There is lot's of room for improvement though. All of the commercial nanopore tech is based on biological nanopores, which have the advantage of having very straight-forward to fabricate. But they are limited to the ionic current signal, which is very noisy and weak. Once these companies start introducing solid-state devices though, things will begin to get very interesting as alternative signal transduction mechanisms come into play.
It seems like a fun problem in information theory. Can you point us to some articles or papers about current approaches to solving it?
Also I think the method you are describing is their other sequencing approach - this one, as far as I know, passes an intact strand through each pore and examines the electrical conductivity of overlapping 6 base pair sequences.
People may be specifically interested in the first and second talks with the corollary of both being likely biased towards the MinIon:
- 1st is by Clive Brown (ONP CTO) and discusses the MinIon platform, background to the Nanopore technology, analysis platforms and the future PromethIon expansions - https://www.youtube.com/watch?feature=player_detailpage&v=Ut...
- 2nd has Nick Loman (one of the MAP researchers but admitted 'fanboy') discussing the performance on the MinIon in his lab in 'real world' conditions - https://www.youtube.com/watch?feature=player_detailpage&v=Ut...
Incredible work, the stamina on display is extremely impressive and I hope they and their backers will reap the rewards from all the hard work very many times over.
To put in laymans terms what this thing is: it's a tape-playback machine for DNA.
Now they need to fix the bugs (hard, but probably not nearly as hard as getting to this stage in the first place).
http://www.economist.com/news/technology-quarterly/21615029-...
"One of these, Genia, is commercialising a process called nanopore sequencing that Dr Church first devised in 1988. Distinct polymer tags are attached to each of the four nucleotides poised to contribute to a single molecule of replicating DNA. As they react, the tags are released near a protein layer full of tiny holes called nanopores. Each tag blocks the flow of electrical ions across the layer in a different way. Because it relies on electronics rather than optics, nanopore sequencing promises faster, cheaper sequencing. Dr Church holds up a fingernail-sized chip containing 128,000 nanopores that he reckons will bring the cost of sequencing down to $100. In June, Genia was acquired by Roche, a Swiss pharmaceuticals giant."
Having said that, for cancer genomics, the vast majority of archival tumour tissue in the world is stored in a way that auto-fragments the DNA, so being able to do long reads won't actually help...
It's a great achievement and I'm hopeful, but the accuracy needs to go up before it competes with current standard.
Like that is not even a blip in a normal research budget for equipment.
