Maximum number of cycles from 300-cycle MiSeq kit...

[ECO]

Just finished this run this morning...it was an attempt at 151x251 to determine just how many cycles are contained in the 300-cycle reagent cartridge.



After a successful 308 cycle run (2x151,1x6index), there is quite a bit of reagents left over, mainly the incorporation, scan, and cleave mixes which is all that's needed to sequence and and image.

The answer seems pretty clearly that at about cycle 360-370 (haven't looked at images yet), things fall off a cliff. So 2x175 would be pushing it but probably possible.

Next time we have a run that fails early, I'm going to overfill a new reagent cartridge to see what's possible...2x200 seems really easy. 2x250 probably stretching it but we'll see!

[hedas2303]

Any quality scores available?

Very cool!
I wonder if you had any data on Q-scores during the run? From the Broad presentation (300bp) they ended at app 65% Q30, it would be interesting to see the drop off in your run too.

[krobison]

Any reason to fear this could damage the instrument?

[Tanya12]

miseq Q-scores

Very cool!
I wonder if you had any data on Q-scores during the run? From the Broad presentation (300bp) they ended at app 65% Q30, it would be interesting to see the drop off in your run too.

I also really would like to know about the Q-score at cycle150 or 200.

Thanks advance.
Cheers!
Tanya

[ECO]

Any reason to fear this could damage the instrument?

No ill effects noticed yet.

I wouldn't let it run dry and sit overnight...as (from reading the recipe xml files) it flushes the template position with incorporation buffer several times, likely to prevent template from drying on the sipper.

I also really would like to know about the Q-score at cycle150 or 200.

I'll try to get the fastqc report up asap.

[ians]

Have you had a chance to get a fastqc profile?

[ECO]

Yes! I just have to remember to post it! Setting reminder now!

[ECO]

Finally dug it out...enjoy:

http://seqanswers.com/forums/images/content/fastqc/111102_M00182_0006_AMS0002210-00300/s_G1_L001_R1_fastqc/fastqc_report.html

[ECO]

By the way, this run was from 12x libraries of 768 amplicons each (250bp long)...

[gringer]

The answer seems pretty clearly that at about cycle 360-370 (haven't looked at images yet), things fall off a cliff. So 2x175 would be pushing it but probably possible.

About 372, perhaps? That's enough for 6 50bp runs with two barcode reads each (i.e. the newer dual-indexed Nextera barcodes).

[ECO]

About 372, perhaps? That's enough for 6 50bp runs with two barcode reads each (i.e. the newer dual-indexed Nextera barcodes).

That would require hacking their RFID system...in the current form you can only use a cartridge once.

[Nitrogen-DNE-sulfer]

Any had luck hacking the RFID system?
I dont see any mention of it in the recipe XML files

[ECO]

Easiest way is just to break out the RFID chip (under the square "Illumina" pad on the top of the cartridge)...and when the scan fails, get a bypass code from icom.illumina.com. They are good for a week...I'm sure if you made a huge practice of this ILMN would crack down but it's easy enough for playing around.

[yaximik]

I wonder if it would be possible to reuse the flow cell, not to run completely new sequencing, but to repeat sequencing with clusters already there. Sequencing errors are generated at random, so repeated sequencing will not generate the same dataset. Or a flow cell with the good cluster density of a good library could be used to obtain longer reads. Removing just synthesized strand should not be a problem as it is done anyway in paired end runs. I am not sure about paired end chemistry though, could the cell after a paired end run sequenced in the reverse order? Certainly the current RFID system disallows this. Any thoughts?