Why choose the PGM and not MiSeq?

[docsascha]

Dear all,

Hello my name is Alexander.
I am knew to NGS, the main interest of my Lab is bacterial typing.

Currently I have two obvious choices Ion Torrent or Myseq. Out of reading papers and posts I would say great a Myseq it is, but what about the PGM?

Is it worth walking its road?
Interestingly the running costs of the PGM seem somewhat higher ( they are even selling this as an advantage over the Myseq) than then Myseq ones.
The hands on time is higher also there is the Quality issue?

Is there a reliable argument for this System?
Does it still have a future, expect beeing a prop in an "Elementary" episode
Does anyone have used the new chemistry?

Alexander

[AllSeq]

Dear all,

Hello my name is Alexander.
I am knew to NGS, the main interest of my Lab is bacterial typing.

Currently I have two obvious choices Ion Torrent or Myseq. Out of reading papers and posts I would say great a Myseq it is, but what about the PGM?

Is it worth walking its road?
Interestingly the running costs of the PGM seem somewhat higher ( they are even selling this as an advantage over the Myseq) than then Myseq ones.
The hands on time is higher also there is the Quality issue?

Is there a reliable argument for this System?
Does it still have a future, expect beeing a prop in an "Elementary" episode
Does anyone have used the new chemistry?

Alexander

Alexander,

The main advantage that the Ion Torrent platforms have is speed. Runs on the PGM can be as little as two hours. They used to have an advantage with read length, but MiSeq has really caught up (and even surpassed with the latest paired end kits). The other advantage they had was the promise of future improvements. Unfortunately, the promises are starting to ring a little hollow. The PGM has maxed out with the 318 chip and ~2Gb of output per run (compared with ~15Gb output from the MiSeq). The Ion Proton continued the tradition of improvement with the PI chip which generates ~10Gb per run. But the PII chip, which was due out late 2012, has been pushed back multiple times, with the latest launch target set to mid-2014.

In terms of data quality, the have different error models: Ion Torrent produces indel errors (due to their homopolymer issue) while the MiSeq tends to produce base substitution errors. MiSeq is generally seen as producing higher quality data.

I would say that both are competing heavily in the desktop sequencer space, but the delays on the PII chip is allowing the MiSeq to pull ahead.

Shawn
AllSeq, Inc.

[ymc]

Alexander,
But the PII chip, which was due out late 2012, has been pushed back multiple times, with the latest launch target set to mid-2014.

Wow. Another postponement.

What happened? Shouldn't Protons improve throughput as fast as CPUs? What is the technical obstacles that prevent their advance? Or is it human factor?

[AllSeq]

Wow. Another postponement.

What happened? Shouldn't Protons improve throughput as fast as CPUs? What is the technical obstacles that prevent their advance? Or is it human factor?

I wasn't there to attend in person, but the Ion Torrent folks seemed to spend a lot of time at ASHG explaining how hard it is to make these higher density chips (building the electronics to capture all the data that's being pumped out, etc.)

Basically, they've reached the limit where shrinking things down is getting harder and harder. Just building the chip itself probably isn't that hard, but consistently getting good, high quality data out of it is probably another matter. Maybe someone who attended Ion Torrent workshop at ASHG can chime in with the details.

Shawn

[thomasblomquist]

Just to add some hear-say to the pile. But the delay in increased density may be a function of bleed-over from adjacent beads pumping out H+ ions during sequencing. The signal-noise ratio is getting lower with the higher density chips. May require significant redesigns to overcome this issue of H+ ion diffusion. This is a rumor at this point, heard from someone, this onetime at a conference, who talked to a tech, that used to know this one person... ... ... you get the point.

It seemed like a believable explanation though. Not sure if it's true.

-Tom Blomquist

[MrGuy]

Here's their marketing guy talking about it:

http://www.yuzuki.org/ion-chef-pii-chip-isothermal-template-prep-and-a-new-hi-q-enzyme-at-ion-world-2013/

Synopsis for those who don't want to click:
Chef: ships now (anyone seen one?)
PII: ships March 2014 (he explains some stuff about it)
Avalanche: didn't see a date, but pretty sure it is either out or early access
Hi-Q: apparently in "tech access". Allegedly gives Q30+ over 400+ bp, thus allegedly better than Illumina. Anyone have some data to show?

As for the OP's question:
PGM, Miseq, or Proton PI will work for what you are doing. If Chef, Avalanche and HiQ work as advertized, then the whole picture gets foggy as to "who is best". If they don't, Miseq is easier from a labtech perspective, but turn around time will suffer. Read length, no big deal, since you are doing whole genome on a reference. If you do any kind of subset of a whole genome, then a PGM will be enough.

To clarify more about what you are doing, you'll have to give an idea of how many samples, response time, expectations within data reporting, and sample types you need to process. The sequencer really isn't important. You might even want to wait for the Qiagen box or some other gadget coming to the market.

Don't forget that this forum is primarily Illumina users, so you get illumina influenced repsonses. Ion community has Ion users. Good luck!

[IonTorrent]

Regarding the latest updates on a few key product innovation areas (Ion Chef, Isothermal Amplification, Hi-Q sequencing chemistry, Ion PII Chip), please take a look at these two Ion World presentations that were given by our VPs of R&D

Alan Williams discussing the latest on Ion Chef, Hi-Q sequencing chemistry:
https://www.youtube.com/watch?v=Ygjm2mZq7LM

Mike McKenna discussing the latest on Isothermal amplification, PII:
https://www.youtube.com/watch?v=K7RgSe14-Uw

You don't have to rely on hearsay. We always post videos of key talks/presentations/announcements on our YouTube channel. There are also many Ion customer-led discussions and Ion-specific content including protocols/manuals at the Ion Community in addition to the insightful discussions that take place here on SeqAnswers.

YT Channel:
https://www.youtube.com/user/iontorrent

IC (you don't need to be a paying customer to enter, just register):
http://ioncommunity.lifetechnologies.com/welcome