Combined optical genome mapping and cnv-seq identify complex Y-chromosome rearrangements and ectopy in 46,XX testicular disorder of sex development Authors: Hai Wang, Hao Wang, Zitong Xu, Penglong Chen, Xianjue Zheng, Haojie Pan, Hongping Zhang, Jiayong Zheng Published: 2026 ABSTRACT AbstractObjectiveTo define the genomic architecture of Y‑chromosome ectopy in a child with 46,XX testicular disorders of sex development (DSD) and illustrate the complementary roles of low‑coverage CNV sequencing (CNV‑seq) and optical genome mapping (OGM).MethodsA one‑and‑a‑half‑year‑old child with male phenotype, unfused scrotum, and hypospadias underwent clinical assessment and Doppler ultrasound, karyotyping, whole‑exome sequencing (WES), CNV‑seq, and OGM. CNV‑seq data were aligned to GRCh37; OGM maps were aligned to GRCh38 with T2T‑CHM13 consulted for Y display. Coordinates were harmonized for interpretation. FISH and long‑range PCR targeting the full SRY locus were not performed.ResultUltrasound showed bilateral scrotal testes, with no uterus or ovarian tissue. Karyotype was 46,XX, inv(9)(p12q13). CNV-seq identified called losses at Xp22.33 (287,328–3,605,487) and within Yp11.31–p11.2 (2,649,472–6,065,425) and Yp11.2 (7,270,527–9,599,178). OGM resolved a derivative X containing an insertion of Yp sequence at Xp22.33, with three re-join events in Yp11.2, including a deleted interval (~ 6.2–7.3 Mb) and a local inversion (~ 7.3–9.9 Mb); the call was ogm[GRCh38] der(X)(Ypter_Yp11.2::Yp11.2::Xp22.33_Xqter)(Ypter_6248448::9915169_7362950::3618476_Xqter). Harmonized coordinates indicate that the retained Yp11.3 interval includes SRY and is inserted into der(Xp22.33), supporting SRYpositive 46,XX testicular DSD on a coordinate basis. The Xp22.33 deletion supports counseling regarding SHOX haploinsufficiency–related growth surveillance.ConclusionCNV-seq and OGM deliver nonredundant, genomewide dosage and architectural resolution that localizes the insertion site, defines breakpoint orientation, and provides counseling-relevant information beyond karyotype and WES. The coordinate-based inference of SRY and PAR1 involvement demonstrates a practical framework for integrating structural genomics into DSD diagnostics.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13039-026-00771-6.Keywords: Disorders of sex development; Sex-determining region Y protein; Y Chromosome; Copy number variation, OGM Introduction Disorders of sex development (DSD) is a condition in which a person’s chromosome karyotype does not match his or her phenotypic sex, presenting as 46,XX males or 46,XY females. Among them, 46,XX testicular DSD is most often caused by SRY gene translocation, while a subset of cases may involve abnormalities in the SOX9 sex-determining pathway [1, 2]. Affected individuals are typically male in phenotype and may show testicular hypoplasia, low testosterone, hypospadias, markedly elevated gonadotropins, short stature, azoospermia, subtle undervirilization, and variable gynecomastia. The estimated incidence is ~ 1 in 20,000 male births [3–6]. In clinical practice, establishing the genetic basis can be challenging because Y‑derived sequences may be transferred to the X chromosome in diverse and sometimes complex ways. Mechanistically, X–Y exchange events cluster near the pseudoautosomal region (PAR1) and its boundary, where intense meiotic recombination and non‑allelic homologous recombination (NAHR) can precipitate atypical crossovers, boundary shifts, or extension of recombining domains [7, 8]. However, NAHR is unlikely to fully explain multi‑break architectures with internal inversions/deletions; therefore, NAHR should be viewed as one plausible contributor near PAR1 rather than a sufficient explanation for complex SV formation. As a consequence, breakpoints are heterogeneous, and the SRY‑containing Yp segment may be (i) translocated to Xp22.33 with variable junctions, (ii) inserted as an insertional translocation (including insertion onto an autosome), (iii) incorporated into dicentric X–Y chromosomes, or (iv) embedded within multi‑break rearrangements that may include local inversions [9, 10]. Mosaicism, when present, reflects the timing of the rearrangement (e.g., post‑zygotic occurrence) rather than a specific structural configuration and can therefore accompany any of these categories. Approaches that only indicate the presence of Y-derived material do not necessarily localize insertion sites, resolve multi-break architectures, or delineate dosage changes in PAR1, information that can sharpen genetic counseling and guide clinical follow-up. Against this background, high‑resolution, genome‑wide structural methods can complement classical cytogenetics. Optical genome mapping (OGM) is an emerging high‑resolution