Skip to main content
NCBI home page
Scientific Data logoLink to Scientific Data
. 2026 Apr 21;13:922. doi: 10.1038/s41597-026-07286-0

A telomere-to-telomere genome assembly of Chinese water deer (Hydropotes inermis)

Haotian Wu 1,2,#, Hengquan Zhao 1,2,#, Qiuting Chen 3,4, Min Chen 3,4,✉, Ying Lu 1,2,✉
PMCID: PMC13284385  PMID: 42014755

Abstract

Hydropotes inermis (Chinese water deer, 2n = 70) is a relatively primitive small Cervidae species naturally distributed along the eastern coast of China and the Korean Peninsula. High-quality genomic resources are essential for investigating its unique adaptive and biological traits. In this study, we assembled the telomere-to-telomere (T2T) gap-free genome of a female H. inermis using PacBio HiFi, Oxford Nanopore Technologies (ONT), and Hi-C sequencing technologies. The assemblies were 3.45 Gb with a contig N50 of 101.09 Mb in size, which were anchored on 35 chromosomes, carrying 60 telomeres and 35 centromeres. Genome annotation annotated 33.36% repetitive sequences and 24,398 protein-coding genes. Comparative analyses with genomes of closely related species confirmed high genome integrity, continuity, and accuracy, supported by a quality value (QV) of 52.56 and a BUSCO completeness of 99.40%. This study provides a valuable genetic resource for H. inermis and serves as an important reference for investigating the evolutionary history of Cervidae.

Background & Summary

The Chinese water deer (Hydropotes inermis) represents one of the most primitive species within the Artiodactyla family Cervidae, characterized by its small body size, absence of antlers, and well-developed upper canine teeth in males that form prominent tusks. It is primarily distributed in eastern China and the Korean Peninsula1. H. inermis exhibits several unique biological traits. For instance, it reaches sexual maturity within just 6–7 months and has the highest reproductive rate among cervids, with an average litter size of 2–3 and up to 5 offspring per birth2. Corresponding to its high reproductive capacity, fawns employ specialized survival strategies: newborns typically conceal themselves in dense vegetation and are nursed by the mother once every one to two days. Following nursing, the milk coagulates into curds through the action of various enzymes, enabling prolonged digestion and reducing feeding frequency3. As an antlerless deer, H. inermis holds a distinctive position in studies of cervid evolution and antler development4. Classical karyotypic studies have identified that the Chinese water deer possesses a diploid number of 2n = 70, with a predominantly acrocentric chromosome complement5. In recent decades, wild populations have declined sharply due to human disturbances and habitat loss, leading to its classification as a Vulnerable species on the IUCN Red List of Threatened Species6,7. Its adaptation to wetland plain ecosystems makes it a critical target for ecological conservation and research on reproductive and survival strategies. Moreover, H. inermis serves as an important model for understanding antler development and cervid husbandry. However, research on this species remains limited, and although two genome versions have been published, many genomic characteristics are still unclear.

Whole-genome sequencing is a widely used approach for deciphering complex genomic regions, comprehensively identifying structural variations, and elucidating chromosomal structure and evolution in mammals, as applied in models such as mice, sheep, and pigs. For H. inermis, a high-quality genome could provide insights into its unique reproductive behaviors and digestive adaptations. Nevertheless, existing genome assemblies contain numerous gaps and incomplete regions, particularly in repetitive areas like telomeres and centromeres. Telomeres, which reflect organismal health and lifespan8, play crucial roles in aging9 immunity10, and reproduction. Studies in humans11, Soay sheep12, and cattle bulls13 have demonstrated correlations between telomere length and reproductive fitness, underscoring the importance of telomere analysis in H. inermis given its exceptional reproductive rate. Centromeres are essential for kinetochore assembly during cell division, particularly in the rapid and stable cell divisions of early embryonic development14. Thus, complete centromere assembly is critical for understanding the high reproductive capacity of H. inermis. As of November 2025, GenBank hosts 26 chromosome-level cervid genomes spanning four subfamilies, including species such as sika deer (Cervus nippon), red deer (Cervus elaphus), white-lipped deer (Cervus albirostris), reindeer (Rangifer tarandus), white-tailed deer (Odocoileus virginianus), mule deer (Odocoileus hemionus), black muntjac (Muntiacus crinifrons), Reeves’s muntjac (Muntiacus reevesi), and Chinese water deer (H. inermis). However, few approach T2T standards; only four species—Cervus nippon15, Cervus elaphus, Muntiacus reevesi, and Odocoileus virginianus—have assemblies combining long-read sequencing and Hi-C data that approach T2T quality, and no published study has reported a complete T2T genome for any cervid.

In the present study, we used Pacific Biosciences (PacBio) HiFi sequencing, Oxford Nanopore Technologies (ONT) ultra-long sequencing, and Hi-C-assisted assembly to construct the first high-quality T2T genome of H. inermis. We also characterized the telomeres and centromeres of each chromosome, representing the first telomere-to-telomere genome in Cervidae. This work contributes to the conservation and evolutionary analysis of H. inermis and provides critical data for investigating the genetic mechanisms underlying its reproductive capacity and adaptive traits.

Materials and Methods

Ethics statement

The animals used in this study were approved by the ethics committee of laboratory animals of Shanghai Ocean University under the protocol number SHOU-DW-2023-216.

