Use of stereotypical mutational motifs to define resolution limits for the ultra-deep resequencing of mitochondrial DNA.

Change log
Gardner, Kristian 
Payne, Brendan AI 
Chinnery, Patrick F 

Massively parallel resequencing of mitochondrial DNA (mtDNA) has led to significant advances in the study of heteroplasmic mtDNA variants in health and disease, but confident resolution of very low-level variants (<2% heteroplasmy) remains challenging due to the difficulty in distinguishing signal from noise at this depth. However, it is likely that such variants are precisely those of greatest interest in the study of somatic (acquired) mtDNA mutations. Previous approaches to this issue have included the use of controls such as phage DNA and mtDNA clones, both of which may not accurately recapitulate natural mtDNA. We have therefore explored a novel approach, taking advantage of mtDNA with a known stereotyped mutational motif (nAT > C, from patient with MNGIE, mitochondrial neurogastrointestinal encephalomyopathy) and comparing mutational pattern distribution with healthy mtDNA by ligation-mediated deep resequencing (Applied Biosystems SOLiD). We empirically derived mtDNA-mutant heteroplasmy detection limits, demonstrating that the presence of stereotypical mutational motif could be statistically validated for heteroplasmy thresholds ≥ 0.22% (P = 0.034). We therefore provide empirical evidence from biological samples that very low-level mtDNA mutants can be meaningfully resolved by massively parallel resequencing, confirming the utility of the approach for studying somatic mtDNA mutation in health and disease. Our approach could also usefully be employed in other settings to derive platform-specific deep resequencing resolution limits.

Case-Control Studies, DNA, Mitochondrial, High-Throughput Nucleotide Sequencing, Humans, Intestinal Pseudo-Obstruction, Mitochondrial Encephalomyopathies, Muscular Dystrophy, Oculopharyngeal, Mutation, Nucleotide Motifs, Ophthalmoplegia, Sensitivity and Specificity
Journal Title
Eur J Hum Genet
Conference Name
Journal ISSN
Volume Title
Springer Science and Business Media LLC
Wellcome Trust (101876/Z/13/Z)