Search Results (1,011)

Feline kobuvirus (FKoV) has been linked to diarrheal disease in cats and has recently emerged in several Asian countries. This study aimed to investigate the epidemiological and genetic characterization of FKoV in domestic cats in northern Vietnam from 2022 to 2025. In total, 244 fecal samples were obtained from both healthy and sick cats in Hanoi, Hungyen, Bacninh, and Ninhbinh in the north of Vietnam. Of 244 fecal samples examined, eight (3.28%) samples were found to be positive for the FKoV genome by using the conventional PCR method. Viral genomes were identified in both healthy and sick cats. The highest FKoV-positive rate was 12.50% in 6–12-month-old cats, significantly higher than those detected in cats aged 3–6 months and >12 months. Eight positive samples were successfully sequenced and characterized for the partial 3D gene and full-length VP1 genes. Results showed that the level of nucleotide identity of the partial 3D gene and full-length VP1 genes among the FKoV strains in this study ranged from 95.86% to 100% and 94.09% to 99.73%, respectively. Phylogenetic analysis revealed that the eight Vietnamese viral strains obtained in this study formed a novel cluster within FKoV and are genetically related to Chinese FKoV strains. Two positive selections were found in the VP1 protein of the Vietnamese FKoV strains. Full article

Predicting CRISPR-Cas9 guide RNA efficiency and off-target activity is a precondition for precise genome editing. Computational models have progressively incorporated chromatin accessibility and epigenetic descriptors into their feature sets, yet synthesising findings from independently published studies—especially when those studies contradict one another—remains an unresolved methodological gap. Large Language Models (LLMs) have been proposed as a route to automate cross-study synthesis, but their utility depends on a constraint that receives less attention than model architecture: how much of the source text actually reaches the model at inference time. Cloud-based models process 48,000-token corpora without hardware limitations, but at the cost of data leaving the local environment and with limited reproducibility across API versions. Local RAG systems avoid the cloud dependency while fragmenting the input, discarding the global context needed to link biological arguments that are distributed across separate papers. We benchmark these strategies using a corpus of four CRISPR-Cas9 efficiency prediction studies and apply the Reduced Interaction Sampling (RIS) engine—a local sparse attention method—to retain the full sequence within the memory envelope of a laboratory server. Preserving that context uncovers three latent inconsistencies. The static epigenetic markers used in DeepCRISPR (CTCF, DNase I) show near-zero Spearman correlations with off-target cleavage ($\rho \le 0.07$), while nucleosome positioning scores from the Block Decomposition Method reach $\rho =0.388$–$0.423$. The sequence-only Apindel model was published in June 2022 without incorporating nucleosome descriptors reported in the concurrent literature. The benchmark review by Konstantakos et al. attributed 10–20% of rank correlation to epigenetics—a figure that reflects the weak feature subset evaluated, not a ceiling on chromatin influence. These discrepancies are invisible when papers are read individually or retrieved as chunks; they become traceable only when the full corpus is processed as a single context window. An independent empirical analysis of 2000 CRISPR-Cas9 off-target cleavage events provides evidence consistent with this pattern: static epigenetic markers yield $\left|\rho \right|\le 0.11$, whereas computed NuPoP Affinity descriptors reach $r=-0.622$ ($p<{10}^{-210}$). On a 30-question cross-study synthesis benchmark (5 independent seeds), baseline accuracy is 53.33%, RAG 60.00%, and RIS (30 seeds, 3% density) 70.00% ($p<0.0001$, t-test vs. RAG, $\sigma =0.00\%$ for all configurations). Full article

(1) Background: The Hepatitis B virus (HBV) is characterized by extensive genetic diversity, including low-frequency variants that contribute to disease progression. We aimed to characterize low-frequency variants and evaluate their potential clinical impact. (2) Methods: We utilized 104 HBV near-full-length sequences generated using next-generation sequencing (NGS) from people living with HIV (PLHIV). We used an in-house bioinformatics suite (HBVgenomeR v5.9.7) to filter for low-frequency variants (5–50%), which were compared to escape and drug resistance mutations (DRMs) and hepatocellular carcinoma (HCC)-associated mutations reported at the consensus level. Unclassified variants were characterized by HBV open reading frames (ORFs) to determine mutation frequency per genomic region. (3) Results: A total of six escape mutations were detected in 8/104 (7.7%) sequences, with _surfaceN131T being the most prevalent (5/8). We also observed six DRMs in 30/104 (28.8%), with _rtV173L being the most prevalent (21/30). Truncation mutations were also observed with _rtA181T/_sW172* and _rtM204I/_sW196L being the most prevalent. A total of 8/104 (7.7%) sequences had four variants associated with HCC. The _xP46S was the highest observed HCC-associated mutation at 5/8. We report 1152 unique uncharacterized variants across all ORFs, and these were found in 94/104 (90.4%) sequences. The RNaseH domain had the highest burden (330/1152, 28.6%). (4) Conclusions: Deep sequencing results identified clinically significant mutations, including those below the 20% detection limit of traditional sequencing, that would go unreported. This highlights the possible underreporting of mutational burden in people living with HBV/HIV, indicating the importance of deep sequencing to aid in HBV/HIV understanding and management. Full article