Sample collection and DNA extraction

Muscle tissue was sampled from a 2-year-old adult female Chinese water deer (Hydropotes inermis, 2n = 7016) that died of natural causes (pneumonia) on June 7, 2024. The individual was housed at Shanghai Huaxia Park (Fig. 1), a government-affiliated conservation and breeding center in Shanghai, China. High-molecular-weight genomic DNA (gDNA) was isolated using the Monarch® HMW DNA Extraction Kit for Tissue (New England Biolabs, T3060) following the manufacturer’s protocol. The extraction service was provided by Biomarker Technologies Co., Ltd (Beijing, China). DNA purity was assessed with a NanoDrop spectrophotometer (Thermo Fisher Scientific, USA), and concentration was accurately quantified using a Qubit fluorometer with the dsDNA HS Assay Kit (Invitrogen, USA, Q32854).

Fig. 1.

Fig. 1

Open in a new tab

A male H. inermis in the Shanghai Huaxia Park, from the same population of the sequenced individual.

Library construction and sequencing

Approximately 10 µg of gDNA was sheared into ~15 kb fragments using a Megaruptor® 2 system (Diagenode). A SMRTbell library was constructed using the SMRTbell Express Template Prep Kit 2.0 (Pacific Biosciences, 100-938-900). The sheared DNA was subjected to enzymatic reactions to remove single-stranded overhangs and repair DNA damage. The DNA ends were then polished and ligated with SMRTbell adapters. Following ligation, the library was treated with exonuclease to remove unligated products and purified with 0.45X AMPure PB beads. Library size distribution and concentration were assessed using the FEMTO Pulse system (Agilent Technologies) and Qubit 3.0 Fluorometer, respectively. A size selection step targeting 15–18 kb SMRTbells was performed using the Sage ELF system (Sage Science). The final library was bound with sequencing primer and Sequel II DNA polymerase and sequenced on a PacBio Sequel II system using one 8 M SMRT Cell with the Sequel II Sequencing Kit 2.0 (Pacific Biosciences, 101-820-200), collecting 1800-minute movies. Circular consensus sequencing (CCS) reads (HiFi reads) were generated from the subread data.

For the ONT ultra-long read sequencing, high molecular-weight gDNA was size-selected (>50 kb) using the Sage HLS HMW Library System (Sage Science, USA), The qualified gDNA was then fragmented and used to construct a library with the Ultra-Long DNA Sequencing Kit (Oxford Nanopore Technologies, SQK-ULK001) according to the manufacturer’s instructions. The library was purified using magnetic beads included in the kit. Approximately 400 ng of the final library was loaded and sequenced on a PromethION R9 flow cell (Oxford Nanopore Technologies)17.

The Hi-C libraries were constructed to scaffold the genome to chromosome-length. Muscle tissue was cross-linked with 1% formaldehyde, quenched with 0.2 M glycine, and then lysed to isolate nuclei. Chromatin was digested with 150 U of DpnII restriction enzyme (New England Biolabs) overnight at 37 °C. The resulting sticky ends were filled in with biotin-14-dATP and ligated. After reversing cross-links by incubation with 200 µg/mL proteinase K (Thermo) at 65 °C overnight, the DNA was purified using QIAamp DNA Mini Kit (Qiagen, 51304). The purified DNA was sheared to ~400 bp fragments, and biotin-labeled ligation junctions were captured using Dynabeads MyOne Streptavidin C1 beads (Thermo Fisher Scientific). The final library was sequenced on an Illumina NovaSeq 6000 platform with a paired-end 150 bp strategy. Following the above processes, a total of 103.97 GB (~27× coverage) of ONT data with a contig N50 of 100 kb and 147.0 Gb GB (~40× coverage) of PacBio HiFi CCS data with a contig N50 of 15.31 kb were generated, as well as 380.13 GB of Hi-C data (~110× coverage) (Table 1).

Table 1.

Statistics of the sequencing data.

Sequencing Platform Total bases (bp) Sequencing Depth (×) Mean read len. (bp) Maximum read len. (bp) N50 read len. (bp)
ONT 103,972,763,519 27 99,950 1,218,387 100,002
PacBio HiFi 147,000,117,901 40 15,440 47,853 15,312
Hi-C 380,130,711,271 110 149.60 150 150
NGS 188,015,337,296 45 149.30 150 150
RNA-seq 40,846,245,937 — 149.20 150 150
Open in a new tab

RNA-seq library sequencing

Total RNA was extracted from muscle tissue using the TRIzol Universal reagent (Tiangen). Poly(A)+ mRNA was enriched using oligo(dT) magnetic beads and fragmented. First-strand cDNA was synthesized using random hexamer primers, followed by second-strand synthesis. The double-stranded cDNA was purified with AMPure XP beads (Beckman Coulter). After end repair, A-tailing, and adapter ligation, the library was amplified by PCR. Library quality was assessed using a Qubit 2.0 Fluorometer and an Agilent 2100 Bioanalyzer. Quantification by qPCR ensured accurate pooling for sequencing, which was performed on an Illumina platform with a paired-end 150 bp strategy.