cytogenetic method that generates genome‑wide single‑molecule maps, and together with low‑coverage copy‑number sequencing (CNV‑seq) can detect a broad spectrum of clinically relevant structural variants (aneuploidies, copy‑number changes, balanced and unbalanced translocations, inversions, insertions, circular chromosomes, and regions of loss of heterozygosity) with high sensitivity [11, 12]. In this context, OGM and CNV‑seq are positioned as complementary tools: CNV‑seq offers dosage profiling at the whole‑genome scale, while OGM reconstructs the structural architecture and pinpoints breakpoints at kilobase‑to‑megabase resolution. Here, we report a male child with a 46,XX karyotype. By combining CNV‑seq and OGM, we localized Y‑chromosome ectopy to the short arm of the X chromosome and reconstructed the underlying Yp architecture, clarifying the structural mechanism of 46,XX testicular DSD in this case. Materials and methods Participants and clinical evaluation A child with male phenotype, unfused scrotum, and hypospadias was evaluated at presentation. Clinical assessment included physical and genital examination, Doppler ultrasound of the scrotum and pelvis, and baseline endocrine testing (including dehydroepiandrosterone sulfate, aldosterone in the supine position, and plasma renin in the supine position) performed in the same clinical laboratory. To investigate chromosomal and single‑gene etiologies, peripheral‑blood karyotyping and whole‑exome sequencing (WES) were ordered concurrently at initial work‑up. Parental demographic information was recorded (father 32 years, mother 29 years). Parental karyotyping was obtained according to standard practice. Maternal prenatal screening and ultrasound records (NIPT-plus and obstetric sonography) were retrieved from the medical record. All procedures adhered to routine clinical protocols. Specimen collection and DNA preparation Fresh EDTA‑anticoagulated peripheral blood from the proband was collected for genetic testing. Genomic DNA for sequencing‑based assays was extracted using standard column‑based methods per manufacturer instructions. For optical genome mapping (OGM), ultra‑high molecular weight (UHMW) DNA was prepared using the Bionano Prep SP Blood and Cell DNA Isolation Kit v2 (Part 80042) following the vendor protocol. DNA quality and quantity were assessed prior to downstream assays. Conventional cytogenetics Peripheral‑blood lymphocyte cultures were established, and GTG‑banded metaphases were analyzed at a target 450–550‑band resolution according to ISCN guidelines. At least 20 metaphases were reviewed. Parental samples were processed with the same procedures in the clinical cytogenetics laboratory. Whole‑exome sequencing WES was performed in a clinical testing laboratory using standard capture‑and‑sequence workflows. Reads were aligned to the GRCh37/hg19 reference, and single‑nucleotide variants and small insertions/deletions were called and annotated using a routine pipeline. Variant interpretation followed contemporary clinical standards, with attention to genes implicated in disorders/differences of sex development (DSD) and hypospadias. WES was used for single‑gene variant assessment; chromosomal copy‑number or karyotype inference from WES data was not the purpose of this assay. CNV sequencing Chromosome copy‑number analysis was performed on the proband’s peripheral‑blood genomic DNA (clinical laboratory: Shenzhen Huada/BGI) using low-coverage CNV sequencing (CNV-seq), which serves as a pragmatic first-line genome‑wide dosage screen in our workflow prior to higher-resolution structural reconstruction by OGM. Libraries were constructed using enzymatic fragmentation, end‑repair and adaptor ligation, and sequenced as single‑end 35‑bp reads at an average depth of approximately 0.4× on the DNBSEQ‑T7 platform (cPAS chemistry). Reads were aligned to GRCh37/hg19, followed by GC correction and duplicate removal. Copy‑number profiles were segmented using a read‑depth–based algorithm, with a nominal analytical resolution of ≥ 100 kb under the laboratory’s quality criteria. Key platform QC metrics for CNV‑seq (e.g., sequencing depth and Q30) and OGM are summarized in Supplementary Table S1. CNV‑seq intervals reported on GRCh37 were converted to GRCh38 using UCSC LiftOver; the harmonized coordinates used for cross‑platform comparison are provided in Supplementary Table S2. Genomic intervals were annotated with reference to OMIM (https://www.omim.org/), DECIPHER (https://www.deciphergenomics.org/), and ClinGen dosage resources (https://www.ncbi.nlm.nih.gov/projects/dbvar/clingen/). CNV pathogenicity was interpreted according to American College of Medical Genetics and Genomics(ACMG)/ClinGen technical standards. Optical genome mapping OGM was performed