The SWEET (Sugars Will Eventually be Exported Transporter) family, a class of sugar transporters identified in recent years, plays important roles in plant growth and development. Increasing evidence suggests that SWEET genes are also involved in plant responses to abiotic stresses, including salt stress. However, the genome-wide composition and salt-responsive functions of SWEET genes in Elymus nutans Griseb. have not been systematically characterized to date. This study presents the first genome-wide identification of the SWEET gene family in E. nutans and functional characterization of EnSWEET15 in salt stress response, providing candidate genes and a theoretical basis for improving salt tolerance in this species. A total of 12 EnSWEET genes were identified from the E. nutans genome and were unevenly distributed across eight chromosomes. Phylogenetic analysis classified them into four subfamilies, and promoter analysis revealed abundant cis-elements related to hormone signaling and stress responses, suggesting roles in abiotic stress adaptation. Expression analysis showed that EnSWEET15 was significantly upregulated in both roots and leaves under salt stress. The full-length EnSWEET15 coding sequence (930 bp ORF) was cloned and functionally characterized. Heterologous overexpression in Arabidopsis thaliana indicated that EnSWEET15 enhances salt tolerance. Transgenic lines showed reduced reactive oxygen species (ROS) accumulation and increased activities of antioxidant enzymes, including superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT), compared with wild-type plants under salt stress. Overall, this study identifies EnSWEET15 as a positive regulator of salt tolerance via modulating ROS homeostasis, providing novel insights into SWEET-mediated salt stress response in perennial forages and valuable genetic resources for salt-tolerant breeding of E. nutans. Full article

Wild almond relatives are valuable reservoirs of allelic variation for crop improvement and conservation, yet Kazakhstan’s wild-almond genetic resources remain poorly characterised. We conducted an exploratory SSR assessment of 80 putative individuals from four taxon-locality groups (20 per group), each representing one sampled population: Prunus ledebouriana, P. tenella, P. petunnikowii, and P. spinosissima. Of 22 nuclear simple sequence repeat loci screened for cross-taxon transferability, 15 generated reproducible profiles and were retained; their even genome-wide distribution was not verified. Across the full dataset, the mean number of alleles was 3.55, the effective number of alleles was 2.62, expected heterozygosity (He) was 0.544, and 95.0% of loci were polymorphic. Missing genotypes ranged from 0.0% to 34.7% among groups, and six loci had at least 20% missing data. AMOVA attributed 76.5% of variation to within-group differences and 23.5% to among-group differences (PhiPT = 0.235, p = 0.001). PCoA, unbiased Nei distances, UPGMA, and descriptive Bayesian clustering separated the four sampled groups. A nine-locus sensitivity analysis that excluded the six high-missing loci retained P. spinosissima as the group with the highest mean He (0.699), whereas P. petunnikowii increased from 0.471 to 0.609. Thus, the low full-panel estimate for P. petunnikowii was not robust to missing data. Because taxon identity was fully confounded with locality and the marker panel was limited, the results are interpreted as a regional marker-transferability and methodological baseline rather than as species-wide or genome-wide inference. Full article