Genome survey and assembly

The genome size, heterozygosity and duplication ratio were estimated based on PacBio HiFi reads using Jellyfish (v2.3.0)18 and GenomeScope (v2.0)19. As determined from a 21-mers histogram, the estimated genome size was 3546.5 Mb, with a heterozygous ratio of 0.32% (Fig. 2). The estimated genome size was much greater than the previous assembly GCA_020226075.1 (a genome sized of 2.5 Gb) by the vast newly discovered repetitive sequences within telomeres and centromeres with PacBio HiFi long reads. A contig-level genome was assembled from the PacBio HiFi reads and ultra-long ONT reads using hifiasm (v0.19)20 with the parameters -t40–ul. The initial assembly was screened for contamination by aligning contigs against the NCBI GenBank nucleotide (nt), mitochondrial, and plastid databases using BLAST (v2.9.0+)21. The contigs identified as foreign sequences, mitochondria, or plastids were filtered out. The purified contigs were then scaffolded into chromosome-length assemblies using Hi-C data with the software Lachesis22 (Fig. 3). The resulting scaffolds underwent gap closing and telomere extension, utilizing the other assembly versions and long reads, to produce the final gapless T2T genome localized on 35 chromosomes (Table 2) with a genome size of 3450.42 Mb, which is highly consistent with the k-mer-based estimate, an N50 length of 101.09 Mb, and GC content of 39.11% (Fig. 4).

Fig. 2.

Fig. 2

Open in a new tab

GenomeScope plots of the 21-mer frequency. The curves in black indicate the distribution of the k-mer frequency. The red curves near the y-axis show the error of the k-mers, which normally occur at low coverage. The estimated genome sizes (Len) and heterozygosity rates (Het) are shown at the top of each plot, as is the percentage of uniquely mapped reads (Uniq), k-mer coverage (Kcov), PCR error rate (Err), percentage of PCR-derived duplicates (Dup), and k-mer length (K).

Fig. 3.

Fig. 3

Open in a new tab

Hi-C contact heatmap under 200-Kb resolution from the female H. inermis sequenced in this study. Strength of intra- and inter-chromosomal interaction frequencies is indicated by a color gradient scale displayed with strong interactions in red and weaker interactions in lighter colors.

Table 2.

Overview of the genome assemblies.

Assembled genome size (bp) 3,450,426,773
Number of Chromosomes 35
Number of gap chromosomes 0
Length of anchored chromosomes (bp) 3,450,426,773
Contig N50 Length (bp) 96,144,934
Number of contigs 35
Number of telomeres 60
Number of centromeres 35
QV 52.56
Complete BUSCO (%) 99.40
Open in a new tab

Fig. 4.

Fig. 4

Open in a new tab

Landscape of the genome calculated in 500-kb non-overlapping windows. The circles from outside to inside: (a) Length scale of the chromosomes, (b) TE density, (c) SSR density, (d) Gene density, (e) GC content. The lines in the central area of the circles link the syntenic regions.

Localization of telomeres and centromeres

Telomere and centromere sequences were identified to investigate genomic structure of Hydropotes inermis. The telomere repeat units were predicted with TIDK (v0.1.5)23, which determined the consensus sequences as 5′-AACCCT-3′ and 5′-AGGGTT-3′. A total of 60 telomeres were detected in the 35 chromosomes scaffolds (Table 3). Centromeric regions were characterized using Centromics (v0.3)24. Initially, long-read sequencing data (third-generation reads) were utilized to detect potential centromere repeats. These repeats were then aligned back to the genome assembly to obtain accurate centromere positions and sequences. Among the 35 chromosomes, the centromeric and pericentromeric heterochromatin sequences were predominantly located at the chromosomal ends. The centromere positioning is generally consistent with previous karyotypic study, as predominantly acrocentric. The centromeres had an average length of 21.59 Mb (Fig. 5) and an average GC content of 47.03%. Chinese water deer is an ancient species of the Cervidae family, characterized by over 20% centromeres and pericentromeric heterochromatin regions, which usually resulted in an under-esimation of the genome size in the k-mer analysis using the illumina short reads.

Table 3.

Localization of identified telomeres and centromeres in the H. inermis assembly.

