Molecular epidemiology and whole-genome characterization of human bocavirus in hospitalized children with pneumonia in Suzhou, China, 2022–2025

Abstract

Background

Human bocavirus (HBoV) is frequently detected in pediatric respiratory infections, yet its molecular epidemiology and genomic diversity remain incompletely characterized in many regions. We investigated HBoV circulation among hospitalized children with pneumonia in Suzhou, China, from 2022 to 2025 and assessed VP1/VP2 variation for potential physicochemical and functional impacts.

Methods

Oropharyngeal swabs were collected from hospitalized pediatric pneumonia cases and screened for HBoV using multiplex real-time PCR. HBoV-positive samples with Ct < 30 were subjected to amplicon-based whole-genome sequencing using a 17-primer-pair scheme covering > 98% of the genome, followed by assembly and comparative genomic analysis. Genotyping was performed by maximum-likelihood phylogenetic inference based on VP1/VP2 sequences. Deduplicated VP1/VP2 amino acid sequences were analyzed for hydrophobicity (ProtScale, Kyte–Doolittle, window = 9) and predicted functional impact (PROVEAN, cutoff − 2.5).

Results

Among 1,372 specimens, 83 were HBoV-positive (6.05%), with increased detection in autumn (September–November). Co-detection of other respiratory pathogens occurred in 59/83 (71.08%) samples. The most frequently co-detected pathogens were human rhinovirus (27/83, 32.53%), Haemophilus influenzae (13/83, 15.66%), respiratory syncytial virus (13/83, 15.66%), and Streptococcus pneumoniae (13/83, 15.66%). We obtained 45 complete HBoV genomes (6 in 2022, 5 in 2023, 15 in 2024, and 19 in 2025). VP1/VP2 phylogenetic analysis classified all strains as HBoV-1 and further into subclades Ⅰa (n = 6) and Ⅰb (n = 39), with Ⅰb predominating. Nucleotide substitutions and amino acid changes were mainly concentrated in the VP1/VP2 region. Hydrophobicity profiles of VP1/VP2 were largely conserved relative to reference strains, with only local shifts adjacent to substitution sites. PROVEAN predicted major substitutions (e.g., R17K, A149T, T590S) as neutral.

Conclusions

HBoV was detected in 6.05% of hospitalized children with pneumonia in Suzhou during 2022–2025, with frequent co-detections. Genomic surveillance based on 45 complete genomes indicates the predominance of HBoV-1 subclade Ⅰb, and VP1/VP2 variation was mainly capsid-region–enriched but exhibited overall stable hydrophobicity patterns and largely neutral predicted functional impacts. These findings provide a genomic baseline for continued molecular surveillance in the region.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12985-026-03149-7.

Background

Human bocavirus (HBoV), a member of the family Parvoviridae (genus Bocaparvovirus), has been increasingly recognized as a common pathogen detected in pediatric respiratory tract infections since its initial identification in lower respiratory samples. HBoV is frequently detected in young children presenting with cough, fever, wheezing [1], and may be associated with lower respiratory tract disease such as pneumonia and bronchitis [2–3], making it an important target for pediatric clinical and public health surveillance [4].

To date, HBoV is classified into four species (HBoV1–4). HBoV1 is predominantly associated with respiratory infections, whereas HBoV2–4 are more often linked to gastrointestinal disease [5–6]. The HBoV1 genome is approximately 5.2 kb in length and contains three open reading frames encoding non-structural proteins (NS1 and NP1) and structural capsid proteins (VP1/VP2), which are central to viral entry, assembly, and host interactions [7]. However, the precise pathogenic role of HBoV in respiratory disease remains debated, partly due to frequent co-detection with other respiratory pathogens and limited routine genomic surveillance in many settings [8–9].

In this study, we conducted molecular surveillance of HBoV among hospitalized children with pneumonia in Suzhou, China (2022–2025), and performed amplicon-based whole-genome sequencing (Ct < 30) using a 17-primer-pair scheme covering > 98% of the viral genome to characterize genomic diversity, VP1/VP2 phylogeny, and the potential physicochemical/ functional impacts of VP1/VP2 amino-acid variations.

Materials and methods

Study design and specimens

This retrospective study included oropharyngeal swab specimens collected from hospitalized children diagnosed with pneumonia at Suzhou Children’s Hospital, Suzhou, China, from 2022 to 2025. All specimens were submitted for routine respiratory pathogen testing. Demographic information (sex and age group) and specimen collection time were recorded for epidemiological analysis.