Background/Objectives: DNA-based identification of degraded human remains represents a major challenge in forensic science, particularly in cases involving burned, fragmented, or commingled bodies. Advances in forensic genetics have expanded the analytical capabilities for such samples; however, the effectiveness of different approaches and their integration within Disaster Victim Identification (DVI) workflows remain heterogeneous. This systematic review aims to critically evaluate current evidence on DNA-based identification of degraded remains, focusing on methodological strategies, emerging genomic technologies, and DVI applications, while integrating laboratory evidence and operational forensic practice into a structured analytical framework. Methods: A systematic literature search was conducted in Scopus and Web of Science from database inception to 5 June 2026, following PRISMA 2020 guidelines. Eligible studies included original research addressing DNA analysis of degraded, thermally altered, or highly compromised human remains in forensic or DVI contexts. After a multistep screening process involving title/abstract and full-text evaluation, 37 studies were included. Data were extracted and organized into three thematic categories: (i) core DNA analysis, (ii) advanced molecular technologies, and (iii) DVI case applications. Results: The findings demonstrate that DNA recovery from degraded remains is influenced by thermal exposure, tissue type, and sampling strategy. Teeth and dense cortical bone consistently provide higher DNA yield. While autosomal STR profiling remains the primary analytical approach, its limitations in highly degraded samples are mitigated through the complementary use of mitochondrial DNA (mtDNA), Y-chromosome STRs (Y-STRs), and SNP markers, together with advanced sequencing technologies such as massively parallel sequencing (MPS). Emerging technologies, including rapid DNA systems and predictive models based on macroscopic indicators, significantly enhance efficiency and success rates. DVI studies report identification rates exceeding 90–95% when multidisciplinary and structured workflows are applied. The evidence further supports a flexible triage-based analytical strategy, in which marker selection is guided by tissue preservation and degradation level. Conclusions: DNA-based identification of degraded human remains has evolved into an adaptive, multi-level forensic process. Successful outcomes rely on the integration of optimized sampling, hierarchical genetic analysis, and coordinated DVI strategies. The findings support a triage-based framework that links tissue selection, degradation assessment, and analytical methodology to maximize identification success. Future developments should focus on predictive models, advanced genomic tools, and standardized workflows to further improve identification in challenging forensic scenarios. Full article

The quail is a valuable model organism for avian developmental biology and genetic improvement; however, the full-length transcriptomic landscape and regulatory networks underlying critical embryonic developmental stages remain poorly characterized. In this study, we employed Oxford Nanopore Technologies full-length transcriptome sequencing to systematically analyze leg muscle tissues from quail embryos at embryonic day 8 (E8, pre-differentiation stage) and embryonic day 14 (E14, later differentiation stage), with four biological replicates per stage (eight libraries in total). Nanopore sequencing identified a total of 54,938 transcripts and 21,574 genes, including 49,678 known transcripts, 20,980 known genes, 5260 novel transcripts, and 594 novel genes. Analysis of transcriptome structure revealed that alternative first exon and exon skipping represented the predominant stage-specific alternative splicing events, and we identified 2692 lncRNAs. Differential expression analysis identified substantial transcriptome dynamics from E8 to E14, with 5934 differentially expressed transcripts and 3745 differentially expressed genes identified. Functional enrichment analyses, including KEGG, GSEA and PPI network analysis, revealed that the E8 transcriptome was predominantly enriched for pathways associated with cell proliferation, whereas the E14 transcriptome shifted toward the coordinated refinement of myofiber structural assembly and energy metabolism. Here, we present a high-resolution full-length transcriptional atlas of quail skeletal muscle development and identify candidate regulatory genes, offering a valuable theoretical framework and data resource for refining quail genome annotation. Full article

Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of aggressive non-Hodgkin lymphoma and is characterized by pronounced molecular heterogeneity that is not always fully captured by standard histopathological assessment. Circulating tumor DNA (ctDNA) is increasingly regarded as a promising liquid-biopsy biomarker that enables non-invasive molecular tumor profiling, assessment of tumor burden, dynamic monitoring of treatment response, and detection of measurable/minimal residual disease (MRD). Modern analytical platforms, ranging from PCR-based assays to next-generation sequencing approaches, including CAPP-Seq and PhasED-Seq, have substantially expanded the possibilities of molecular monitoring in DLBCL. This review summarizes current data on the biological characteristics of ctDNA, contemporary methods for its analysis, concordance between ctDNA and tumor-tissue mutational profiles, and the clinical significance of baseline ctDNA levels, early molecular response, post-treatment MRD status, and molecular surveillance during remission. Special attention is given to ctDNA monitoring in patients receiving novel immunotherapies, including CAR-T cell therapy, bispecific antibodies, and antibody–drug conjugates. Emerging multi-omic approaches integrating genomic, epigenomic, and fragmentomic data are discussed as promising future directions. Key limitations of clinical implementation include insufficient standardization of preanalytical and analytical workflows, the confounding effect of clonal hematopoiesis of indeterminate potential, variability across technological platforms, and the lack of completed prospective randomized interventional studies demonstrating improved outcomes when therapy is modified according to ctDNA status. Overall, ctDNA is currently a highly informative prognostic biomarker in DLBCL; however, its full implementation as a predictive tool for treatment selection requires further harmonization, prospective validation, and confirmation in interventional clinical trials. Full article

