Unlike others, who are disappointed by the first presentation of Oxford Nanopore data today, we feel excited. Why? It is because now we can move beyond all kinds of speculations regarding what miracles nanopore sequencing will achieve and focus on actual science of data quality, noise reduction algorithms, reconstruction and realistic road-map. Also, we can discuss the merits of competing technologies by Oxford Nanopore, Quantum Bioscience, PacBio and maybe even Moleculo (reconstructed long read), if getting long reads is the real goal.
The blog nextgenseek covers all tweets from David Jaffe’s talk in storified version.
Allseq blog presented a nice summary of the talk reconstructed from tweetstreams.
Mostly what weve heard before there are 512 nanopores/cm2 (on the MinION) and each pore can read sequence at a rate of 1-100 bps (the presented dataset was run at 25 bps). The new thing was the read profile of overlapping 6mers -presumably meaning that 6 bases at a time occupy the nanopore.
In the presented dataset the average read length with 5kb, with some reads up to 20kb. Those are nice long reads, but not that impressive in the face of what PacBio is doing these days. It was stated that the read length was limited by the quality of the sample and that longer reads will be possible with better quality samples or more careful prep. This is in line with what weve heard from PacBio prep is now the limiting factor for read length. (However, it should be noted that Nabsys seems to not have a problem getting
The read quality seems to have an unusual profile. There are stretches of perfect reads (84% of >5kb reads have at least a perfect 50mer and 100% have at least a perfect 25mer) that are interrupted by error-blocks that appear to be systematic artifacts. These error-blocks are dominated by indels, mostly deletions. ONT is working on improving this through the use of multiple pore types (which presumably will have non-overlapping error profiles).
Lex Nederbragt also chimed in.
Here are a few of my thoughts about this presentation. On its own, this is a fantastic breakthrough for nanopore based DNA sequencing. Not earlier has a nanopore platform been able to show this kind of data. So, congratulations are in place for Oxford Nanopore.
The lengths of reads presented may seem disappointing given the presentation at AGBT two years ago, where 100 kbp reads were mentioned. However, the limit may have been the size distributions after fragmentation. As users of the PacBio platform know, fragmentation is an art for obtaining long reads, so there is a clear potential for improvement here.
The error-model, which seems to be biased and systematic, is worrying from an assembly and variant calling point of view. It will probably improve over time. On the other hand, there are, as Jaffe eluded to, lots of applications where having (near)-perfect reads are not crucial. Think scaffolding, structural variation detection, isoforms for RNA (cDNA) sequencing, etc.
There remain a few open questions, though. Nothing was mentioned on coverage of the genomes sequenced (a clear miss by Jaffe). Hopefully, there will be no GC bias, for example. A careful look at the errors is needed, for example, are there chimeric reads, adaptors present, long-range errors in the reads?
What about the comparison with other long-read platforms? The data presented is somewhat similar to moleculo data in terms of length, with a lower quality and a yet-to-be determined GC bias. PacBio reads are significantly longer, and of a better quality after self error-correction. The big difference is of course price. A PacBio instrument, and a HiSeq, are very much more expensive than the MinION. A small USB stick also takes up much less space, and can be carried around to the field, especially if library prep can be made small enough.
Erika Check Hayden of Nature wrote -
MinION results are promising, but fall short of high expectations.
Jaffe also found that the MinION appeared to have difficulty sequencing particular parts of the bacterial genomes that he studied. That is worrisome to bioinformaticians, because it is more challenging to correct for systematic sequencing errors than random ones. In Jaffe’s case, the recurring errors prevented him from assembling the complete genome sequences of the two bacteria from scratch using only MinION data; instead, he used MinION sequences to supplement data generated by Illumina machines.
Yet he sounded an optimistic note about the MinION’s future. Oxford has said that higher-quality DNA or different preparation methods should increase the average read length. Jaffe says that the company can work to eliminate errors, perhaps by using a mix of pores with different properties. But even now, he says, the vast majority of long MinION reads had lengthy stretches without any mistakes. The systematic errors are a temporary feature that we hope they’ll figure out how to solve.
Other researchers will soon have a chance to form their own opinions about the new device, as Oxford today launched its early-access programme. Researchers who pay a US$1000 deposit, plus $250 for shipping costs, can receive MinIONs on which they can run their own experiments.