on the Saphyr system (Bionano Genomics, San Diego, CA). UHMW DNA was direct‑labelled (enzyme recognizing the CTTAAG motif) and stained per the manufacturer’s Direct Label and Stain protocol. Labelled DNA molecules were loaded onto Saphyr chips, linearized in nanochannels, and imaged at high resolution. Raw images were converted into single‑molecule maps and assembled de novo. Structural variants (SVs) encompassing insertions, deletions, inversions and translocations were identified using the vendor’s analysis software (Bionano Solve/Access) with standard parameters, and aligned to the GRCh38 reference genome. Genome‑wide Circos plots were generated in Bionano Access from SV calls passing the platform’s default confidence/quality filters and are configured to highlight inter‑chromosomal connections (e.g., translocations/insertional links), which can result in only a subset of chromosomes being visualized. For case‑level review beyond the primary Y→X event, SVs were additionally filtered using an internal database population frequency ≤ 1% and overlap with OMIM morbid genes.For improved Y‑chromosome visualization, T2T‑CHM13 reference annotations were used for display as noted in the relevant figure legends. OGM‑derived copy‑number profiles were also generated by the platform’s depth‑of‑label algorithm under default parameters. The dataset met recommended quality thresholds, with a molecule N50 length of 238.5 kb, label density of 15.27 labels/100 kb, map rate of 84%, and effective reference coverage of 155.02×. A summary of these quality metrics is provided in Supplementary Table S1 together with additional run parameters. Results Clinical findings and ultrasound Prenatal screening by NIPT-plus was reported as low risk. Obstetric ultrasound in early pregnancy noted bilateral choroid plexus cysts (2022‑11‑03; left ~ 3 × 4 × 4 mm; right ~ 2 × 3 × 3 mm), while subsequent systematic ultrasound (2022‑12‑29) and 4D ultrasound (2023‑02‑07) were reported as unremarkable. The proband was born as a singleton at 38 + 4 weeks’ gestation via vaginal delivery, with birth weight 2800 g and birth length 50 cm. Apgar scores were 10/10 at 1 and 5 min. No neonatal complications or NICU admission were documented. Sex was registered as male at birth based on external genital appearance, and hypospadias with an unfused scrotum was noted during infancy and prompted clinical evaluation. Clinical presentation and imaging: At clinical assessment, the child had male external genitalia with hypospadias and an unfused scrotum. Doppler ultrasound demonstrated bilateral testes located within the scrotum, without obvious abnormalities of the epididymides. No uterine echo was observed behind the bladder, and no ovarian tissue was identified in the abdomen (Fig. 1). Following disclosure of the cytogenetic/structural genomic results, the parents received genetic counseling, and the child has continued follow‑up with male sex of rearing unchanged to the most recent visit. Doppler ultrasound images. The ultrasound image showed obvious testicle echo at the intersection of white dotted line, and no obvious abnormality in other parts Baseline endocrine testing was obtained on 2023‑07‑10. LH was 4.49 mIU/mL (reference 1.24–8.62), FSH was 2.60 mIU/mL (reference 1.27–19.26), and total testosterone was 5.55 nmol/L (reference 6.07–27.10). 17‑hydroxyprogesterone was 608.6 pg/mL. Aldosterone (supine) was 48.6 ng/dL, and plasma renin (supine) was 48.92 uIU/mL; ACTH was 18.5 ng/L. Birth head circumference was not available in the medical record. Longitudinal auxology through the most recent follow‑up (2025‑10‑29) documented height measurements of 58.6 cm (2023‑07‑28), 81.3 cm (2024‑11‑05), 83.8 cm (2025‑01‑20), and 91.2 cm (2025‑10‑29), with a body weight of 12.1 kg at the latest visit. No skeletal findings characteristic of SHOX haploinsufficiency (e.g., mesomelia or Madelung deformity) were documented in available clinical records; however, targeted orthopedic examination and forearm/wrist radiographs have not yet been performed. Continued growth monitoring and skeletal surveillance were recommended as part of follow‑up. Conventional cytogenetics G‑banded karyotyping of peripheral blood lymphocytes showed 46,XX, inv(9) (p12q13) at a 450–550‑band level. The pericentric inv (9) is a common heteromorphism and was considered likely benign with respect to the clinical presentation. CNV-seq test results The results suggested that seq [GRCh37] Xp22.33p22.33(287328–3605487) × 1, Yp11.31p11.2 (2649472–6065425) × 1, Yp11.2p11.2 (7270527–9599178) ×1 (Fig. 2). These calls indicate a heterozygous loss at Xp22.33 (PAR1) and copy‑number losses within Yp11.31–p11.2 and Yp11.2 relative to the Y reference, consistent with the presence of Y‑derived material