Early Data on Human Sequencing with Oxford Nanopore MinION

By Bio-IT World Staff

January 22, 2015 | A team based at the University of Toronto has published the first reported case of using the Oxford Nanopore MinION to sequence human DNA, in a study focused on three genes with known relevance to drug dosing. The team was led by Gary Bader, a computational biologist who previously worked with Chris Sander at Memorial Sloan-Kettering Cancer Center. The MinION, currently in early access, is the first DNA sequencer based on nanopore technology, which could soon open up genetics to a much broader set of users with its low cost and relative ease of use. (For Bio-IT World's deep feature on the instrument, see "Nanopore Sequencing Is Here to Stay.")

Because of the MinION's low throughput, the device is best-suited to sequencing bacterial and viral genomes, or those of simpler eukaryotes like yeast (as in this de novo assembly of a yeast genome posted to bioRxiv). However, it could also be used for narrow applications in human genetics. For instance, its long read profile makes the MinION a promising instrument for filling gaps in the human genome in regions with very low complexity or long stretches of tandem repeats. Bader's team chose to concentrate on a key clinical need: long-read sequencing of genes with high structural variation, where variants affect the metabolism of certain drugs.

"Currently, only a few hospitals with large research programs are able to purchase and use massively parallel sequencing instruments," said Bader and first author Ron Ammar in an interview with F1000Research. "In anticipation of a revolution in DNA-based diagnostics, we sought to be among the first researchers to investigate the power and pitfalls of nanopore sequencing for human diagnostic applications."

The team amplified DNA from the HLA-A, HLA-B and CYP2D6 genes, using the well-characterized cell line NA12878 so their results could be confirmed. Sequencing on the MinION, and using the BLASR program for alignment, Bader and colleagues were able to completely sequence and haplotype the three targeted regions, although significant minorities of reads also supported incorrect haplotypes. This result suggests that the MinION could one day be appropriate for clinical use in making prescription decisions, but that its high error rate remains a concern. The full study, "Long read nanopore sequencing for detection of HLA and CYP2D6 variants and haplotypes," is posted on F1000Research, an open access, open peer review journal, and the raw data is available through Figshare for readers to use in replication. The paper has not yet gone through peer review.

"I’m hoping that our rapid publication model can be a venue for new nanopore data," said F1000Research Associate Publisher Michael Markie by email. "To get it out quickly and have all the data readily available is crucial at this point... I think it’s a good way to get all the MAP [MinION Access Program] users working together on potential issues as they can see one another's data."

While this study presents the first publicly available data on human sequencing with the MinION, Bader's team is not the only group pursuing human applications with the technology, and further research like this is likely to be published in the coming months, even before the MinION's commercial release. MAP members are hard at work finding niche areas of human genetics that could benefit from nanopore sequencing with a low-throughput instrument. Meanwhile, those interested in whole human genome nanopore sequencing will have to wait for the launch of Oxford Nanopore's first high-throughput instrument, the PromethION, which may enter early access this year.