Thiamine-responsive megaloblastic anemia (TRMA) is a rare autosomal recessive disorder caused by biallelic loss of function variants in the SLC19A2 gene. It typically presents with a triad of megaloblastic anemia, diabetes mellitus, and sensorineural deafness. In this work, we analyzed ten children with suspected TRMA: nine exhibited the full triad and one, a younger sibling of a patient with full triad, did not develop hearing loss by the age of 18 months. All patients became transfusion-independent on high-dose thiamine therapy. Whole-genome sequencing identified homozygosity for the canonical splice variant SLC19A2 c.1223+1G>A in eight patients. One patient was homozygous for a known SLC19A2 c.196G>T variant, and the other was compound heterozygous for two novel variants, c.120C>G and c.584T>C. All patients with the SLC19A2 c.1223+1G>A variant were ethnic Ingush. In the reference Ingush cohort, 9/328 unrelated adults were c.1223+1G>A carriers (heterozygous carrier frequency 2.7%; carrier frequency ≈ 1/36), and shared a 2.3 Mb ATP1B1–FMO2 haplotype on chromosome 1, demonstrating a strong founder effect. These findings identify Ingushetia as a new TRMA-endemic region and support targeted SLC19A2 screening and early thiamine therapy in patients with macrocytic anemia and diabetes of unclear origin in this population. Full article

Background and Objectives: Fabry disease (FD) is an X-linked lysosomal storage disorder caused by pathogenic GLA gene variants, leading to progressive multi-organ damage including end-stage renal disease. Although dialysis patients represent a high-risk population for undiagnosed FD, data from Türkiye using genetic analysis as the primary screening method remain limited. This study aimed to determine FD prevalence among hemodialysis patients across multiple centers in Türkiye and to perform cascade family screening of confirmed cases. Materials and Methods: This multicenter cross-sectional study screened 1359 adult hemodialysis patients across 8 centers in Ankara, Türkiye, using complete GLA gene sequencing. Variants were classified per American College of Medical Genetics and Genomics criteria. Patients with pathogenic variants underwent confirmatory biochemical testing (α-galactosidase A activity and plasma lyso-Gb3). Cascade screening was performed for confirmed index cases. Results: Among 1359 patients (mean age 62.3 ± 14.3 years; 38.5% female), GLA variants were identified in 12 (0.88%): 8 benign/likely benign (including 7 p.D313Y pseudodeficiency alleles), 2 unclassified variants, 1 variant of uncertain significance, and 1 confirmed classic FD (prevalence: 0.07%; 95% CI: 0.002–0.41%). Cascade screening of the index patient identified 6 carriers among 10 relatives tested (60% yield). Three of 7 carriers (43%) were initiated on enzyme replacement therapy. Conclusions: Among 1359 hemodialysis patients, GLA gene sequencing identified 12 variants (0.88%), yet only one was confirmed as a disease-causing mutation responsible for end-stage renal disease (prevalence: 0.07%). The remaining variants comprised polymorphisms, likely benign pseudodeficiency alleles and variants of uncertain significance; most of which would not have been detected by enzyme-based screening alone, as enzyme activity was normal in these carriers. Cascade screening of the single confirmed index case yielded 6 carriers among 10 relatives tested (60%), including one hemizygous male with classic FD on hemodialysis, and three carriers were initiated on enzyme replacement therapy. These findings demonstrate that GLA gene analysis is a valuable instrument for screening in dialysis populations, as it captures the full variant spectrum while enabling rigorous distinction between the overall GLA variant carrier rate and the true disease prevalence defined by variants causing end-stage renal disease. Future screening studies should report prevalence based exclusively on confirmed disease-causing variants rather than total variant counts, which have inflated prevalence estimates in prior literature. Full article