with internal breakpoints (Fig. 2). CNV-seq test results for patients. (A)Chromosome X copy‑number profile showing a called loss at Xp22.33 (PAR1). (B) Chromosome Y copy‑number profile indicating called losses within Yp11.31–p11.2 and Yp11.2 Optical genome mapping OGM identified a derivative X chromosome harboring an insertion of Yp-derived sequence at Xp22.33. Breakpoint reconstruction delineated three re-join events within Yp11.2: a deleted interval of approximately 6.2–7.3 Mb, a locally inverted interval of approximately 7.3–9.9 Mb, and a junction connecting the remaining Yp segment with the Xp22.33 insertion site (Fig. 3). Because both CNV-seq and the OGM copy-number profile consistently indicate a deletion encompassing Xp22.33/PAR1, the most parsimonious interpretation is that the inserted Y-derived fragment does not include PAR1 and does not restore PAR1 dosage. Optical genome mapping (OGM) reconstruction of the derivative X and Yp architecture. (A) Genome‑wide Circos plot (GRCh38) showing inter‑chromosomal connections between chromosome X and Y consistent with a Yp→Xp insertion. (B) Schematic of the derivative X indicating the insertion site at Xp22.33 and the orientation of the Y‑derived segment. (C) Y‑chromosome (Yp) breakpoint map (aligned to GRCh38, with T2T‑CHM13 annotation consulted for display); three break‑re-join events are shown in Yp11.2, including a deleted interval (~6.2–7.3 Mb) and a local inversion (~7.3–9.9 Mb). The SRY locus (Yp11.3) is indicated on the track for reference. (D) X‑chromosome view (GRCh38) localizing the insertion on Xp22.33. (E) OGM-derived copy-number profile supporting the Xp22.33/PAR1 deletion Based on harmonized coordinates, the retained Yp11.3 interval includes the SRY locus and is inserted into der(Xp22.33), supporting an SRY‑positive 46,XX testicular DSD in this patient. This inference is coordinate‑based from CNV‑seq and OGM integration. Additional filtered SVs detected genome‑wide by OGM are provided in Supplementary Table S3, and none of these SVs (other than the Yp→Xp event) were considered associated with DSD onset after review. Discussion The integrated analysis in this child with a 46,XX karyotype defines a coordinate‑based model in which a Y‑p arm segment, encompassing the SRY locus, is inserted into Xp22.33 and configured by three break‑re-join events within Yp11.2, including a local inversion. This arrangement aligns with the known concentration of X–Y exchange near PAR1 and its boundary and with the documented heterogeneity of breakpoints and junction architectures in SRY‑positive 46,XX testicular DSD [13, 14]. Use of the T2T‑informed Y reference facilitates interpretation of complex Y‑p structure and corrects legacy gaps and misassemblies that affect GRCh38‑Y, which is relevant when localizing genes such as SRY and evaluating ampliconic and repetitive content [15]. At the level of clinical phenotype, the ultrasound pattern—bilateral scrotal testes, absent uterus, and no ovarian tissue—supports testicular differentiation, consistent with SRY‑driven testis determination despite an XX karyotype, while the pericentric inv(9)(p12q13) is likely a benign heteromorphism without causal bearing on the DSD presentation [16]. Within this framework, our data add to the literature by reconstructing the multi‑break Y→X insertion at kilobase‑to‑megabase scale and by linking dosage loss at Xp22.33 to a structural mechanism centered on PAR1. The combined CNV‑seq and OGM strategy advances cytogenetic resolution beyond karyotyping and complements WES by resolving architecture as well as dosage. CNV‑seq at low coverage provides an efficient whole‑genome dosage screen to flag Xp22.33 loss and internal Y‑p deletions that anchor the hypothesis of Y‑material transfer; this role is increasingly recognized in clinical pipelines as an economical alternative or adjunct to array‑based approaches [17, 18]. Although OGM also generates a copy‑number profile, in our workflow it is primarily used as a second‑line assay because it requires UHMW DNA and is intended to resolve derivative‑chromosome architecture rather than to provide the initial dosage screen. Compared with chromosomal microarray analysis (CMA), the low‑coverage CNV‑seq approach used here provided sufficient genome‑wide dosage sensitivity for the megabase‑scale abnormalities observed in this case at lower per‑sample cost in our clinical workflow, although its validated analytical resolution (≥ 100 kb) is not intended to exceed that of high‑density CMA for very small focal CNVs. Both CNV‑seq and CMA can detect mosaic dosage abnormalities, but sensitivity depends on clone fraction, event size, read depth/probe density, and laboratory validation thresholds; importantly, neither method resolves balanced