Chr Chr Len. (bp) Telomeres Centromeres
5p start - end (bp) 5p len. (bp) 3p start - end (bp) 3p len. (bp) Start - end (bp) Len. (bp)
01 142,486,036 1–33317 33,317 142429758–142486036 56,279 33347–24392220 24,358,874
02 139,917,033 0–34764 34,765 139901126–139917013 15,888 34765–26042509 26,007,745
03 129,855,783 93–17628 17,536 129852022–129855782 3,761 33441–24701931 24,668,491
04 128,286,804 1–10911 10,911 128285213–128286050 838 102312476–128286804 25,974,329
05 118,335,903 1–1713 1,713 118334674–118335819 1,146 97660438–118319326 20,658,889
06 114,892,324 1–12204 12,204 114878568–114892324 13,757 95727766–114878567 19,150,802
07 123,510,549 0–56578 56,579 123504389–123510548 6,160 56579–30355153 30,298,575
08 120,181,151 Not detected 120169286–120181133 11,848 7346746–37752181 30,405,436
09 105,662,057 0–6696 6,697 105654058–105661956 7,899 80227035–105638305 25,411,271
10 111,594,072 0–40112 40,113 111588990–111594071 5,082 40129–31974453 31,934,325
11 107,605,195 Not detected 107594074–107605185 11,112 2–27731390 27,731,389
12 99,274,201 1–33230 33,230 99268445–99274201 5,757 33246–22278995 22,245,750
13 101,097,117 9–5720 5,712 101057202–101097116 39,915 77126322–101057199 23,930,878
14 93,439,062 10–5117 5,108 93431724–93438974 7,251 69457358–93411238 23,953,881
15 93,400,245 60–6366 6,307 93394812–93400245 5,434 69024376–93382452 24,358,077
16 96,144,934 0–19584 19,585 96135052–96144933 9,882 37058–29497914 29,460,857
17 94,525,344 1–2455 2,455 Not detected 63031266–94525344 31,494,079
18 95,967,640 18–832 815 95960040–95967639 7,600 14510361–30392012 15,881,652
19 103,620,410 1–2960 2,960 103609309–103620410 11,102 15319–30067007 30,051,689
20 94,193,941 1–14550 14,550 94184650–94193941 9,292 14551–28994731 28,980,181
21 83,252,625 40–3566 3,527 83246616–83252625 6,010 60647415–83234256 22,586,842
22 69,199,717 133–6914 6,782 Not detected 61151140–69199710 8,048,571
23 79,928,934 0–1879 1,880 Not detected 59174686–79491370 20,316,685
24 85,257,821 18–3765 3,748 85250386–85257820 7,435 58309975–85176830 26,866,856
25 77,650,735 Not detected 77645695–77650734 5,040 0–23365546 23,365,547
26 86,409,184 1–3609 3,609 86401413–86409184 7,772 8812107–32321602 23,509,496
27 72,366,311 1–7458 7,458 Not detected 52753608–71874121 19,120,514
28 69,842,294 Not detected 69834965–69842293 7,329 3833360–19025793 15,192,434
29 77,787,461 1–1672 1,672 77780911–77787442 6,532 14463758–29357440 14,893,683
30 49,035,948 Not detected 49029487–49035947 6,461 1117309–1125178 7,870
31 80,783,886 1–4798 4,798 80781276–80783886 2,611 44900414–57610761 12,710,348
32 78,943,777 1–3902 3,902 78929141–78943777 14,637 44506073–65180646 20,674,574
33 72,921,780 1–7543 7,543 72919966–72920771 806 45332929–56218804 10,885,876
34 66,426,552 0–990 991 66423141–66426551 3,411 12850711–26048597 13,197,887
X 186,629,947 Not detected 186614263–186629947 15,685 1–7356090 7,356,090
Open in a new tab

Fig. 5.

Fig. 5

Open in a new tab

Distribution of centromere and telomeres in each chromosome scaffold of the H. inermis assembly.

Annotation of repetitive sequences

Repetitive sequences were annotated by integrating de novo prediction with homology-based approaches. De novo prediction was performed using RepeatModeler2 (v2.0.1)25, which leveraged RECON (v1.0.8)26 and RepeatScout (v1.0.6)27 to identify repetitive elements. The initial predictions were then classified using RepeatClassifier with the Dfam database (v3.5)28 to categorize known and novel repeats. LTR retrotransposons were specifically annotated using LTR_retriever (v2.9.0)29. This tool integrated outputs from LTRharvest (v1.5.10)30 and LTR_FINDER (v1.07)31 to generate comprehensive LTR element predictions. The de novo prediction results from RepeatModeler2 and LTR_retriever were merged with known repetitive sequences from Dfam28 and Repbase32 databases. Redundant entries were removed to construct a species-specific repeat library. The compiled repeat library was used to annotate transposable elements (TEs) across the genome using RepeatMasker (v4.1.2)33. This step identified and masked repetitive regions based on the custom library. The annotation results show that repetitive sequence accounts for 33.36% of the genome. Repeat sequences comprise 33.36% of the Hydropotes inermis genome. Among these, Long Interspersed Nuclear Elements (LINEs), Long Terminal Repeat (LTR) elements, DNA transposons, and Short Interspersed Nuclear Elements (SINEs) account for 16.40%, 7.33%, 5.56%, and 2.72%, respectively (Table 4). These proportions are notably lower than in other Cervidae species. In particular, comparison with the widely distributed red deer (Cervus elaphus) reveals lower repetitive content in the Chinese water deer: LINEs (16.40% vs. 37.76%), LTR retrotransposons (7.33% vs. 11.70%), and SINEs (2.72% vs. 4.22%)34. The distinctive repeat landscape of the H. inermis genome—characterized by reduced proportions of LINEs and LTRs—may reflect genomic adaptations associated with its antlerless phenotype and semi-aquatic lifestyle.

Table 4.

Annotation of repetitive sequences.

Type Quantity Length of sequence (bp) Percentage of the genome (%)
Total TEs 3,836,262 1,104,604,612 32.01
LINE 1,202,364 565,948,731 16.40
LTR 670,491 252,868,432 7.33
SINE 522,238 93,978,324 2.72
DNA 1,441,169 191,809,125 5.56
Tandem Repeats 1,241,794 39,983,513 1.15
Satellites 17,883 6,424,339 0.19
Others 1,488 237,211 0.01
Sum 5,097,427 1,151,249,675 33.36
Open in a new tab

Prediction of protein-coding genes

To predict protein-coding genes in the Hydropotes inermis genome, we employed an integrated approach combining de novo, homology-based, and transcriptome-based strategies. De novo prediction was conducted using Augustus (v3.1.0)35 and SNAP (2006-07-28)36. Homology-based prediction was performed with GeMoMa (v1.7)37, utilizing reference genomes from closely related species, including Cervus elaphus, Bos taurus, Muntiacus reevesi, and Rangifer tarandus. Two strategies were employed for Transcriptome-Based Prediction. First, transcripts were assembled using Hisat2 (v2.1.0)38 and Stringtie (v2.1.4)39, followed by gene prediction with GeneMarkS-T (v5.1)40. Second, transcripts were assembled using RNA-Bloom (v2.0.0)41, and gene prediction was carried out with PASA (v2.4.1)42. The gene predictions from the above three methods (de novo, homology-based, and transcriptome-based) were integrated using EVM (v1.1.1)43. The integrated gene set was further refined and modified with PASA (v2.4.1) to improve accuracy and completeness. Then, predicted gene sequences were functionally annotated against multiple databases, including NR, eggNOG44, GO, KEGG45, TrEMBL46, KOG, SWISS-PROT46, and Pfam47. This comprehensive annotation provided insights into gene functions, pathways, and protein domains. Finally, a total of 24,398 protein-coding genes were predicted, with an average gene length of 40,297.17 bp (Fig. 6). Among these, 95.03% of the genes were successfully annotated in at least one of the databases.