Nucleic acid extraction and multiplex real-time PCR screening

Total nucleic acids were extracted from throat swab specimens using a nucleic acid extraction kit (bio-germ, China). Multiplex real-time PCR screening for respiratory pathogens, including HBoV, was performed using a respiratory pathogen multiplex real-time fluorescence PCR detection system (Zhuocheng, China). The target pathogens tested included influenza viruses A and B (FluA and FluB), human coronaviruses (HCoV), parainfluenza virus (PIV), adenovirus (AdV), human rhinovirus (HRV), human metapneumovirus (hMPV), human bocavirus (HBoV), enterovirus (EV), respiratory syncytial virus (RSV), Haemophilus influenzae (Hi), Streptococcus pneumoniae (Spn), Chlamydia pneumoniae (CP), and Mycoplasma pneumoniae (MP). HBoV-positive samples were identified based on cycle threshold (Ct) values. For whole-genome sequencing, only HBoV-positive specimens with Ct < 30 were selected.

Amplicon-based whole-genome sequencing of HBoV

Primer design and multiplex PCR

To obtain complete HBoV genomes, an amplicon-based sequencing strategy was developed. A total of 20 complete HBoV genome sequences were retrieved from GenBank and used as templates for primer design. Seventeen primer pairs were designed to generate overlapping amplicons spanning the entire HBoV genome, achieving > 98% theoretical genome coverage. Primer sequences are provided in Supplementary Table S1.

Multiplex PCR was performed in a single tube using 2× Phanta Max Master Mix (Vazyme, China).The thermal cycling conditions were: 98 °C for 30 s; 35 cycles of 98 °C for 15 s, 63 °C for 5 min, and 65 °C for 1 min; followed by a final extension at 65 °C for 5 min.

Library preparation and sequencing

Amplicon products were purified and used for sequencing library preparation with the Illumina DNA Prep kit (Illumina, USA) according to the manufacturer’s instructions. Libraries were sequenced on an Illumina MiSeq platform.

Genome assembly

Raw sequencing reads were quality-filtered, and consensus genomes were assembled using CLC Genomics Workbench v20.0.3 (Qiagen, Denmark).Prototype strains ST1 (DQ000495) and ST2 (DQ000496) were used as reference strains for subclades Ⅰa and Ⅰb, respectively. Quality control and read-processing steps were performed using the built-in workflows in CLC Genomics Workbench v20.0.3. Briefly, reads were automatically filtered/trimmed according to default quality settings, and the resulting reads were used for mapping/assembly and consensus generation. For each consensus genome, we inspected breadth of genome coverage, read-depth distribution, and the proportion of ambiguous bases (Ns). Consensus sequences were called using the CLC majority-vote approach, and low-support positions were masked as Ns to minimize spurious substitutions. Only high-quality near-complete genomes passing these QC checks were retained for downstream phylogenetic and comparative analyses.

Variation profiling and positional entropy

Nucleotide and amino-acid variations were identified by alignment-based comparison with reference strains. Positional entropy (Shannon entropy) was calculated from aligned nucleotide sequences to characterize genome-wide variability patterns, using CLC Genomics Workbench v20.0.3 (Qiagen, Denmark).

Sequence alignment and phylogenetic analysis

For phylogenetic classification and comparative analyses, study genomes were analyzed together with representative global reference sequences. Multiple sequence alignments were performed using MAFFT v7.511. For genotyping/species assignment, VP1/VP2 nucleotide sequences from this study were analyzed together with representative reference sequences of HBoV1–4 retrieved from GenBank. Maximum-likelihood (ML) phylogenetic trees were inferred using IQ-TREE v2.1.2, with the best-fit substitution model selected by ModelFinder, and branch support evaluated using 1,000 ultrafast bootstrap replicates. All study sequences were assigned to HBoV-1 based on their clustering with HBoV-1 references; subclade classification within HBoV-1 was determined by robust clustering with the prototype strains ST1 (DQ000495) and ST2 (DQ000496) as anchors for subclades Ⅰa and Ⅰb, respectively.Reference sequences were selected to represent major HBoV-1 subclades (Ⅰa–Ⅰd) and recent domestic/international strains. All GenBank reference strains used for the VP1/VP2 phylogenetic analyses are listed in Supplementary Table S2 (accession, subclade, year [GenBank submission], and country/region).

