MiSeq Mid-2012 Upgrade Rumors?

[ECO]

Little birds (some of which have orange feathers ) are telling me that the second surface of the flowcell will be imagable in the next month-ish (double throughput!), and that 2x250 is close behind that, with 2x400 by the end of the year.

Unknowns yet are the overall speed penalty that may happen (more chemistry, more imaging), and cost increases.

Anyone heard anything else?

[GW_OK]

How is this different from the long promised upgrade announce earlier this year?

From what I've heard from similarly plumed birds the Miseq upgrade will occur officially in July, but may not be practically on the street until October. It's being scheduled to coincide with PM's and service visits (since it's a "free" upgrade) so you might get it faster if something happens to the Miseq. You can also get in the queue earlier if you have a 2500 upgrade pending, since you've shelled out for that.

Concerning the 2500 upgrade you can also shell out 5k to get put on the priority list...

[ECO]

Different only in that it's a bit more clear how they plan to get from "2GB to 7GB" as they've quoted in marketing. Maybe I'm sadly a bit behind the marketing curve.

And I agree, it seems there are many different ways to buy yourself to the front of the upgrade (and 2500) queue.

[ians]

that 2x250 is close behind that, with 2x400 by the end of the year.

Assume good quality bases, i'd expect Roche to be very nervous with these lengths, especially when sloptigs are made.

Although these read sizes are awesome, there may be quite a lag in analysis software that can handle these increased read lengths algorithmically.

[dariober]

Although these read sizes are awesome, there may be quite a lag in analysis software that can handle these increased read lengths algorithmically.

As far as the alignment is concerned bowtie2 seems to handle reads of 1000s of bases. Once the alignment is done I guess downstream programs care less about the original read length?

[krobison]

Assume good quality bases, i'd expect Roche to be very nervous with these lengths, especially when sloptigs are made.

Although these read sizes are awesome, there may be quite a lag in analysis software that can handle these increased read lengths algorithmically.

I'm sure it will break some pipelines, but a number of tools shouldn't have a problem. bwasw handles long reads on the alignment side; for de novo assembly there are a number of tools (CABOG, Ray, MIRA) which shouldn't have trouble with these as they handle Sanger data.

[ians]

I mainly had assemblers in mind, where kmer length is king.

[ECO]

Just got this in email this morning, thought people would find it interesting. I am so excited for 2x250, let alone the ability to run 2x150 even faster:

As part of Illumina’s commitment to bring our customers the latest in innovation and technology, we are releasing both hardware and software enhancements to the existing MiSeq system to increase the performance of this already powerful platform. Enhancements include:
·Up to 3x increase in the number of clusters per run
·66% increase in read length (enables 2 x 250 reads)
·Faster run times vs. existing MiSeq run times
These enhancements will confer benefits in a number of areas including:
·Greater sequencing coverage for somatic mutation detection in gene panels
·Higher multiplexing capability for sequencing of bacterial genomes
·Profile several gene expression samples per run
·Improved de novo assembly of small genomes
Hardware upgrades will be performed by an Illumina field representative on a scheduled basis. Your upgrade window is currently scheduled for the month of August. Upgrades are being scheduled based primarily on original installation date and location of your lab. In an attempt to complete the upgrades in a timely manner for our customers, we will be rotating our upgrade team through various cities across North America. If this schedule is not convenient for your laboratory, please contact customer service for details on a revised appointment at:
[email protected]

New Software

Illumina will release a new revision of the MiSeq software suite to coincide with new MiSeq hardware. This revision of software along with new MiSeq v2 reagent kits will provide higher Q scores, decreased cycle times, and enable 2 x 250 reads. The software suite will include:
·MCS v.2.0
·MSR v.2.0
·Recipes v.2.0
·RTA v1.16
Prior to receiving the hardware upgrade, it is possible to upgrade your current MiSeq to the newest version of MiSeq software to gain a portion of the throughput benefit that will be fully enabled with the hardware upgrade. The new software suite will be available through Illumina’s cloud computing environment BaseSpace and Illumina’s support page. To help you prepare for this upgrade, Illumina will release software notes in the following weeks. Look for these on the MyIllumina Bulletin Board and the Illumina website.



The upgrade process takes an average of 1–2 days to complete and will include the MiSeq software suite update, installation of the new hardware parts, and system checks to verify that the instrument meets specifications. After these system checks are complete, the MiSeq can resume normal operation.
Additionally, your local ILMN field representative will be able to assist you in any questions you might have on the upgrade.

[ians]

awesome!
Let us know when you get a look at your first 2x250 run!

[snetmcom]

Did they increase the imaging speed at all? At the current rate, wouldn't 2x250 dual imaging put you well over 2 days?

[ymc]

An Illumina technician I interacted with told me there will be an upgrade that involves additional hardware to MiSeq very soon. But he said he didn't know how high the new throughput will be.

[ymc]

http://www.illumina.com/Documents/products/datasheets/datasheet_miseq.pdf

7gb run in 35hr?

[ymc]

Assume good quality bases, i'd expect Roche to be very nervous with these lengths, especially when sloptigs are made.

Although these read sizes are awesome, there may be quite a lag in analysis software that can handle these increased read lengths algorithmically.

I heard that many microbe researchers are using MiSeq instead of 454 now.

[MrGuy]

Well, only if you are looking at whole genome and never interested in amplicons.

[TonyBrooks]

Internally, Illumina have run 400bp with the same 250bp chemistry and instrument upgrades and hit 80% >Q30 at 400bp (which I think is totally acceptable when comparing to 454 data - especially considering read overlap increasing the score further).
I guess increasing read length can only go so far as there's a limit to what length insert can be successfully brPCR'd.
One thing I would like to see them do is to adjust assays (especially, TSCA) to make use of more and longer reads. Increase dual indexing to 384 and cover more sequence.