or complex multi‑break rearrangement architecture. OGM then reconstructs the derivative chromosome, pinpoints breakpoints, and exposes local inversions that are invisible to dosage‑only methods, with recent multi‑site validations and implementation frameworks demonstrating analytic reliability and practical guidance for clinical laboratories [19, 20]. Structural clarification is not merely descriptive. It localizes the insertion to Xp22.33, orients the Y‑segment, and enumerates Yp11.2 breakpoints, all of which strengthen variant interpretation and inform counseling about recurrence risks and parental testing strategies [21]. Interpreting Y‑derived structure against improved Y references further reduces alignment ambiguity, which is critical for regions rich in palindromes and segmental duplications [22]. Together, these elements show why architecture‑level evidence can be decisive in DSD work‑ups that begin with broad differential etiologies and may not be solved by locus‑specific assays alone. The junction at Xp22.33 brings PAR1 dosage explicitly into consideration and raises anticipatory guidance about growth and skeletal outcomes. Deletions involving PAR1 frequently encompass SHOX and may extend into neighboring genes within Xp22.33, including ARSL (HGNC‑approved symbol; historically reported as ARSE in parts of the literature) [23, 24]. Importantly, the evidence for short stature enrichment in PAR1 deletions is primarily attributable to SHOX haploinsufficiency; therefore, we frame growth surveillance around SHOX rather than inferring an independent growth contribution from ARSL [25]. Beyond canonical PAR1, population‑scale studies show that the human PAR1 boundary exhibits evolutionary and inter‑individual variability, including evidence for shifts in recombining domains and the presence of X‑derived “extended PAR” segments on some Y chromosomes [26]. These findings reinforce the biological plausibility of heterogeneous breakpoint locations in 46,XX testicular DSD and support a counseling stance that integrates growth surveillance with structural genotype, rather than treating SRY transfer as the sole determinant of outcome. In this case, architecture‑level mapping therefore has immediate clinical utility: it explains the dosage finding at Xp22.33, highlights genes of follow‑up interest, and delineates the inserted Y‑segment in a way that can be communicated clearly to the family. Consistent with this counseling focus, longitudinal growth data through 2025‑10‑29 are provided in the Results, and no SHOX‑typical skeletal abnormalities have been documented to date (with ongoing surveillance planned). Positioning of methods within a pragmatic diagnostic pathway is also informed by current DSD guidance. Consensus‑style frameworks emphasize staged but flexible evaluation that combines clinical phenotyping, endocrine testing, and cytogenetics with molecular assays tailored to the presentation [27–29]. In that context, PCR or QF‑PCR remain low‑cost confirmatory assays for SRY per se, particularly when the differential diagnosis has already converged on suspected SRY transfer; however, they do not resolve multi‑break architectures, insertion sites, or orientation, nor do they delineate PAR1 dosage or gene content [30]. Our coordinate‑based inference of SRY inclusion and localization arises from convergent CNV‑seq and OGM evidence and illustrates a scenario in which genome‑wide structural methods establish the map first and locus‑specific tests become optional for confirmation rather than discovery [31]. Similar logic has been applied in prenatal and constitutional settings, where OGM has clarified complex rearrangements that were ambiguous by conventional methods and in which rapid, architecture‑aware results directly shaped counseling and management [32, 33]. Importantly, the present case underscores that “presence of SRY” and “phenotype” are not interchangeable; phenotypic range exists even with SRY on an X derivative, and mapping helps explain that range by showing what else moved, flipped, or was lost at the breakpoint neighborhood. Mechanistically, while NAHR near PAR1 can account for many canonical X–Y exchanges, the PAR1 boundary itself shows inter‑individual variability, which can complicate inference of telomeric/PAR involvement across individuals and reference resources [34]. In the present case, the multi‑break architecture (internal deletion plus local inversion) suggests that NAHR alone is insufficient; replication‑based mechanisms (e.g., FoSTeS/MMBIR) and error‑prone end‑joining, as well as catastrophic chromoanagenesis events such as chromothripsis, have been proposed to generate clustered breakpoints and complex rearrangements [35, 36]. Definitive mechanism assignment would require