The problem:
If a technology described as next-generation sequencing is being published in multiple papers per month, then surely it is no longer “next generation” but “current generation”.
What horrors will be spawned by future advances in sequencing technology?
- Next-next-generation sequencing?
- Yet-another-generation DNA sequencing?
- Sequencing: the wrath of Sanger?
The solution:
In order to avoid an intractably messy literature, how about assigning clearly defined generation numbers, as the cell phone industry have done?
A bet is riding on “sequencing any-digit-point-any-other-digit” over here. But I think I’ve seen next-next somewhere already. There was also “ultra-deep” for a while.
Oh, duh, I should follow links better. Sorry about that.
“next-gen” sequencing is already “last-gen” sequencing. It’s now possible to do single-molecule, real-time sequencing.
It’s definitely time to adopt a version numbering system!
Don’t forget… your talking about re-sequencing, “next-gen” relies on having sequence to compare against in order to assemble anything meaningful
Ooooooh. Shiny.
(Last 2 comments crossed – it’s the single molecule stuff that’s shiny).
Yes, indeed, on the re-sequencing thing. Any numbering scheme would hopefully reflect that!
Heather, it amuses me that we both linked to the same article!
Beware talking of generations. Anyone remember fifth generation computing ?
A lost generation?
It just occured to me (via Heather’s 1st comment) that there are generations of generation-numbering conventions. From I, II, III to 1G, 2G, 3G, to 1.0, 1.1, 2.0.
How about using genetic nomenclature (G0, F1 etc)? That implies one spawns the next though….but also remember that Trek-world avoided the issue by forgetting they ever TNG’ed (Voyager?!!).
I like it!
I never watched Voyager, or DS9, or the new one, or even much of the original. It’s all about Jean-Luc!
If it’s anything like the Star Trek franchise, next-generation sequencing will give way to a darker, earlier age, where a youthful Fred Sanger attempts to rid the universe from radioisotopes, and negotiate an armistice with cling-on fragments of DNA.
Hey, I just got a press ticket to see the new Star Trek at the London IMAX. How pleased am I?
cling-on fragments of DNAGroan
The biggest difficulty arising from next-gen sequencing is, of course, how to manage Data.
He’s Lieutenant Commander of the Starship Venterprise, right?
Nice Troi, but that was Worf-ul.
Argh.
I myself have used the term “next next-generation” to refer to the nanopore kind of stuff that Darren linked to. Which, by the way, doesn’t yet work for anything useful, but might by the time PacBio releases their instrument next year (presuming they hit that scheduled milestone like they’ve been saying).
I also hear “third-generation” from time to time. I like the catch-all term, “massively parallel sequencing”.
Oh, and Darren – you mean you’re not interested in de novo assembly using short reads, where you get 8 billion contigs of N50 length about 100 bases out of your assembly? What’s wrong with you? 😉
P.S. I Khaaaaaaaaaaaaan’t take it any more. I’m gorn home, Chek’n out ov here.
Using descriptives like massively parallel work well.
Yes I agree, Data is a Crusher to manage when you move away from bare Bones sequencing. I’m getting a call over the com from JT Kirk. He wants a word with you regarding It’s all about Jean-Luc!.
Ah, if only my brain could process information in a massively parallel way!
On the subject of data… apparently we will soon overloaded the cooling capacity of the room in which our server/cluster is located (ie the one that stores/processes all the data from our sequencing and high-throughout screening.
Darren – it’s a problem, isn’t it? 😉
You could have a pool where people could bet on whether the disks will be full before the cooling fails, or not… 😛
On another note, I visited a cluster recently (IBM Power6 stuff as I recall) where the cabinets had chilled water running into them, over manifolds on top of the processors, and through a radiator-like affair in the back door of the cabinet. Crazy stuff. Nuts.
I’ll post a photo if I remember, when I’m back somewhere I can get at my Flickr account.
On the face of it, that doesn’t sound like the most inspired piece of engineering!?
apparently we will soon overloaded the cooling capacity of the room in which our server/cluster is locatedIs this why they’re making us do all the fire safety training?
I hope you’re poised by the fire alarm…
As the boss pointed out, Moore’s Law doesn’t apply to air conditioning systems!
Darren – neither does it apply to NGS throughputs (or density of features on microarrays, come to think of it).
The water jacketing was pretty spiffy – well contained and very nicely engineered. It sounds wrong though – running water through a lot of very expensive, high-voltage computing equipment. Eeeekz!
Ah, here we go – Phase I of SciNet. 3,328 IBM Power6 CPUs and something over 13 Terabytes of RAM. The disk cluster is in a separate set of racks (this photo is all diskless CPU):
All conveniently located in an anonymous industrial building, about 15 minutes from my home.
NGS porn alert…