Fig. 6.

Fig. 6

Open in a new tab

Comparison of gene structure among different bovid (Bos taurus) and cervid (Cervus elaphus, Hydropotes inermis, Muntiacus reevesi, and Rangifer tarandus) species, encompassing length of genes, CDSs (coding sequences), exons. and introns.

Prediction of conserved non-coding RNAs

Non-coding RNAs (ncRNAs) were predicted using distinct strategies based on their structural characteristics. The software tRNAscan-SE (v1.3.1) was used to identify tRNAs, while rRNAs were primarily predicted using barrnap (v0.9). For miRNA, snoRNA, and snRNA, these ncRNAs were predicted based on the Rfam database (v14.9)48 and inferred using Infernal (v1.1)49. Overall, this annotation pipeline resulted in the prediction of 193,279 tRNAs, 546 rRNAs, and 239 miRNAs, account for 0.4% of the genome size.

Data Records

All the sequencing data generated in this study have been deposited in the Genome Sequence Archive (GSA) in National Data Center (https://ngdc.cncb.ac.cn/gsa) under the BioProject accession PRJCA050706 and GSA accession CRA03334250. These include the PacBio HiFi sequencing data and the ONT ultra-long read data (FASTQ format; Run accessions CRR2303155 and CRR2303154), short-read Hi-C data and next-generation sequencing data (FASTQ format; Run accessions CRR2303157 and CRR2303156), and RNA-Seq datasets obtained from short-read sequencing (FASTQ format; Run accessions CRR2303158). These raw sequencing data have also been co-deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) under the accession SRP63115151.The final T2T gapless genome assembly of Hydropotes inermis is openly available in public repositories. Specifically, the assembly sequence has been deposited in the NCBI GenBank database52. The associated gene structure annotation files, coding sequences, and protein sequences have been deposited in Figshare53. All data are publicly available.

Technical Validation

The accuracy and integrity of the Hydropotes inermis genome assembly were rigorously validated using multiple complementary approaches. First, Hi-C analysis revealed a high degree of consistency across all chromosomes, confirming the precision of contig ordering and orientation during genome assembly. Additionally, structural integrity was demonstrated by the identification of 35 centromeres and 60 telomeres, which correspond to the expected chromosomal features in a telomere-to-telomere (T2T) assembly. To assess sequence-level accuracy, read mapping analyses were performed. Second-generation short-reads achieved a mapping rate of 99.59%, coverage of 99.96%, and an average sequencing depth of 45 × (Fig. 7). For third-generation circular consensus sequencing (CCS) reads, the mapping rate was 99.99%, with coverage of 99.98% and an average depth of 40 × . Similarly, Oxford Nanopore Technologies (ONT) long-reads exhibited a mapping rate of 98.02%, coverage of 99.96%, and an average depth of 27×. These results indicate comprehensive and reliable sequence representation across the genome. Further quality metrics supported the assembly’s robustness. The genome achieved a quality value (QV) of 52.56, reflecting high base-level accuracy. BUSCO assessment against the single-copy orthologous gene set (Artiodactyla_odb12) from OrthoDB identified 99.4% and 95.4% of the 12,594 conserved genes, underscoring the genome’s completeness and functional relevance (Table 5).Most of the used softwares were list in the (Table 6). Overall, these validation steps confirmed that the Hydropotes inermis genome assembly was of high quality, providing a solid foundation for studies on wetland adaptation in small cervids and reproductive biology in deer species.

Fig. 7.

Fig. 7

Open in a new tab

Sequencing coverage of HiFi, ONT, and NGS reads on each chromosome scaffold of the H. inermis assembly. Coverage depth is plotted with 50-kb windows.

Table 5.

Completeness test of genome assemblies and annotations assessed by BUSCO.

Type Assembly Annotation
Orthologs Percentage (%) Orthologs Percentage (%)
Complete BUSCOs 12517 99.4% 12,020 95.4%
Complete Single-Copy BUSCOs 12347 98.0% 11,873 94.3%
Complete Duplicated BUSCOs 170 1.3% 147 1.2%
Fragmented BUSCOs 22 0.2% 238 1.9%
Missing BUSCOs 55 0.4% 336 2.7%
Total BUSCO groups searched 12594 100% 12,594 100%
Open in a new tab

Table 6.

Software and parameters used for genome assembly, annotation, and validation.

