Characterization of Large-Scale Mitochondrial DNA Rearrangements Using Complementary Short- and Long-Read Sequencing...

Large-scale mitochondrial DNA (mtDNA) rearrangement, including deletion or duplication, is one of the important causes of primary mitochondrial disease. In clinical practice, short-read sequencing (SRS)-based assays typically detect these rearrangements by identifying regional mtDNA copy-number changes but generally do not directly distinguish deletions from duplications. As a result, mtDNA duplications may be misclassified as deletions, potentially leading to inaccurate molecular diagnoses and genotype-phenotype correlations. This study aimed to evaluate the utility of long-read sequencing (LRS) for the definitive characterization of large-scale mtDNA rearrangements.

Two individuals who had been reported to possess single large-scale mtDNA deletions using SRS-based clinical tests were re-evaluated because their clinical presentations were not consistent with typical mtDNA deletion syndromes. Existing SRS data were reanalyzed, and the suspected rearrangements were independently characterized using both Pacific Biosciences (PacBio) and Oxford Nanopore Technologies long-read sequencing platforms.

Reanalysis of SRS data suggested that the reported deletions were more consistent with mtDNA duplications. This finding was confirmed by both PacBio and Oxford Nanopore LRS in one case. Reclassifying the deletion to a duplication in this case substantially improved genotype-phenotype interpretation and better explained the atypical clinical features observed in this individual. However, LRS could not capture sufficient long reads that give direct evidence of duplication in the second case due to low heteroplasmy of the mtDNA structural variant. Therefore, this case remains inconclusive.

SRS and LRS provide complementary information for the evaluation of large-scale mtDNA rearrangements. SRS can sensitively identify mtDNA large deletions, but it cannot distinguish mtDNA deletion from duplication. On the other hand, LRS can provide direct structural information and confirm duplication. However, LRS may have limited sensitivity for low-level heteroplasmic structural variants. An integrated approach incorporating both methods, together with clinical correlation and reanalysis of potentially discordant findings, may improve the characterization and interpretation of mtDNA rearrangements.