New Tech at AGBT

[BBoy]

Originally Posted by JamesH

I've been thinking more about the disruption that the nanopore tech could cause, there's been a bit of discussion regarding the "niche" of de novo genome assembly, and granted this will be the first thing that people want to get their hands on a system to do (because of the long reads). But for me, "run until", and no library prep are what will probably make the bigger difference, you don't have to just do long reads: there's no reason not to put shorter fragments in. In our lab we do a lot of population genetics involving non-model plants and animals, this mostly involves genotyping a lot of individuals (usually at a cost of $10-30 per individual). While the cost of sequencing has been coming down the time/cost/hassle of library prep has been the main barrier for us in going NGS (having to maintain 100's of barcoded primers etc). Straight away I can see several ways that I can make restricted libraries for each individual, not bother with PCR and any associated errors, then run a Gb or two until sufficient coverage is reached and do genotyping by sequencing for close to the cost of what we currently do. The ability to do RNAseq gene expression studies in the lab without having to send off to a core facility (even if you have to make cDNA) will also be pretty awesome.

Yes, this is something that in my sniping about parallelism I seriously overlooked. There is nothing that prevents you from heavily fragmenting DNA and running short reads. In fact, there is nothing preventing you from running PCR and then sequencing if you have a specific need for a specific short segment. Shows you how easily I develop myopia...

Originally Posted by JamesH

I can also see applications for the USB stick in agriculture to allow field-based monitoring of resistance to insecticides/herbicides. I think that putting this kind of sequencing power straight into the hands of researchers is going to be a big game changer, and the more I think about it the less I can see myself doing a lot of the things that I currently do in the lab!

Absolutely, as long as the bio-informatics eventually are brought up to iPhone level, i.e., a big green and a big red button with little thinking in between.

[Jeremy]

Originally Posted by clivey

I can see the are a lot of questions and misconceptions which we can easily answer. If you have further questions please email me directly from your institutions email account.

Many of us are very interested in the discussion and your answers. Why not continue them here instead of answering the same questions over and over by email?

I wasn't able to attend the talk, but am interested in how much sequence you can expect out of a single gridION cartridge. From what I have read on the site, if you want to maximise the amount of sequence you set it to run until read quality falls below X. How many days would that typically take (assuming you set x to whatever would be reasonable for a large (2-3 Gb) whole genome assembly) and what sort of sequence volume are we talking?

[gringer]

How many days would that typically take (assuming you set x to whatever would be reasonable for a large (2-3 Gb) whole genome assembly) and what sort of sequence volume are we talking?

In my post on the "disruptive technology" thread, I came up with about 200kb per pore per hour:

http://seqanswers.com/forums/showpos...3&postcount=44

This was from an estimate of GridION processing speed of 10Gb per day on 2000 pores ("tens of gigabases per day"). Just for comparison, using this 10Gb estimate the MinION (which has a run-time limit of about 6 hours, and 1/4 the number of pores) should be able to process about 600Mb before the reagents run out.

[Jeremy]

10Gb per day does sound very impressive. But everything I have read still only says a run can take "days" and that you just stop it when you are ready. It doesn't actually state how many days you can get good quality data from a single cartridge assuming enough good quality sample is available. That's a pretty significant factor if you want to sequence a whole genome.

[gringer]

Originally Posted by Jeremy

It doesn't actually state how many days you can get good quality data from a single cartridge assuming enough good quality sample is available. That's a pretty significant factor if you want to sequence a whole genome.

This is not particularly important. Quoting myself from this thread:

Samples can be ejected during a run and shifted to other cartridges, cartridges can be replaced, runs can be started at any time, and stopped due to a number of different desired factors, so any drop-off in occupancy probably won't have too much impact on result output (as long as there's still money to burn).

As long as you have money and monkeys, you can sequence as many bananas as you want. With the "release date" 2000-pore GridION, a 20-node system should do a human genome in an hour (you'd need to split the sample into 20 for this), or a 5-node system in 4 hours:

http://www.nanoporetech.com/news/press-releases/view/39

I guess that's assuming 5-20x coverage for the human genome, depending on how you interpret "tens of gigabases", and also suggests a genome should be possible in one day on a single node with reasonable coverage.

With 20 MinIONs, it would take about 4 hours assuming similar processing speed.

[Markiyan]

I remember reading somewhere (2-3 month ago), that there were two main types of the nanopores in their's R&D -
1. Protein nanopore fused with Exonuclease, and the nucleotides are first cleaved away by the exonuclease, than they flow through the pore, and the change in the current flowing through it is detected. Difference in the ion charge is then analysed and is used for basecalling.
2. Purely solid state nanopores - not much details on those - whether they are chopping the DNA up (as in 1, or threading it through)?

From our current error profile report it confirms 1, that the deletions are caused by the bases, that had "run away" from the pore.
I think going for the low temperatures enzymes may reduce the noise caused by the molecular motion, but then the secondary structures may became even bigger issue. (any reaction conditions reports)?