Software Version Parameter Citation
Augustus v3.1.0 autoAug.pl default 35
barrnap v0.9 –kingdom euk–threads 1 54
BLAST v2.9.0 +  default 21
Centromics v0.3 default 24
diamond v0.9.29.130 diamond blastp–masking 0 -e 0.001 55
eggnog-mapper v1.0 emapper.py -m diamond 56
EVidenceModeler v1.1.1 default 43
GeMoMa v1.7 run.sh mmseqs 37
GenBlastA v1.0.4 genblasta -P wublast -pg tblastn 57
GeneMarkS-T v5.1 default 40
GeneWise v2.4.1 genewise -both -pseudo 58
GenomeScope v2.0 -k 21 -p 2 19
hifiasm v0.19 default;–ul 20
Hisat2 v2.1.0 –dta -p 10 38
Infernal v1.1 cmscan–cpu 3–rfam 49
InterProScan v5.34-73.0 interproscan.sh -iprlookup -pa -f xml -dp -t p -cpu 10 59
Jellyfish v2.3.0 count -C -m 21;histo -h 10000000000 18
Lachesis v1 default 22
LTR_retriever v2.9.0 default 29
PASA v2.4.1 default 42
RepeatMasker v4.1.2 repeatmasker -nolow -no_is -norna -engine wublast -parallel 8 -qq 33
RepeatModeler v2.0.1 BuildDatabase -name & RepeatModeler -pa 12 25
SNAP 2006/7/28 default 36
Stringtie v2.1.4 -p 2 39
TIDK v0.1.5 default 23
tRNAscan-SE v1.3.1 default 60
Open in a new tab

Acknowledgements

We appreciated Ms. Chen Xiaorong, Mr. Nie Dayi, and Shanghai Huaxia Park for their assistance in the sampling. This study was funded by the Inner Mongolia Autonomous Region “Jiebangguashuai” project of China (No. 2022JBGS0021).

Author contributions

Ying Lu and Min Chen conceived and designed the study. Min Chen and Qiuting Chen collected the samples. Haotian Wu and Hengquan Zhao performed the data analyses. Ying Lu, Haotian Wu, and Qiuting Chen wrote the manuscript. Ying Lu and Min Chen revise the manuscript.

Data availability

All the sequencing data generated in this study have been deposited in the Genome Sequence Archive in National Data Center under the BioProject accession PRJCA050706 and GSA accession CRA033342 (https://ngdc.cncb.ac.cn/gsa/browse/CRA033342). These include PacBooio HiFi sequencing data and ONT ultra-long read data (FASTQ format; Run accessions CRR2303155 and CRR2303154), short read Hi-C data and next-generation sequencing data (FASTQ format; Run accessions CRR2303157 and CRR2303156), and RNA-Seq datasets obtained from short-read sequencing (FASTQ format; Run accessions CRR2303158). These raw sequencing data have also been co-deposited in the NCBI Sequence Read Archive under the accession SRP631151 (https://identifiers.org/ncbi/insdc.sra:SRP631151).The final T2T gapless genome assembly of Hydropotes inermis is openly available in public repositories. Specifically, the assembly sequence has been deposited in the NCBI GenBank database (https://identifiers.org/ncbi/insdc:JBQQXF000000000). The associated gene structure annotation files, coding sequences, and protein sequences have been deposited in Figshare (10.6084/m9.figshare.30655721.v3). All data are publicly available.

Code availability

All commands and pipelines used in data processing were executed according to the manual and protocols of the corresponding bioinformatic software. No specific code has been developed for this study.

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.

These authors contributed equally: Haotian Wu, Hengquan Zhao.

Contributor Information

Min Chen, Email: mchen@bio.ecnu.edu.cn.

Ying Lu, Email: yinglu@shou.edu.cn.

