Ibo Gut from CNAG-CRG in Spain presented at London Calling 2022 a session entitled “Novel method for multiplexed, full-length single-molecule sequencing of human mitochondrial DNA using Cas9-mediated enrichment.” I don’t know much about Cas9-mediated enrichment and have been curious. Gut spoke about mitochondrial DNA and how the human mitochondrial (mtDNA) is circular, 16.6 kb, has 37 genes, is inherited maternally… and does not have a repair mechanism. Gut explained that it interacts with numerous nuclear genes (>250). Their team used Cas9 to target and enrich for mtDNA. They take genomic DNA, linearize it, and dephosphorylate it to “move it out of the picture.” They then use Cas9 for cleavage and barcoding of mtDNA. They use different cut sites and are then able to barcode multiple samples. Demultiplexing uses minimap. Gut explained how they map to a synthetic “double mitochondrial DNA reference” to help with the challenges of circularity and read “crossover.” The analysis defines the breakpoints, and the output is a VCF file. Gut shared a complex flowchart of the tools they use to analyze reads and produce VCF files and reports of variants. Gut shared an example of detection of low frequency pathogenic variants that are ~2 copies in 1,000. I was intrigued by Gut’s comparison of Illumina and long-range PCR-ONT sequencing of mtDNA and the resulting coverage There are still gaps because of the bias of long-range PCR for short fragments. Cas9 enrichment with specific sites allowed for coverage of the entire mtDNA. Gut summarized that the Cas9-mtDNA-enrichment and bioinformatics pipeline allows for :
- multiplexing in a cost-effective way
- full-length single-molecule mitochondrial genome sequencing
- variant calling
- detection of heteroplasmy below 1%
- detection and quantification of multiple deletions
- solving of haplotypes
While I don’t understand some of the details, the use of Cas9 and dephosphorylation of genomic linearized DNA seems like a very effective approach. Further, custom Cas9 guides can help ensure appropriate coverage. I wonder how we can bring this approach into classes to enrich for and sequence genomic regions of interest.