PS: near the end of the run you should see the increase in the insertions (caused by the saturation of the run-away bases, being finally dragged in by the electric current).

PPS:
Fortunately it doesn't have the fundamental flaw of the current pacbio implementation, there the presence of the base in the active centre doesn't 99% warranty it's incorporation (it does in ~87% cases :-).

[gringer]

Originally Posted by Markiyan

I remember reading somewhere (2-3 month ago), that there were two main types of the nanopores in their's R&D -
1. Protein nanopore fused with Exonuclease, and the nucleotides are first cleaved away by the exonuclease, than they flow through the pore, and the change in the current flowing through it is detected. Difference in the ion charge is then analysed and is used for basecalling.
2. Purely solid state nanopores - not much details on those - whether they are chopping the DNA up (as in 1, or threading it through)?

From our current error profile report it confirms 1, that the deletions are caused by the bases, that had "run away" from the pore.

If you read their website, you might notice it's actually (2) that they're marketing (I might as well link to it once again). The exonuclease is being done in something approximating a partnership with Illumina:

http://www.nanoporetech.com/technolo...ore-sequencing

• "Strand sequencing" is a technique that passes intact DNA polymers through a protein nanopore, sequencing in real time as the DNA translocates the pore. Oxford Nanopore intends to commercialise this technology independently within 2012.
• "Exonuclease sequencing" is a technique that passes individual nucleotides through a protein nanopore, aided by a processive exonuclease enzyme. Oxford Nanopore signed a commercialisation agreement with Illumina for this technology however commercialisation timelines have not been disclosed.

Of course, it's not quite "pure" solid state. They manufacture the membranes, and a solution of nanopores (created from bioreactors) is added in just before the run is started. The pure solid-state stuff may come in the future when things like graphene nanopores are worked out.

[Jeremy]

Originally Posted by gringer

As long as you have money and monkeys, you can sequence as many bananas as you want. With the "release date" 2000-pore GridION, a 20-node system should do a human genome in an hour (you'd need to split the sample into 20 for this), or a 5-node system in 4 hours:

http://www.nanoporetech.com/news/press-releases/view/39

I guess that's assuming 5-20x coverage for the human genome, depending on how you interpret "tens of gigabases", and also suggests a genome should be possible in one day on a single node with reasonable coverage.

With 20 MinIONs, it would take about 4 hours assuming similar processing speed.

Splitting the sample across 20 different cartridges to reduce the time of an automated process is all well and good. But it's not very practical. I am yet to find a lab where the budget is not the most limiting factor for what can be done. With a limited budget, the output of a single cartridge is very important.

[gringer]

If money is the limiting factor, rather than time, then carrying out sequencing using a single cartridge is possible, it will just take longer. A single MinION probably won't (yet) produce enough sequence in a 40h run, but two or three might be enough.

Then again, if money is the sole factor of consideration, neither Oxford Nanopore nor Pacific Biosciences will be an appropriate solution.

[Jeremy]

Originally Posted by gringer

If money is the limiting factor, rather than time, then carrying out sequencing using a single cartridge is possible, it will just take longer. A single MinION probably won't (yet) produce enough sequence in a 40h run, but two or three might be enough.

Then again, if money is the sole factor of consideration, neither Oxford Nanopore nor Pacific Biosciences will be an appropriate solution.

Nothing is ever the sole factor of consideration. And in my experience, money is ALWAYS a limiting factor, as usual it is a trade off. Pac Bio and hopefully Oxford Nanopore have a huge advantage in de novo assembly. Assemblies from just Illumina data can be insanely fragmented, the rubber tree genome for example consists of 1.2 million contigs. That is what could be done at the time for a reasonable cost and time expense, maybe when the gridION (hopefully at this point it is a case of when and not if) is released, more complete genomes will be obtainable for reasonable costs.

[gringer]

A coverage of 30X is not necessary to reduce 1.5M contigs down to something much more manageable -- a single MinION run, even at less than 1X coverage, will be more than a bit useful for cleaning up an assembly like that.

[gringer]

There is now a [pre]-publication out, authored by Nicholas J. Loman and
Aaron R. Quinlan, that introduces a toolkit for working with nanopore data:

http://bioinformatics.oxfordjournals...s.btu555.short

If you squint really hard at figure 1 (or zoom in lots), you can see the read length graph and a graph of cumulative output over time (see attached figure). It shows a median read length of about 8kbp, maximum read length of over 30kbp, with just under 300 Mbp generated in about 28 hours.

There's also another alternative R toolkit from Mick Watson's lab that has been released with a slightly different method of dealing with the data:

http://biorxiv.org/content/early/2014/07/29/007567

[Deheresedese]

Hi, It looks very cool, it seems to me that this is a very important thing. I am interested in information technology and found an article where they compare Kemp VLM Series vs F5 BIG-IP Load Balancers and it was very exciting for me. I am sure that it will be interesting to you.