References

  • 1.Tan, G., Zhang, G. & Lu, Q. Daily Activity Rhythm and Time Budget of Captive Female Chinese Water Deer in Lactation. Sichuan Journal of Zoology34, 548–554 (2015). [Google Scholar]
  • 2.Li, Y. et al. Research Advances on the Ecology of Chinese Water Deer (Hydropotes inermis). Chinese Journal of Wildlife34, 270–273 (2013). [Google Scholar]
  • 3.Gu, J. & Gu, G. The breeding techniques of Chinese water deer. China Animal Industry. 82-83 (2013).
  • 4.Ba, H.X., Chen, M. & Li, C.Y. Cross-Species Analysis Reveals Co-Expressed Genes Regulating Antler Development in Cervidae. Front Genet. 13, (2022). [DOI] [PMC free article] [PubMed]
  • 5.Fontana, F. & Rubini, M. Chromosomal evolution in cervidae. Biosystems24, 157–174 (1990). [DOI] [PubMed] [Google Scholar]
  • 6.Zhao, L., Ma, H., Yin, J. & Liang, Z. Study on the Apparent Digestibility of Protein in Hydropotes inermis Fed with Different Types of Diets. Journal of Anhui Agricultural Sciences42, 2941–2957 (2014). [Google Scholar]
  • 7.Sun, Y. et al. Habitat Selection: Autumn and Winter Behavioral Preferences of Water Deer (Hydropotes inermis) in Northeast China. Sustainability-Basel. 15, (2023).
  • 8.Zhang, N., Baker, E.C., Welsh, T.J.r. & Riley, D.G. Telomere Dynamics in Livestock. Biology-Basel. 12, (2023). [DOI] [PMC free article] [PubMed]
  • 9.Ji, G.Z., Liu, K., Okuka, M., Liu, N. & Liu, L. Association of telomere instability with senescence of porcine cells. Bmc Cell Biol. 13, (2012). [DOI] [PMC free article] [PubMed]
  • 10.Katepalli, M. P., Adams, A. A., Lear, T. L. & Horohov, D. W. The effect of age and telomere length on immune function in the horse. Dev Comp Immunol32, 1409–1415 (2008). [DOI] [PubMed] [Google Scholar]
  • 11.Santa Rocca, M., Foresta, C. & Ferlin, A. Telomere length: lights and shadows on their role in human reproduction. Biol Reprod100, 305–317 (2019). [DOI] [PubMed] [Google Scholar]
  • 12.Ravindran, S. et al. The association between female reproductive performance and leukocyte telomere length in wild Soay sheep. Mol Ecol31, 6184–6196 (2022). [DOI] [PubMed] [Google Scholar]
  • 13.Iannuzzi, A. et al. Evaluation of bovine sperm telomere length and association with semen quality. Theriogenology158, 227–232 (2020). [DOI] [PubMed] [Google Scholar]
  • 14.Grzonka, M., Black, B. & Lampson, M.A. Centromere regulation in the germline and early embryo. Curr Opin Genet Dev. 94, (2025). [DOI] [PMC free article] [PubMed]
  • 15.Yang, Y.A. et al. A high-quality Chromosome-level reference genome assembly of white-lipped deer (Przewalskium albirostris). Sci. Data. 12, (2025). [DOI] [PMC free article] [PubMed]
  • 16.Yin, Y. et al. Molecular mechanisms and topological consequences of drastic chromosomal rearrangements of muntjac deer. Nat Commun. 12, (2021). [DOI] [PMC free article] [PubMed]
  • 17.Wick, R.R., Judd, L.M., Gorrie, C.L. & Holt, K.E. Completing bacterial genome assemblies with multiplex MinION sequencing. Microb Genomics. 3, (2017). [DOI] [PMC free article] [PubMed]
  • 18.Marcais, G. & Kingsford, C. A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. Bioinformatics27, 764–770 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ranallo-Benavidez, T.R., Jaron, K.S. & Schatz, M.C. GenomeScope 2.0 and Smudgeplot for reference-free profiling of polyploid genomes. Nat Commun. 11, (2020). [DOI] [PMC free article] [PubMed]
  • 20.Cheng, H. Y., Concepcion, G. T., Feng, X. W., Zhang, H. W. & Li, H. Haplotype-resolved de novo assembly using phased assembly graphs with hifiasm. Nat Methods18, 170-+ (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Camacho, C. et al. BLAST plus: architecture and applications. Bmc Bioinformatics. 10, (2009). [DOI] [PMC free article] [PubMed]
  • 22.Burton, J. N. et al. Chromosome-scale scaffolding of de novo genome assemblies based on chromatin interactions. Nat Biotechnol31, 1119-+ (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Brown, M.R., de la Rosa, P.M.G. & Blaxter, M. tidk: a toolkit to rapidly identify telomeric repeats from genomic datasets. Bioinformatics. 41, (2025). [DOI] [PMC free article] [PubMed]
  • 24.Zhang, R. Source code for: Centromics: visualing centromeres with multiple omics data. githubhttps://github.com/zhangrengang/Centromics (2023).
  • 25.Flynn, J. M. et al. RepeatModeler2 for automated genomic discovery of transposable element families. P Natl Acad Sci USA.117, 9451–9457 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Bao, Z. R. & Eddy, S. R. Automated de novo identification of repeat sequence families in sequenced genomes. Genome Research12, 1269–1276 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Price, A. L., Jones, N. C. & Pevzner, P. A. De novo identification of repeat families in large genomes. Bioinformatics21, i351–i358 (2005). [DOI] [PubMed] [Google Scholar]
  • 28.Storer, J., Hubley, R., Rosen, J., Wheeler, T.J. & Smit, A.F. The Dfam community resource of transposable element families, sequence models, and genome annotations. Mobile DNA-Uk. 12, (2021). [DOI] [PMC free article] [PubMed]
  • 29.Ou, S. J. & Jiang, N. LTR_retriever: A Highly Accurate and Sensitive Program for Identification of Long Terminal Repeat Retrotransposons. Plant Physiol176, 1410–1422 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Ellinghaus, D., Kurtz, S. & Willhoeft, U. LTRharvest, an efficient and flexible software for de novo detection of LTR retrotransposons. Bmc Bioinformatics9, 18 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Xu, Z. & Wang, H. LTR_FINDER: an efficient tool for the prediction of full-length LTR retrotransposons. Nucleic acids research35, W265–W268 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Bao, W., Kojima, K. K. & Kohany, O. Repbase Update, a database of repetitive elements in eukaryotic genomes. Mobile DNA-Uk6, 11 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Chen, N. Using Repeat Masker to identify repetitive elements in genomic sequences. Current protocols in bioinformatics5, 4.10. 11–14.10. 14 (2004). [DOI] [PubMed] [Google Scholar]
  • 34.Ba, H. et al. Chromosome-level genome assembly of Tarim red deer, Cervus elaphus yarkandensis. Sci. Data.7, 187 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Stanke, M., Diekhans, M., Baertsch, R. & Haussler, D. Using native and syntenically mapped cDNA alignments to improve de novo gene finding. Bioinformatics24, 637–644 (2008). [DOI] [PubMed] [Google Scholar]
  • 36.Korf, I. Gene finding in novel genomes. Bmc Bioinformatics5, 59 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Keilwagen, J., Hartung, F. & Grau, J. GeMoMa: homology-based gene prediction utilizing intron position conservation and RNA-seq data. in Gene prediction: Methods and protocols 161-177 (Springer, 2019). [DOI] [PubMed]
  • 38.Kim, D., Paggi, J. M., Park, C., Bennett, C. & Salzberg, S. L. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol37, 907–915 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Pertea, M. et al. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat Biotechnol33, 290–295 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Tang, S., Lomsadze, A. & Borodovsky, M. Identification of protein coding regions in RNA transcripts. Nucleic acids research43, e78–e78 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Nip, K. M. et al. RNA-Bloom enables reference-free and reference-guided sequence assembly for single-cell transcriptomes. Genome research30, 1191–1200 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Haas, B. J. et al. Improving the Arabidopsis genome annotation using maximal transcript alignment assemblies. Nucleic acids research31, 5654–5666 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Haas, B. J. et al. Automated eukaryotic gene structure annotation using EVidenceModeler and the Program to Assemble Spliced Alignments. Genome biology9, R7 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Huerta-Cepas, J. et al. eggNOG 5.0: a hierarchical, functionally and phylogenetically annotated orthology resource based on 5090 organisms and 2502 viruses. Nucleic acids research47, D309–D314 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Kanehisa, M., Furumichi, M., Tanabe, M., Sato, Y. & Morishima, K. KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic acids research45, D353–D361 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Boeckmann, B. et al. The SWISS-PROT protein knowledgebase and its supplement TrEMBL in 2003. Nucleic acids research31, 365–370 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Finn, R. D. et al. Pfam: clans, web tools and services. Nucleic acids research34, D247–D251 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Griffiths-Jones, S. et al. Rfam: annotating non-coding RNAs in complete genomes. Nucleic acids research33, D121–D124 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Nawrocki, E. P. & Eddy, S. R. Infernal 1.1: 100-fold faster RNA homology searches. Bioinformatics29, 2933–2935 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.CNCB-NGDC Genome Sequence Archivehttps://ngdc.cncb.ac.cn/gsa/browse/CRA033342 (2025).
  • 51.NCBI Sequence Read Archivehttps://identifiers.org/ncbi/insdc.sra:SRP631151 (2025).
  • 52.NCBI GenBankhttps://identifiers.org/ncbi/insdc:JBQQXF000000000 (2026).
  • 53.Zhao, H., Wu, H. & Lu, Y. A telomere-to-telomere genome assembly of Chinese water deer (Hydropotes inermis). Figshare10.6084/m9.figshare.30655721.v3 (2025). [DOI] [PMC free article] [PubMed]
  • 54.Loman, T. Lund University. A novel method for predicting ribosomal RNA genes in prokaryotic genomes, (2017).
  • 55.Buchfink, B., Xie, C. & Huson, D. H. Fast and sensitive protein alignment using DIAMOND. Nat Methods12, 59–60 (2015). [DOI] [PubMed] [Google Scholar]
  • 56.Cantalapiedra, C. P., Hernández-Plaza, A., Letunic, I., Bork, P. & Huerta-Cepas, J. eggNOG-mapper v2: functional annotation, orthology assignments, and domain prediction at the metagenomic scale. Molecular biology and evolution38, 5825–5829 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.She, R., Chu, J. S.-C., Wang, K., Pei, J. & Chen, N. GenBlastA: enabling BLAST to identify homologous gene sequences. Genome research19, 143–149 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Birney, E., Clamp, M. & Durbin, R. GeneWise and genomewise. Genome research14, 988–995 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Jones, P. et al. InterProScan 5: genome-scale protein function classification. Bioinformatics30, 1236–1240 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Lowe, T. M. & Eddy, S. R. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic acids research25, 955–964 (1997). [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.