Wild waterfowl constitute the primary natural reservoir of influenza A viruses, and wetlands at the convergence of major migratory flyways serve as critical hubs for viral genetic exchange. Baikal Siberia, situated at the intersection of the East African–West Asian, Central Asian, and East Asian–Australasian flyways, represents a unique yet understudied region in this context. Here we report the results of long-term virological surveillance of wild birds in the Lake Baikal basin conducted between 2018 and 2024. A total of 1036 cloacal swab samples from 28 bird species were screened, yielding 42 influenza A virus isolates belonging to 12 HA/NA subtype combinations: H1N1, H3N1, H3N2, H3N5, H3N6, H3N8, H4N6, H6N1, H6N2, H6N3, H6N8, and H12N5. Among the detected subtypes, H6 viruses—identified with four distinct neuraminidase combinations (N1, N2, N3, N8)—are of particular public health relevance owing to their documented capacity for dual-receptor binding and potential for zoonotic transmission to mammals, including humans. Full-genome sequencing followed by cluster analysis of internal gene segments identified 16 distinct segment constellations, indicating extensive reassortment. BLAST searches against the GISAID database revealed closest genetic relatives in Mongolia, South Korea, Japan, China, and Western Siberia, with more distant links to Bangladesh, Europe, and a possible intercontinental connection via the Pacific flyway. Maximum-likelihood phylogenetic analysis of the HA and NA segments confirmed that all isolates belong to the Eurasian genetic lineage, yet they are distributed across multiple clades rather than forming a single monophyletic group, reflecting the role of Buryatia as a mixing zone for genetically diverse viral populations. These findings substantially expand the understanding of influenza A virus ecology in the Lake Baikal basin and underscore the importance of continued surveillance at this key migratory crossroads in Northern Asia. Full article

Low-density SNP panels are widely used to reduce genotyping costs in genomic selection (GS) in aquaculture, but sparse marker density can constrain prediction accuracy, particularly in species with rapid linkage disequilibrium decay. In this study, we evaluated a locally enriched multi-SNP (mSNP) strategy derived from a conventional 1K SNP panel in a family-based breeding population of Pacific white shrimp (Penaeus vannamei). Targeted sequencing was used to recover multiple nearby variants surrounding each marker locus, and conventional SNP and mSNP panels were compared for genomic features, imputation performance, and genomic prediction accuracy for body weight. The mSNP panel exhibited higher polymorphism and broader gene-region coverage than the corresponding SNP panel. The original mSNP panel improved prediction accuracy by 11.6% over the original 1K SNP panel. This advantage was maintained or slightly improved when mSNPs were restricted to on-target variants within ±300 bp of the target markers. A locus-matched comparison further showed that enriched mSNPs consistently outperformed conventional SNPs derived from the same target loci, with a 12.0% improvement at the full 961-locus panel. Although imputation improved the 1K SNP panel, the original mSNP panel already outperformed the imputed SNP panel and reached 0.509 after imputation. These findings demonstrate that locally enriched mSNP markers provide an effective and practical strategy for improving low-density GS in aquaculture species with rapid LD decay. Full article

Genome and exome sequencing have become central to diagnosing rare hereditary diseases, but each test returns thousands of variants that a clinical scientist must review by hand to find the one responsible for the patient’s condition. This manual interpretation is the main bottleneck in clinical genomics. To reduce it, we developed DiagAI, a machine-learning system that ranks the variants found in a patient and returns a short list of the most likely causal candidates. DiagAI combines three sources of evidence: a pathogenicity score from the Universal Pathogenicity Predictor (UP2), a model we trained to estimate how damaging a variant is on the five-tier scale of the American College of Medical Genetics and Genomics (ACMG); a phenotype-matching score from PhenoGenius, which weighs how well a gene’s known clinical features match the patient’s symptoms (encoded as Human Phenotype Ontology, or HPO, terms); and expert rules covering inheritance pattern and sequencing quality. We evaluated DiagAI on 966 exomes from adults investigated for kidney disease of unknown cause, of which 196 had a confirmed genetic diagnosis. We first tested UP2 on its own by ranking 62 confirmed disease-causing missense variants that were absent from its training data: UP2 placed the causal variant within the top 100 candidates in 87% of cases, compared with 61% for the widely used tool REVEL. Across the 196 diagnosed exomes, the full DiagAI shortlist contained the causal variant in 94.9% of cases when the patient’s symptoms were provided and in 90.8% when they were not, with a typical shortlist of about 10 variants. When symptoms were provided, the single top-ranked variant was the correct diagnosis in 74% of cases, versus 42% without symptoms, exceeding the performance of the established tools Exomiser and AI-MARRVEL on the same cohort. DiagAI produces compact, accurate shortlists that can reduce the manual interpretation workload as diagnostic sequencing volumes continue to grow. Full article