nucleotide‑level breakpoint sequencing, which was not performed in this case. Several caveats qualify these conclusions. First, we did not perform PCR or FISH to directly verify the presence of SRY or to cytogenetically localize it because no residual biological material was available for retrospective testing. Therefore, SRY inclusion and its placement on the derivative X are inferred from the concordant CNV‑seq/OGM breakpoint coordinates and may be influenced by reference annotation biases and the repetitive architecture of Yp. Second, breakpoint junctions were not resolved at nucleotide level, which limits definitive inference of the underlying rearrangement mechanism. In addition, T2T‑CHM13 is derived from a single individual and may not fully capture population diversity, potentially affecting interpretation in highly repetitive Y‑chromosome regions. Core technologies also have inherent limits: OGM has reduced sensitivity in heterochromatin and low‑complexity intervals, while low‑depth CNV‑seq primarily detects dosage changes and may miss balanced events, small indels, or low‑level mosaicism associated with X–Y breakpoints. Importantly, all analyses were performed on peripheral blood; tissue‑restricted mosaicism (including gonadal mosaicism) cannot be excluded. Moreover, this report describes a single patient, and Yp ectopy and breakpoint configurations are highly heterogeneous across individuals; thus, the observed “Yp insertion into Xp22.33 with three Yp11.2 breakpoints” should be interpreted as exploratory and not necessarily representative of a recurrent rearrangement pattern. Despite these limitations, the combination of CNV‑seq and OGM yields a coherent structural explanation for 46,XX testicular DSD in this case, delineates counseling‑relevant dosage information at Xp22.33, and offers a practical framework for integrating genome‑wide structural methods into DSD diagnostics. Future optimization should include targeted SRY confirmation (e.g., PCR/FISH), breakpoint‑level sequencing when feasible, standardized reference/annotation resources and harmonized analysis pipelines, and larger cohorts (ideally with parental breakpoint evaluation) to support broader clinical integration. Conclusion Our integrated CNV‑seq and OGM analysis delineates the structural mechanism of SRY‑positive 46,XX testicular DSD by reconstructing a multi‑break Yp segment inserted into Xp22.33. Coordinate‑based mapping demonstrates that the retained Yp interval includes SRY and defines breakpoint orientation together with a PAR1 deletion on Xp22.33. This architecture‑level evidence shows the complementary value of low‑coverage CNV‑seq for genome‑wide dosage screening and OGM for high‑resolution structural reconstruction beyond karyotyping and locus‑specific assays. Clinically, the map informs genetic counseling and follow‑up by highlighting growth surveillance when PAR1/SHOX is affected and by guiding discussions of recurrence risk and parental testing. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Acknowledgements We would like to thank the patient and his family for their participation and cooperation in this study. We also acknowledge the support of the staff at the Center for Reproductive Medicine, Wenzhou People’s Hospital. Author contributions HW and JZ drafted the manuscript. HW and WH were involved in obtaining the sequencing results. ZX, XZ, HP and HZ contributed to the collection of clinical data. JZ designed and supervised the study. All authors read and approved the final manuscript. OGM imaging and CNV‑seq library construction/sequencing were performed as clinical services by Bionano/affiliated laboratory and Shenzhen Huada/BGI, respectively. Funding This work was supported by Wenzhou City Major Scientific and Technological Innovation Project (Research on New Technologies for Women, Children and Reproductive Diseases) (ZY2024023), and Lishui Key Research and Development Program Project (2025ZDYF24). Data availability The datasets generated and/or analyzed during the current study are not publicly available due to patient confidentiality and privacy regulations but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The study was approved by the Wenzhou People’s Hospital (Approval No.KY-202508-035). Written informed consent for genetic testing and publication of de‑identified clinical data and images was obtained from the patient’s parents/guardians. 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. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Data Availability Statement The datasets generated and/or analyzed during the current study are not publicly available due to patient confidentiality and privacy regulations but are available from the corresponding author on reasonable request.