To assess whether genome-wide phylogeny differed from capsid-gene phylogeny, an additional ML tree was also constructed based on the complete genome alignment of the 45 study genomes together with representative global references, using the same IQ-TREE/ModelFinder pipeline and ultrafast bootstrap settings (Supplementary Fig. S1).

Hydrophobicity analysis and functional impact prediction

Deduplicated VP1/VP2 amino-acid sequences were analyzed for hydrophobicity using ProtScale (ExPASy) with the Kyte–Doolittle method and a window size of 9.

The potential functional impact of amino-acid substitutions was predicted using PROVEAN, with a cutoff score of − 2.5 to classify variants as neutral or deleterious.

Statistical analysis

Statistical analyses were performed using SPSS v25.0 (IBM, USA). Categorical variables were compared using the chi-square test, and P values < 0.05 were considered statistically significant.

Results

Detection rate and epidemiological characteristics of HBoV

A total of 1,372 oropharyngeal swab specimens were collected from hospitalized children with pneumonia in Suzhou, China, between 2022 and 2025 and screened for HBoV by multiplex real-time PCR. Overall, 83 specimens were HBoV-positive, yielding a positivity rate of 6.05% (83/1,372). The detection rate did not differ significantly by sex (P > 0.05), whereas a significant difference was observed across age groups (P < 0.001). (Table 1).Temporal analysis showed increased HBoV detection in autumn (September to November) (Fig. 1).

Table 1.

Epidemiological characteristics of HBoV detection among hospitalized pediatric pneumonia cases in Suzhou, China (2022–2025)

	HBoV positive cases	Total cases	HBoV positive rate(%)
Age group
≤ 1	5	316	1.58
1–<2	27	285	9.47
2–<3	22	148	14.86
3–<4	20	198	10.10
≥ 4	9	425	2.12
Sex
Male	50	747	6.69
Female	33	625	5.28
Total	83	1372	6.05

Fig. 1.

Monthly distribution of sampling volume and HBoV detection in hospitalized children with pneumonia in Suzhou, 2022–2025

Co-detection patterns of respiratory pathogens in HBoV-positive cases

Co-detection of other respiratory pathogens was common among HBoV-positive specimens. Among the 83 HBoV-positive cases, 59 (71.08%) had at least one additional pathogen detected, whereas 24 (28.92%) were positive for HBoV alone. The co-detection complexity varied: 27 cases showed dual infections, 22 cases showed triple infections, 9 cases showed quadruple infections, and 1 case presented with five pathogens detected simultaneously (Fig. 2). The most frequently co-detected pathogens were HRV (27/83, 32.53%), Hi (13/83, 15.66%), RSV(13/83, 15.66%), and Spn (13/83, 15.66%). Other co-detected pathogens included EV(9/83, 10.84%), PIV (7/83, 8.43%), MP (5/83, 6.02%), ADV(4/83, 4.82%), Flu (4/83, 4.82%), HCoV (4/83, 4.82%), and hMPV (3/83, 3.61%) (Fig. 2).

Fig. 2.

Co-detection patterns of respiratory pathogens among HBoV-positive cases in Suzhou, 2022–2025. Heatmap summarizing pathogen co-detection profiles among HBoV-positive specimens (n = 83). Each column represents one HBoV-positive case, and each row represents one respiratory pathogen target included in multiplex real-time PCR testing. Red indicates a positive detection and white indicates a negative result. The bar plot on the right shows the positivity rate for each pathogen among HBoV-positive cases (%, n/83)

Whole-genome sequencing and dataset overview

To characterize viral genomic diversity at the whole-genome level, HBoV-positive specimens with Ct values < 30 were selected for amplicon-based whole-genome sequencing using a 17-primer-pair multiplex scheme covering > 98% of the HBoV genome. Using this approach, complete genome sequences were obtained from 45 of the 83 HBoV-positive cases, including 6 genomes from 2022, 5 from 2023, 15 from 2024, and 19 from 2025. These genomes were designated as SZ01–06-2022, SZ01–05-2023, SZ01–15-2024, and SZ01–19-2025. The complete genome sequences have been deposited in GenBank (accession numbers PX984692–PX984736). Sequencing accuracy was supported by built-in quality-control checks in CLC Genomics Workbench v20.0.3 during read processing and consensus generation, including inspection of genome coverage, read-depth distribution, and ambiguous bases (Ns). For downstream comparative analyses, the 45 genomes were analyzed together with 39 representative global reference sequences (including ST1 and ST2).