Data Citations

  1. CNCB-NGDC Genome Sequence Archivehttps://ngdc.cncb.ac.cn/gsa/browse/CRA033342 (2025).
  2. NCBI Sequence Read Archivehttps://identifiers.org/ncbi/insdc.sra:SRP631151 (2025).
  3. NCBI GenBankhttps://identifiers.org/ncbi/insdc:JBQQXF000000000 (2026).
  4. Zhao, H., Wu, H. & Lu, Y. A telomere-to-telomere genome assembly of Chinese water deer (Hydropotes inermis). Figshare10.6084/m9.figshare.30655721.v3 (2025). [DOI] [PMC free article] [PubMed]

Data Availability Statement

All the sequencing data generated in this study have been deposited in the Genome Sequence Archive in National Data Center under the BioProject accession PRJCA050706 and GSA accession CRA033342 (https://ngdc.cncb.ac.cn/gsa/browse/CRA033342). These include PacBooio HiFi sequencing data and ONT ultra-long read data (FASTQ format; Run accessions CRR2303155 and CRR2303154), short read Hi-C data and next-generation sequencing data (FASTQ format; Run accessions CRR2303157 and CRR2303156), and RNA-Seq datasets obtained from short-read sequencing (FASTQ format; Run accessions CRR2303158). These raw sequencing data have also been co-deposited in the NCBI Sequence Read Archive under the accession SRP631151 (https://identifiers.org/ncbi/insdc.sra:SRP631151).The final T2T gapless genome assembly of Hydropotes inermis is openly available in public repositories. Specifically, the assembly sequence has been deposited in the NCBI GenBank database (https://identifiers.org/ncbi/insdc:JBQQXF000000000). The associated gene structure annotation files, coding sequences, and protein sequences have been deposited in Figshare (10.6084/m9.figshare.30655721.v3). All data are publicly available.

All commands and pipelines used in data processing were executed according to the manual and protocols of the corresponding bioinformatic software. No specific code has been developed for this study.

Articles from Scientific Data are provided here courtesy of Nature Publishing Group

RESOURCES