JUST AWESOME!!!!!!Gr888 Job

I still see a lot of PGM's being used as doorstops, still in the boxes...not being used for sequencing.

Cannot say the same for MiSeq....all on the bench, being used, generating data. This comes from someone that is in labs on daily basis. No affiliation with either company.

JR was actually born rich...

Lifetech now has workflow on SOLiD that is direct library onto chip with in situ isothermal monoclonal amp. I think it's called Wildfire or something like that. Maybe they are thinking about applying that to ION as well

Poor Q30s for low diversity libraries is a known issue. Supposedly a software fix to RTA is coming soon that should resolve many of these issues, but no idea when that's happening. Overall, we are happy with our MiSeq. No comment on PGM other than that our lab never even considered purchasing one.

I just heard from a friend's lab. They bought Miseq one year ago.
There are still some problems in using MiSeq for 2x250 bp sequencing kit. Their software is sometimes problem, which cannot produce data. Sometimes, only 50% of reads can pass Q30. The company service is good, and they come several times to my friend's lab.
Did somebody have experience in using MiSeq? I wish to find someone to share your experience, since our lab wish to consider a sequencer, either Miseq or Ion Torrent PGM.

I just heard from a friend's lab. They bought Miseq one year ago.
There are still some problems in using MiSeq for 2x250 bp sequencing kit. Their software is sometimes problem, which cannot produce data. Sometimes, only 50% of reads can pass Q30. The company sale service is good, and they come several times to my friend's lab.
Did somebody have experience in using MiSeq? I wish to find someone to share your experience, since our lab wish to consider a sequencer, either Miseq or Ion Torrent PGM.

See response #5 for real life example of Miseq vs PGM

http://seqanswers.com/forums/showthread.php?t=18686

What is it you do for a living? Sounds like you are a real expert especially since you use the NY times as a scientific reference.

who maufacture Ion torrent chip?

All these systems are going to stall if not already. No one even bothered with data deluge... Scientists are the worst managers and now they will caught officially in the third genomic bubble, well done boys, not only are we spewing useless shit from these machines but we have no where to go for answers and no software to help. Don't expect LT or ILMN to get into the bioinfo part of the deal.

http://www.nytimes.com/2011/12/01/business/dna-sequencing-caught-in-delu...

Comparing sales numbers isn't yet fair, the PGM was out prior to MiSeqs even being sold. Illumina is smart, why put out something early that isn't ready, then spend all of your time and money trying to fix placed instruments. How many FAS did LT have to send out for simple prep to all of these newbie users? emPCR with the desktop seq market is a nightmare, even making a standard library for either box is hard for most of these people.

Illumina was smart with Nextera. Simple prep, pipet into machine and hit go. Any high school kid could do it in a science class. emPCR... HA!

Ask any customer what they hate about 454 and SOLiD. Read any forum on NGS, the main problems are emPCR, not library creation.

We can talk sales numbers next spring. I haven't seen the PGM jumping spec every 6 months as promised. Aren't we due for 100M high quality 500 bp reads by now?

Sample prep is extremely critical. Emulsion PCR is definitely one of the key reasons why SOLiD struggled.

Currently, from many customers' view, PGM is more promising than Miseq. That's why PGM was sold much more than Miseq, although debates exits, that's the reality.

Miseq's dilemma is, they lost the opportunity to be the first one, when the market DOES need a small, low-throughput, and simple NGS. And now ,their specifications are not very shining. Sample prep is not the key factor to win. In fact, although LIFE has increditble numerous internal problems, it's very hard for Miseq to turn over.

JR is laughing all the way to the bank. He can walk away now a rich man, too bad LT sucks at due diligence. Greg Lucier can get all emulsional, promise the world. The one metric to track is not initial sales numbers, but return customers and reagent sales. How many of these PGMs will be turned off and put into a closet once people see how hard it is to get sequence ready beads? Who cares if the machine can run in 15 minutes, the prep process really sucks.

Don't worry, JR will fix it!

The PGM is useless in comparison to the Miseq, it loses in EVERY comparison (error, throughput, ease of instrument use, ease of workflow). And with the recent press release from Broad of 300bp reads on the Miseq, the PGM loses in the last place it had a chance.

Illumina seriously sandbagged on their "3M" read spec. 7-9M pairs of reads (2.5GB of mapped data...) with 15 minutes hands on time after the library, compared to 3-4 hours of misery with the leaking Onetouch (yes, the PCR cartridges leak) and the almost non-functional One-touch enrichment "robot". Need to rerun a chip? New emulsion. Run out of argon during a run, new emulsion. Miss your window on the titration? New emulsion. Any of these happens on the Miseq, just thaw another cartridge in 30 minutes and use the same library again to restart the run. The MiSeq happily deals with cluster densities ~2x higher than spec (~1200K).