Phylogenetic classification based on VP1/VP2

Maximum-likelihood phylogenetic analysis based on VP1/VP2 nucleotide sequences showed that all sequenced strains clustered within HBoV-1. The 45 genomes were further classified into two subclades, Ⅰa (n = 6) and Ⅰb (n = 39), with Ⅰb predominating (86.67%) (Fig. 3).The GenBank reference strains included in the VP1/VP2 phylogeny are listed in Supplementary Table S2.

Fig. 3.

VP1/VP2 gene-based phylogenetic analysis of HBoV strains from Suzhou, 2022–2025

Within-subclade sequence identity was high: the VP1/VP2-based nucleotide identity ranged from 99.64% to 100.00% for subclade Ⅰa and from 99.39% to 100.00% for subclade Ⅰb, indicating limited intra-subclade divergence during the study period.

We further reconstructed an ML phylogeny using the complete genome sequences of the 45 strains (Supplementary Fig. S1). The whole-genome tree showed a topology consistent with the VP1/VP2-based phylogeny, assigning all Suzhou strains to HBoV-1 and separating them into subclades Ⅰa (n = 6) and Ⅰb (n = 39), with no meaningful conflicts in genotype/subclade assignment.

Maximum-likelihood phylogenetic tree inferred from VP1/VP2 sequences of strains generated in this study and representative GenBank references. Reference identifiers are provided in Supplementary Table S2. Symbols indicate sequence type: ★, reference strains; ●, strains generated in this study.

Genomic variation landscape and positional entropy

To delineate genomic variation patterns, each subclade was compared with its corresponding prototype reference strain (ST1 for Ⅰa and ST2 for Ⅰb). In subclade Ⅰa (n = 6), a total of 39 nucleotide substitutions were identified relative to ST1, including 11 substitutions in NS1, 4 in NP1, and 19 in the VP1/VP2 region; overall nucleotide identity ranged from 99.64% to 100.00%.

In subclade Ⅰb (n = 39), 83 nucleotide substitutions were identified relative to ST2, comprising 18 substitutions in NS1, 11 in NP1, and 39 in VP1/VP2; nucleotide identity ranged from 99.39% to 100.00%.

Positional entropy analysis based on aligned nucleotide sequences revealed higher entropy values within the VP1/VP2 coding region compared with NS1 and NP1, suggesting that sequence variability was preferentially enriched in the capsid gene region (Fig. 4).

Fig. 4.

Positional entropy of HBoV nucleotide sequences in Suzhou, 2022–2025

Positional entropy calculated from aligned HBoV nucleotide sequences obtained in this study (n = 45). Bars indicate entropy values at each nucleotide position across the genome.

Amino-acid substitutions in VP1/VP2 and substitution frequency patterns

Amino-acid substitutions detected in Suzhou HBoV strains during 2022–2025 were predominantly located in the VP1/VP2 region. In subclade Ⅰa (n = 6), the number of variable sites was limited; however, two high-frequency substitutions were observed within VP1/VP2: A149/20T occurred in all strains (100%, 6/6) and R17K occurred in 66.7% (4/6), while the remaining sites appeared at low frequencies.

In subclade Ⅰb (n = 39), VP1/VP2 variation was characterized by a fixed substitution, N474/345S, present in all strains (100%, 39/39). In addition, a moderate-frequency substitution in NP1 (S79N) was observed in 30.8% (12/39). Other substitutions were sporadic and occurred at low frequencies (Fig. 5).

Fig. 5.

Frequency of amino-acid substitutions in HBoV strains from Suzhou, 2022–2025

Frequencies of amino-acid substitutions identified in HBoV strains obtained in this study. The x-axis lists amino-acid substitution sites, and the y-axis indicates the percentage of strains carrying each substitution within subclade Ⅰa and subclade Ⅰb.

Predicted functional impact (PROVEAN) and hydrophobicity profiling of VP1/VP2

PROVEAN prediction

To evaluate the potential functional impact of VP1/VP2 substitutions, PROVEAN was applied using the default cutoff (− 2.5). In subclade Ⅰa, the major substitutions were predicted as neutral: R17K (score = 0.522), A149/20T (score = − 1.145), and T590/461S (score = − 0.292), indicating that these changes are unlikely to exert deleterious effects under the PROVEAN model.

