panah et al., 2017).Among these molecular markers, five polymorphic nSSRs (i.e., nSSR_901, nSSR_1065, nSSR_1163, nSSR_1491 and nSSR_2102) showing differentiation between BJ and FD, along with five plastome-derived SSRs were selected for validation using PCRbased Sanger sequencing (Supplementary Data).The sequence validation showed 100% similarity, affirming the accuracy and reliability of these identified molecular markers.Undoubtedly, the genomic resources described here formed a foundational platform for future studies in population genetics, evolution, and breeding of D. nipponica, crucial for developing effective conservation strategies, understanding the genetic basis of adaptation, and designing suitable breeding programs.
Single-copy nuclear genes (SCNGs), characterized by bi-parental inheritance, higher evolutionary rates, and numerous unlinked loci (Alvarez et al., 2008;Hojjati et al., 2019), offer significant potential for addressing issues related to hybridization and incomplete lineage sorting, potentially reconciling discrepancies among plastome genes (Small et al., 2004;Doyle, 2022).However, the recovery of SCNGs in NumReads: total number of input reads in the *.fastq files provided; ReadsMapped: total number of input reads that mapped to sequences in the target file; GenesMapped: Number of genes in the target file that had reads mapped to their representative sequences; GenesWithSeq: Number of genes with sequences; GenesAt25pct and GenesAt50pct: Number of genes with sequences > 25% and 50% of the mean target length, respectively.

B A
Intraspecific phylogenetic trees of D nipponica inferred from (A) complete plastome sequences and 80 shared protein coding regions, as well as (B) 17 single copy nuclear genes, based on the methods of maximum likelihood (ML) and Bayesian inference (BI).The ML bootstrap values/BI posterior probabilities were displayed above the lines.1) represented about 8× and 4× coverage of the estimated diploid and tetraploid genomes, respectively.Although the sequencing coverage was somewhat lower than optimal minimum sequencing depth (10×) for highquality SCNG assembly via low-coverage genome sequencing (Liu et al., 2021), the successful retrieval of 17 SCNGs from diverse D. nipponica accessions underscored their potential in resolving phylogenetic relationships within this species.Efforts aimed at obtaining more comprehensive SCNG sequences, possibly through improved sequencing methodologies or increased coverage, will be imperative.Addressing these limitations will undoubtedly enhance the accuracy and robustness of phylogenetic reconstructions, providing a more nuanced understanding of the intraspecific phylogenetic relationships of D. nipponica.

Intraspecific phylogenetic relationship of D. nipponica
The taxonomic classification and intraspecific phylogenetic relationships within D. nipponica have long been the subject of ongoing debate and controversy (Ding and Gilbert, 2000;Gao et al., 2008;Li et al., 2020;Hu et al., 2023).However, previous taxonomic and phylogenetic studies concerning D. nipponica have often grappled with limitations, either relying on short DNA sequences or featuring a restricted sample of taxa, resulting in constrained and sometimes conflicting conclusions (Gao et al., 2008;Li et al., 2020;Hu et al., 2023).Although in this study, phylogenetic analyses based on plastome sequences and concatenated SCNG data, revealed cytonuclear discordance within D. nipponica: the former supported the differentiation of D. nipponica accessions into distinct Chinese and Japanese groups (Figure 6A), while the latter suggested that the Japanese accessions formed a monophyletic group with two Chinese accessions (HuN and ZJ) from Central and South China (Figure 6B), both findings challenged the current subspecies classification as per the Flora of China (Ding and Gilbert, 2000) and the proposition to elevation of D. nipponica subsp.rosthornii to the status of an independent species (Gao et al., 2008;Li et al., 2020).This cytonuclear discordance may be attributed to both ancient and recent hybridization events as well as polyploidization occurrences within this species (Ding and Gilbert, 2000).Alternatively, it could be caused by the lack of phylogenetic resolution in the SCNG data.The nuclear gene-based phylogeny is inadequate in offering definitive insights into intraspecies relationships within D. nipponica (see Figure 6B), highlighting the inherent challenges of resolving deeper phylogenetic relationships with a limited set of nuclear genes.Thus, to gain a more comprehensive understanding of the taxonomy and evolutionary history of D. nipponica, it is imperative to utilize more robust molecular markers, such as nuclear SNPs, and conduct broader sampling across various regions.

Conclusions
In this study, we conducted low-coverage whole genome sequencing of diverse D. nipponica accessions, to retrieve plastome information, including whole plastome sequences, plastome-derived SSRs and plastome-divergent hotspots, as well as nuclear genomic markers, including polymorphic nuclear SSRs and single-copy nuclear genes.Our findings revealed a striking uniformity in plastome features across diverse D. nipponica accessions, with subtle length differences in inverted repeat regions between Chinese and Japanese accessions.A total of 639 plastome-derived SSRs and six divergent hotspots were identified from D. nipponica plastomes.Besides, four highly divergent hotspots were developed as potential markers for distinguishing D. nipponica from its closely related species.In parallel, 988 highquality candidate polymorphic nuclear SSRs and 17 single-copy nuclear genes were obtained.The genomic resources identified here will aid in the conservation and strategic utilization of this economically significant plant.Furthermore, the study shed light on the intraspecific phylogenetic relationships of D. nipponica, challenging the current subspecies classification and highlighting the need for further taxon sampling and the integration of more robust molecular markers to thoroughly unravel the intraspecific relationship and evolutionary history of this species.

TABLE 1
Summary of characteristics of eight Dioscorea nipponica plastomes.

TABLE 2
The recovery efficiency of Angiosperms353 target genes in eight D. nipponica accessions.
Table2), possibly due to low sequencing coverage.Our previous flow cytometry analysis genome sizes of ~550 Mb and ~1.10 Gb for diploids (2n = 20) and tetraploids (2n = 40), respectively (detailed data not shown), thus the total clean data generated in this study (Table