The family provided written informed consent, and this study was approved by the appropriate local institutional review boards on human subject research at the First Affiliated Hospital of Zhengzhou University.

The genetic relationship of the proband and the parents was confirmed by quantitative fluorescent polymerase chain reaction (QF-PCR) using the Goldeneye DNA ID System 20A Kit (Peoplespot, Beijing, China).

A multiplex ligation-dependent probe amplification (MLPA) assay was performed using the SALSA MLPA Kit P034/P035 for DMD (MRC-Holland, Amsterdam, Netherlands).

WES was performed using Illumina library construction and capture kits (Illumina, San Diego, California, USA) according to the manufacturer’s instructions. Briefly, genomic DNA was fragmented with enrichment Bead-Linked Transposomes (eBLT) in Tagmentation Buffer 1 (TB1) (Illumina), with a target fragment size of 200 to 300 bp. The fragment ends were repaired, and an adenosine residue was added to the 3′ end of the fragments. Adaptors were then ligated to the fragments using the Fast DNA Library Prep Set for Illumina CW3045M (CWBIO Inc., China). Exon-containing libraries were captured using IDT xGen exome baits (Integrated DNA Technologies, Inc., USA), and quality and purification were assessed using Qubit 4.0 (Thermo Fisher Scientific Inc., USA). Also, 150 bp pair-end sequencing was conducted on NovaSeq 6000 for sequencing depths greater than 100×.

After sequencing, paired-end reads were trimmed using Trimmomatic version 0.36 (Bolger et al., 2014) to filter out low-quality reads and adaptors from the dataset. High-quality reads were aligned to the human reference genome GRCh37/hg19 using Burrows–Wheeler Aligner MEM version 0.7.17 (Li and Durbin, 2009), and the duplicates were removed using Picard version 2.9.0 (http://broadinstitute.github.io/picard). Small variants were identified using the Genome Analysis Toolkit (GATK) version 3.8 (McKenna et al., 2010). Copy number variants (CNVs) were detected by comparing the coverage of depth between the target sample and a baseline, which was determined using other male control samples in the same pipeline. In addition, the breakpoints of CNVs were called in the patient’s BAM file by detecting soft-clipped and abnormal mapping orientation reads. The script for calling CNVs was run using the R programming language, and details are described in Supplementary File 1.

The breakpoints of the partial exonic deletion were confirmed by PCR and Sanger sequencing of DNA samples from the proband, the proband’s mother, and a healthy man (a control subject unrelated to this family without any muscular diseases). The primers used for sequencing are listed in Supplementary Table S1. Primers were designed using Primer-BLAST and were based on the combination of sequences on both sides of the deletion. The PCR product of the mutant type was approximately 1 kb, while that of the wild type was approximately 12 kb. Lastly, the sequences were aligned to the human reference genome sequence (GRCh37/hg19) using BLAST version 2.9.0 (Camacho et al., 2009) to identify the breakpoints of the large deletion.

The genomic DNA extracted from the fresh peripheral blood of the proband was purified with AMPure XP magnetic beads (Beckman Coulter, CA, USA) and electrophoresed to confirm its integrity. The concentration of the DNA was 44.0 ng μl⁻¹ when measured using Nanodrop 2000 (Thermo, Massachusetts, USA) and 43.4 ng μl⁻¹ when measured using Qubit 4.0. The DNA input mass for the single library preparation without topping-up or refuel was 2.0398 μg. LR-WGS was carried out without fragmentation using the SQK-LSK109 kit (ONT, Oxford, UK) on MinION (ONT) with R9.4 flow cells (#FLO-MIN106, ONT) according to the manufacturer’s instructions (GDE_9603_v109_revW_Aug 14, 2019) for 72 h. Raw fast5 data were basecalled using Guppy GPU (version 4.3.4) with a high-accuracy base calling model. After base calling, the sequencing data were aligned to the GRCh37/hg19 reference genome using minimap2 (version 2.17-r941) with “map-ont” model, and structural variants (SVs) calling was performed using sniffles (version 1.0.11) with the option “-s 1 -r 1,000 -q 20”.

The proband was a 10-year-old boy who had developed more slowly than his peers since birth—for example, he was unable to walk until he was approximately 1.5 years old. As time progressed, both his calves became thick and hard (Figure 1A). He also struggled with walking, running, ascending stairs, squatting, and standing up. Because the proband’s family was living in a remote and underserved area, the proband remained undiagnosed until July 2020, when he was diagnosed with DMD in Henan Children’s Hospital. Electromyogram results confirmed myogenic damage of the nerves and muscles of both lower limbs and the left upper limb of the proband and the mother (Supplementary Table S2). The concentrations of creatine kinase (CK) and creatine kinase isoenzymes (CK-MB) in the venous serum of the proband were 14,984.0 U/L and 803.0 U/L (Supplementary Table S3), respectively. The proband had no family history of DMD (Figure 1B). Since the pathogenetic cause was not confirmed in Henan Children’s Hospital, the family came to our center in August 2020 to seek additional information about the genetic etiology of DMD.

MLPA analysis suggested that there was no deletion or duplication of exons in DMD (NM_004006.2) of the proband and the mother (Supplementary Figures S1A,B, S2A,B). Furthermore, the accuracy of kinship among the proband and the parents was confirmed by QF-PCR (Supplementary Table S4; Supplementary Figure S3). The quality control of the WES data is summarized in Supplementary Table S5. No candidate SNVs or small INDELs in DMD were found, and there was no significant variation in exon coverage. However, a 11,593-bp hemizygous deletion (c.7310-11543_7359del, chrX:g.31792260-31803852) spanning exon 51 and intron 50 in DMD was identified by CNV and breakpoint analysis of the WES data (Figure 1C–E). The 5′ breakpoint was only 50 bp from the 5′ end of exon 51 (Figure 1G).

The breakpoints of the large deletion were further confirmed by agarose gel electrophoresis and Sanger sequencing (Figure 1F). The results confirmed that the 5′ end 50 bp (chrX:31792260–31792309) in exon 51 and 3′ end 11,543 bp in intron 50 of DMD were hemizygous deletion in the proband. Furthermore, this deletion was de novo, since the PCR product of the proband was approximately 1 kb, and those of the mother and the negative control were between 10 and 15 kb (Figure 1H).

The quality control of the LR-WGS data is presented in Supplementary Tables S6–S8 and Supplementary Figure S4. A total of two reads captured the large deletions in DMD, one of ∼40 kb (read name: 97d7f30e-0f30-4ddc-b66d-c8c82c0fb221; chrX:31790731–31792260, chrX:31803853–31842466) and the other of ∼2.3 kb (read name: 1e46feed-d90e-44b8-bc47-56eb0495b76d; chrX:31791318–31792259, chrX:31803853–31805193) (Supplementary Figure S5, red arrows; Supplementary VCF File, line 1,569). The breakpoints of the large deletion were consistent with those identified by WES and Sanger sequencing. The structural variants (SVs; >1 kb) are listed in the Supplementary VCF File (GRCh37/hg19).