Similarly, in subclade Ⅰb, all evaluated substitutions were predicted as neutral, including R17K (0.488), L40S (1.339), D68N (− 1.154), D72N (− 0.569), D137/8 N (− 0.326), R251/122K (− 2.224), and N474/345S (− 0.194). Among these, R251/122K yielded the score closest to the cutoff but remained above − 2.5 and was still classified as neutral.

Hydrophobicity profiling (ProtScale)

ProtScale (Kyte–Doolittle, window size = 9) analysis showed that the overall VP1/VP2 hydrophobicity profiles of deduplicated sequences were largely consistent with their corresponding reference strains, with only localized and reproducible shifts in a few segments.

In subclade Ⅰa, hydrophobicity differences were mainly concentrated in positions 144–152, where all strains showed a consistent downward shift with a maximum amplitude of approximately − 0.278. In contrast, subclade Ⅰb exhibited more dispersed shifts: a downward shift was observed in positions 35–43 (maximum ≈ − 0.511), whereas an upward shift was consistently observed in positions 469–477 across all samples (maximum ≈ + 0.300). Here, an upward (downward) shift indicates higher (lower) hydrophobicity index values relative to the corresponding reference profile. A summary of the hydrophobicity-shift segments and amplitudes is provided in Table 2.

Table 2.

VP1/VP2 hydrophobicity shift segments in HBoV-1 subclades Ⅰa and Ⅰb relative to reference strains

Segment (aa positions)	Direction of change	Maximum Δ amplitude
Ia
12–20	Upward shift	+ 0.067
144–152	Downward shift	−0.278
585–593	Downward shift	−0.011
Ib
12–20	Upward shift	+ 0.067
35–43	Downward shift	−0.511
246–254	Upward shift	+ 0.067
469–477	Upward shift	+ 0.300

Reference profiles were derived from ST1 (DQ000495) for subclade Ⅰa and ST2 (DQ000496) for subclade Ⅰb. Upward (downward) shifts indicate higher (lower) hydrophobicity index values relative to the reference profile; Δ amplitude represents the maximum change within each segment

Discussion

In this study, we conducted molecular surveillance of human bocavirus among hospitalized children with pneumonia in Suzhou, China, from 2022 to 2025. By combining multiplex real-time PCR screening with an amplicon-based whole-genome sequencing strategy, we obtained 45 complete HBoV genomes and characterized their phylogenetic distribution, genomic variation patterns, and VP1/VP2 physicochemical features. Our findings provide a regional genomic baseline for continued HBoV surveillance and contribute additional whole-genome data for global comparative analyses.

The overall HBoV positivity rate in this cohort was 6.05% (83/1,372), which is broadly consistent with previously reported detection rates of HBoV in pediatric respiratory infections [10–12]. We observed a clear seasonal pattern, with increased detection in autumn (September–November), suggesting that HBoV circulation in Suzhou may exhibit predictable seasonal dynamics [13]. In addition, significant differences were observed across age groups, highlighting that HBoV detection is not uniformly distributed across pediatric populations and may be influenced by age-related exposure patterns or immune status [14–15].

A notable finding was the high co-detection rate among HBoV-positive cases (71.08%), with human rhinovirus being the most frequently co-detected virus (32.53%), followed by Haemophilus influenzae, respiratory syncytial virus, and Streptococcus pneumoniae (each 15.66%). Such frequent co-detection has been repeatedly reported in HBoV studies and complicates causal attribution of HBoV as the sole etiological agent of pneumonia [16]. Nevertheless, from a clinical and surveillance perspective, the observed co-detection profile underscores the value of multiplex molecular diagnostics in hospitalized pediatric pneumonia, particularly in settings where bacterial and viral pathogens co-circulate [17].

Whole-genome sequencing generated 45 complete HBoV genomes in this study. VP1/VP2 gene-based maximum-likelihood phylogenetic analysis classified all strains as HBoV-1 and further into two subclades, Ⅰa (n = 6) and Ⅰb (n = 39), with Ⅰb predominating during 2022–2025. Both subclades exhibited high VP1/VP2 sequence identity, and nucleotide substitutions were preferentially enriched in the VP1/VP2 region compared with NS1 and NP1, as supported by positional entropy analysis. These results indicate that capsid genes (VP1/VP2) represent a major hotspot of genetic diversity in circulating HBoV-1 strains in Suzhou, consistent with prior observations that VP1/VP2 exhibits relatively more frequent nucleotide polymorphisms [18].