Average nucleotide identity (ANI) values of 95–96% and digital DNA–DNA hybridization (dDDH) values of 70% are currently regarded as the gold standards for bacterial species delineation. However, the accuracy and applicability of ANI thresholds may vary among genera and depend on genome quality. In this study, three genome quality standards were applied to re-evaluate ANI classification thresholds for the genus Nocardiopsis: >90% completeness and <5% contamination (13 strains, 78 genome pairs), >90% completeness and <10% contamination (34 strains, 561 genome pairs), and >50% completeness and <10% contamination (40 strains, 780 genome pairs). Based on these analyses, we preliminarily estimated the species delineation thresholds for the genus Nocardiopsis as 96.68% for ANIm and 96.15% for ANIb, using the currently available genome datasets. Strain HUAS JQ3^T was isolated from leaves of Camellia oleifera Abel collected in Taoyuan County, northwestern Hunan Province, China. Phylogenetic, genomic, morphological, cultural, and physiological analyses were conducted to determine its taxonomic status. Full-length 16S rRNA gene sequence analysis showed that strain HUAS JQ3^T belongs to the genus Nocardiopsis and shares the highest sequence similarities with N. dassonvillei subsp. crassaminis D1^T (100.00%), N. alborubida NBRC 13392^T (99.86%), N. synnemataformans DSM 44143^T (99.73%), and N. akebiae HDS 12^T (99.73%). Phylogenetic analysis based on 16S rRNA gene sequences clustered strain HUAS JQ3^T with N. alborubida, N. synnemataformans, and N. dassonvillei subsp. crassaminis. In contrast, whole-genome phylogenetic analysis indicated that strain HUAS JQ3^T is most closely related to N. akebiae HDS 12^T. Nevertheless, the ANIm/ANIb and dDDH values between strain HUAS JQ3^T and N. akebiae HDS 12^T were below the proposed species delineation thresholds of 70% dDDH and 96.68%/96.15% ANIm/b for the genus Nocardiopsis. In addition, strain HUAS JQ3^T exhibited distinct morphological, cell culture, physiological, and biochemical characteristics compared with N. akebiae HDS 12^T. The combined genotypic and phenotypic evidence gathered in this study demonstrates that strain HUAS JQ3^T (= MCCC 1K08696^T = JCM 36305^T) represents a novel species of the genus Nocardiopsis, for which the name Nocardiopsis camelliae sp. nov. is proposed. Full article

Although commercial vaccines against porcine circovirus type 2 (PCV2) have been widely implemented globally, PCV2 remains endemic in swine populations, accompanied by ongoing genotype replacement. Meanwhile, the emergence of novel porcine circoviruses (PCVs), including PCV3 and PCV4, has further complicated the prevention and control of porcine circovirus-associated diseases (PCVAD). This study systematically characterized the epidemiological patterns and genetic diversity of PCVs circulating in Eastern China. A total of 739 clinical samples collected between 2010 and 2016 were screened for PCV2. Additionally, 653 samples obtained during 2023–2024 were analyzed using a triplex real-time quantitative PCR (qPCR) assay for the simultaneous detection of PCV2, PCV3, and PCV4. Full-genome amplification and sequencing were subsequently performed on all PCR-positive samples. Epidemiological analysis revealed an overall PCV2 positivity rate of 37.62% during 2010–2016. In the 2023–2024 cohort, the positivity rates for PCV2 and PCV3 were 35.99% and 16.39%, respectively, with a co-infection rate of 10.26%. Notably, no PCV4-positive samples were detected. Phylogenetic analysis demonstrated that PCV2d is the predominant genotype in Eastern China. Furthermore, PCV2g strains were identified in clinical samples for the first time in mainland China, while PCV3b were determined to be the dominant circulating subtype of PCV3. Multiple critical amino acid substitutions were identified within the neutralizing epitopes of the PCV2 Cap protein, and a recombination event involving a PCV2d vaccine strain and a PCV2c reference strain was detected. In contrast, the PCV3 genome exhibited a high degree of genetic conservation. Collectively, these findings expand the molecular epidemiological landscape of PCVs in Eastern China and elucidate the evolutionary dynamics of circulating PCV strains, providing important insights for the development of next-generation vaccines and region-specific PCVAD control strategies. Full article