My personal opinion, you'll be sorry if you buy/lease/have-to-use a PGM. LIFE rushed out an early alpha instrument, and soon as all Ion folk get their payout, it will be even less well supported. Combine that with rumors that the current instrument hardware can't handle the sensor bandwidth of the 318 chip, and I consider the platform as dead as SOLiD.

PGM2 you say? Same workflow. Even with the same throughput (10M loaded and functional sensors?) the workflow is the achilles heel.

Life started to sell IT as free offer along with reagtns and its shows they are very much bit worried lot of Miseq performance.

My guess is that they will use hitachi for pgm2. Don't forget hitachi was also vested in this, not just a basic assembler but GL and his invitrogen idiots really screwed up, if a normal employee stooped only at 5% of that he would be abused, reduced and fired in no time. This is why everyone hates him.

5500xl is unfortunately already dead. It was dead from the start. One of the worst LT project. Worst is the fact that hitachi likely has a minimum pipeline order that will never see the day. GL should be fired on this alone, massive judgement error.

What to do... Pay Hitachi to build the machines, hoping that the Ignite Institute order finally comes through (last resort, give a free 5500xl with every PGM purchase and hope for consumable sales), or buy your way out of the Hitachi contract...

Ion torrent 2 looks cheezy but if its up to specs, then should be interesting. Moore's law on the same PGM was really not possible.

"You will soon have another system to compare here."

Early 2012. Be patient, my friend.

5500xl is unfortunately already dead. It was dead from the start. One of the worst LT project. Worst is the fact that hitachi likely has a minimum pipeline order that will never see the day. GL should be fired on this alone, massive judgement error.

"5500xl"

Really? Why, has one been spotted in the wild?

I think you mean 5500lol

5500xl

What's the new system ?

You will soon have another system to compare here.

LIFE attributed $13M of revenue in Q2 to Ion Torrent. Say its 90/10 systems to consumables and everyone paid $50k for a system that's only 234 units. Considering that 60 units were announced just prior to MiSeq announcement, hard to see how 1000 are in the field especially since installs lag orders.

I noticed that this blog site:

http://omicsomics.blogspot.com/

has many informed discussions of your questions.

My bad... I did forget the IT instrument cost. I believe the ~$100K figure for the all-in cost is about right. The sequencer itself is ~$50K. In terms of read length, the 100 base figure is the current spec. They have released some internal demonstrations out in the 200-250 base range as you mentioned.

Thanks for the infos. Do you have insights into the cost of consumables? total life cycle cost? Although semiconductor technology is definitely scaleable, would the microfluidic scaleability keep up? Accuracy of the sequencing? I think LIFE vision is to have these PGMs in every doctor/clinic office. I just don't see these will be prevalent as PC in everyhome where upgrades are done frequently to keep pace with technology. Sorry for the many questions, Thanks in Advance.

Already > than 200 bases for IT. I also suspect the costs are changing (lower) for IT as I write this, since they are coming out with new chips which offer higher densities.

I understand there are well over 1000 IT's already in the field and many are being used. The one complaint I've read about has been the tech support for the installation and training of the IT...it is slow and people are getting pissed about that. MiSeq has just begun delivering the first instruments to a select number of clients...kind of pre-full release clients. Not sure what that really means.

In the end, IT has a big head start. Not saying MiSeq can't/won't catch up. But if it suffers any major hiccups it could be a pretty large gap to close.

You didn't mention the cost of the IT. I believe it is somewhere near $100K after you add the cost of the instrument and the accessory items...but don't quote me on that...I don't work for LT...it's just a number I saw posted on a different message board.

Design:
MiSeq: Fluorescence-based reversible terminator nucleotides for sequencing-by-synthesis, requires optical detection of four fluorophores that are incorporated into DNA, requires fluidics
IT: Unlabeled nucleotides like pyrosequencing, requires electrical detection of the pH change that occurs when nucleotides are incorporated, requires fluidics

Useability/Simplicity/Workflow:
MiSeq: After library preparation, this is a set-it-and-forget-it system (clonal amplification, paired end module, sequencing, base-calling all in one instrument)
IT: After library preparation, one must clonally amplify with emulsion PCR prior to sequencing

Performance:
MiSeq: 2x150 bases in 27h, ~1Gb, ~3 Mreads, >75% at Q30 (99.9% accuracy)
IT: 100 bases in 2h, ~100Mb, ~1 Mreads, ~Q20 (99% accuracy)
(supposedly headed towards 200-400 bases, 1Gb in the future)

Cost:
MiSeq: 1Gb $750, instrument is $125K
IT: 0.1Gb >$500