To place the Suzhou genomes in a broader genomic context, we compared our 45 complete genomes with publicly available complete HBoV-1 genomes from China and other regions included as references in both the VP1/VP2- and whole-genome phylogenies (Supplementary Fig. S1; Supplementary Table S2). Overall, the Suzhou strains fall within the globally distributed HBoV-1 lineages and are closely related to recent genomes reported from China/Asia, consistent with regional circulation of a predominant lineage during 2022–2025. Notably, the genotype/subclade assignment inferred from VP1/VP2 is concordant with the genome-wide classification, supporting the robustness of VP1/VP2-based typing for HBoV-1. Taken together, these observations suggest limited genome-wide divergence over the study period and highlight that the Suzhou strains represent part of the contemporary HBoV-1 genomic diversity observed in China and worldwide.

Although VP1/VP2 amino-acid substitutions were observed, ProtScale analysis showed that the overall hydrophobicity profiles remained largely conserved relative to reference strains(ST1, DQ000495 for subclade Ⅰa; ST2, DQ000496 for subclade Ⅰb), with only localized shifts adjacent to substitution sites. Furthermore, PROVEAN predicted the major substitutions in both subclades to be neutral, suggesting that these changes are unlikely to cause deleterious functional effects under the model. Taken together, these findings support the notion that recent HBoV-1 evolution in this setting may be characterized by limited functional divergence despite ongoing accumulation of capsid-region substitutions [19–20].

An important question is whether the VP1/VP2 variants observed here involve antigenic epitopes or receptor-binding sites. Published capsid structural and antibody-epitope mapping studies indicate that exposed capsid surface loops constitute major antigenic regions in HBoV [21–22]. In our dataset, substitutions were enriched in VP1/VP2 (e.g., positions 17,149/20 and 474/345), which could be consistent with immune-related selection acting on capsid-exposed regions. However, without structure-guided mapping of these residues onto available capsid models and functional validation, we cannot determine whether specific substitutions overlap defined epitope footprints. In addition, because the precise cellular receptors for HBoV-1 remain incompletely defined, receptor-binding sites cannot be assigned based solely on sequence variation.

This study has several limitations. First, the analysis was based on specimens collected from a single pediatric hospital, which may limit generalizability to other regions. Second, whole-genome sequencing was restricted to samples with Ct values < 30, potentially underrepresenting low-viral-load infections. Third, detailed clinical information (e.g., symptom severity, imaging findings, treatment, and outcomes) was not incorporated, which limited our ability to assess associations between viral genotypes and disease severity. Finally, while in silico analyses (hydrophobicity profiling and PROVEAN prediction) provide useful preliminary insights, additional work using structural modeling, epitope mapping, and functional assays will be required to clarify whether VP1/VP2 variations affect antigenicity or host interactions.

Conclusion

In this study, HBoV was detected in 6.05% of hospitalized children with pneumonia in Suzhou, China, during 2022–2025, and co-detection with other respiratory pathogens was frequent. Using an amplicon-based whole-genome sequencing approach, we generated 45 complete HBoV genomes and showed that circulating strains were exclusively HBoV-1, with a predominance of subclade Ⅰb. Genomic variation was mainly concentrated in the VP1/VP2 capsid region, while hydrophobicity profiles were largely conserved and the major VP1/VP2 substitutions were predicted to be neutral by PROVEAN. These data provide a local genomic reference to support continued molecular surveillance and longitudinal tracking of HBoV in pediatric respiratory infections.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Xuan Yuan and Yayun Tan contributed equally to this work. Xuan Yuan and Yayun Tan: Conceptualization, Methodology, Investigation, Data curation, Writing—original draft. Zhihui Xu and Xuerong Ya: Investigation , Specimen collection. Zefeng Dong and Qiang Shen are co-corresponding authors and contributed to Supervision, Project administration, and Writing—review & editing. All authors read and approved the final manuscript.

Funding

This work was supported by the Project of Science and Education for Promoting Health Development (QNXM2024058) and the Suzhou Gusu Health Talent Program (GSWS2024057).

Data availability

The complete genome sequences generated in this study have been deposited in GenBank under accession numbers PX984692–PX984736. The sequences are scheduled for public release on 1 March 2027.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Suzhou Center for Disease Prevention and Control (Suzhou CDC) (approval no.: SZCDC/IRB/PJ/2025-05). The requirement for written informed consent was waived because residual clinical specimens were used and all data were anonymized.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.