Analysis of the whole mitochondrial genome: translation of the Ion Torrent Personal Genome Machine system to the diagnostic bench?

Mitochondrial (mt) diseases are a heterogeneous group of disorders, caused by both nuclear and mitochondrial genome mutations. A correct diagnosis is challenging, mainly because of the absence of clear phenotype-genotype correlations, the existence of heteroplasmy and the very large number of genes involved. Current traditional molecular diagnoses for disorders caused by a mtDNA defect, relies on the identification of (common) point mutations and large deletions with sequencing and Southern blot procedures. Complete mitochondrial genome resequencing, using Sanger nucleotide sequencing techniques, is reserved only for few well selected patients. The method is laborious, and not very sensitive to detect nucleotide variations below 15-20% heteroplasmy. Next generation sequencing (NGS), an innovative sequencing technology which enables the successful analysis of numerous gene sequences in a massive parallel sequencing approach, has revolutionised the field of molecular biology. Although NGS was introduced in a rather recent past, the technology has already demonstrated its potential and effectiveness in many research projects, and is now on the verge of being introduced into the diagnostic setting of routine laboratories to delineate the molecular basis of genetic disease in undiagnosed patient samples. We tested a benchtop device on retrospective genomic DNA samples of controls and patients with a clinical suspicion of a mtDNA disorder. This Ion Torrent PGM platform is a high throughput sequencer with a fast turn-around time and reasonable running costs. We challenged the chemistry and technology with the analysis and processing of a mutational spectrum composed of samples with single nucleotide substitutions, indels (insertions and deletions) and large single or multiple deletions, occasionally in heteroplasmy. The output data were compared with previously obtained conventional dideoxy sequencing results and the mitochondrial revised Cambridge Reference Sequence (rRCS).We were able to identify the majority of all nucleotide alterations, but three false negative results were also encountered in the data set. Large single deletions, with visualization of their breakpoints, as well as multiple deletions were detected. At the same time, the poor performance of the PGM instrument in regions associated with homopolymeric stretches generated many false positive miscalls demanding additional manual curation of the data.