Dream Machine

[Biotech Rockstar]

What specifications/performance would you all like to see in your perfect (be realistic) sequencer?

I appreciate your time in answering this!

Disclaimer: I'm from a company developing a dream machine (I hope)

[krobison]

I suppose a disposable instrument the size of an smart phone which generates 99.9999% accurate megabase reads covering the entire human genome for $100 wouldn't qualify as reasonable?

As I've written about a number of times (most recently this week, but also in August of last year and the prior fall and certainly other places), I see the world as having multiple sequencing markets. A truly dream machine would inexpensively address all of them (as my proposal above), but valuable new entries would be those that improve on an important dimension (cost, read length, accuracy, number of reads, turn time) for one of them.

Some key markets
1) Generating a lot of high-quality data in up to a few weeks -- e.g. competing with GAII*/HiSeq and SOLiD. Beating them on either upfront cost or per run cost or cost per base are the best routes I would see

2) Generating very long reads for genome finishing, haplotyping, de novo transcriptome. Pretty open market, with Sanger still holding on by its fingernails, 454 grabbing a bunch & PacBio poised to make inroads (if it can really work). The StarLight technology would be another possible player here, though Life Tech has not really put a road map out for consumption. OptGen is a dark horse in this area.

3) Very fast turn-around for mutation detection / clinical apps. Ion Torrent is making a play here; PacBio & 454 are in the space & to some degree even Illumina & SOLiD. Ideally, very low cost-per-sample or cost-per-run.

4) Very high numbers of "long enough, accurate enough" tags for RNA-profiling, CHiP Seq, etc. Cost-per-sample is a major issue here; it's brutal to try to plan a complex multi-arm time series experiment with replicates (as in, "you've got to be kidding").

5) Little-to-no sample prep. Perhaps more of a feature than a market segment, but worth noting. In similar vein, working with minuscule amounts of clinical (i.e. beat-up) DNA.

[Dr. Hybe]

Dream Machine

I enjoyed your previous post about the $100 disposable dream machine. Ok! so here is my dream machine...I call it Variable Ray Holographic Sequencing (VRHS) It works like this...somehow get the sample onto a special compact disk like chip device that when rotated causes the DNA to migrate into specific grooved cannels. (Think ultrafast roulette wheel) When read by a laser of variable wavelength at different RPM's correlated to sample type, the computer registers this as an a,t, g, or c. Combine this with cutting edge holographic data storage techniques and you are looking at petabase capacity in a few hours. Scale this down to a Sony discman like device with wifi and cloud computing and voila - a sequencing dream machine. I have some more ideas about sample prep, bit will have to save those for a future post. Thanks - Dr. Hybridization

[qnc]

Biotech Rockstar if your company was to develop a dream machine and is currently seeking a wish list then first things first... In my opinion there is no point holding back.

Possibly you are not saying your going to develop the ideas here you just want to see what people are thinking.

Unrealistic is hard to define when it comes to design or desire as what was "unrealistic" years ago like traveling faster than the speed of sound for instance or even flying itself is now common place to us.

It all depends on the right idea to get the tech working the way we want it... But if we see it as limiting we risk never getting it right as we may never try it as we think its unrealistic.

What would have happened if the Wright brothers had though flying was unrealistic would they have tried?



So can your company make me...

1) Something that can do a full human sequence (including large repeat regions like LINEs and SINEs and homo-nucleotide stretches and anything else current sequencers can't do) from hair or mouth scrape.

2) The sequence in an hour or less (end to end including the internalised sample prep.. which could be tweaked for difficult cases, still like experimentation... don't like just following the costly propriety instructions...)

3) The machine given for free possibly, the collector/scaper itself is the machine (I do like the disposable option mentioned by krobison (Biodegradable/recyclable off course) and a genome sequenced £0.01? (6 billion on the planet so still £6 million for your company... not huge but this is for humankind and all and its my wish list not your shareholders one Maybe you can pick up some advertising money logos or products from other companies etc.. I don't mind that as long as I don't have to pay too much to get my results.

4) All results automatically integrated into a anonymised but disease tracking planet wide human SNP database. That way we can properly tackle population diseases and look at personalised medicine for all.

Just my two pennies.

Let me know when my free machine is ready and off course some free consumables to get started

And its nice that your asking what we would like rather than just saying hey buy this at 1000x mark up

[Biotech Rockstar]

1. Fluorescent sequencing
2. Semiconductor sequencing

Interesting schematics, which technologies are they from specifically? My companies semi-con sequencing schematic looks nothing like the one in the diagram.

[Biotech Rockstar]

One dream machine coming right up!

So can your company make me...

1) Something that can do a full human sequence (including large repeat regions like LINEs and SINEs and homo-nucleotide stretches and anything else current sequencers can't do) from hair or mouth scrape.

2) The sequence in an hour or less (end to end including the internalised sample prep.. which could be tweaked for difficult cases, still like experimentation... don't like just following the costly propriety instructions...)

3) The machine given for free possibly, the collector/scaper itself is the machine (I do like the disposable option mentioned by krobison (Biodegradable/recyclable off course) and a genome sequenced £0.01? (6 billion on the planet so still £6 million for your company... not huge but this is for humankind and all and its my wish list not your shareholders one Maybe you can pick up some advertising money logos or products from other companies etc.. I don't mind that as long as I don't have to pay too much to get my results.

4) All results automatically integrated into a anonymised but disease tracking planet wide human SNP database. That way we can properly tackle population diseases and look at personalised medicine for all.

Just my two pennies.

Let me know when my free machine is ready and off course some free consumables to get started

And its nice that your asking what we would like rather than just saying hey buy this at 1000x mark up

1. YES, these applications are the very reason I invented it

2. Wow, not asking for much are we.... but yes, end to end, telomere to telomere, T2T Sequencing

3. Hey, I gotta eat!

4. Well someone else can develop this bit, I've done my part... T2T the nanoPOOR killer!


"Just your two Penny's" Your two pennies? So by your cost requests, that's two genomes worth ;-)

You're the market, why wouldn't I ask?!?!

[Chipper]

Wow, killing something before it exists is truly impressive! When will you show how your nanowiresequencing (?) works and why we should believe it is BestGen?