Bad MiSeq Reagent Kits

[mcnelson.phd]

Hey everyone,

Not sure how many people have heard about this, but apparently Illumina had a bad set of MiSeq reagent kits get shipped to customers. I found this out from talking with my FAS after we had a run fail this past weekend and he asked for the lot numbers for all Box 1's (the reagent cartridge) that we had. Turns out that the failed run, and 4 out of our 5 remaining 500 cycle kits have lot numbers that Illumina has identified as possibly having issues.

My FAS told me Illumina quality assurance was supposed to notify affected users yesterday, but I never received any emails about this so anyone thinking of doing a 500 cycle soon might want to hold off or at least check with their FAS to see if their kit is unaffected by the issue.

I'm not going to post lot numbers for the kits we had that were affected, but I will say that they were all purchased/received around the end of last year through the first two weeks of this year. 50 and 300 cycle kits, as far as I know, don't appear to be affected as my FAS didn't mention them (there was an implication that all 300 cycle kits are OK).

Hopefully this gets sorted out soon, or else Illumina could have a pretty bad PR problem on its hands.

[microgirl123]

We received notification late yesterday afternoon. We did have one in our freezer. I'm wondering if there's a way to figure out what the lot number was on past runs.

[mcnelson.phd]

We received notification late yesterday afternoon. We did have one in our freezer. I'm wondering if there's a way to figure out what the lot number was on past runs.

As far as past runs go, Illumina should be able to figure that out based on the reagent barcode.

As for the email, care to share any information about it? My FAS said we should have received one yesterday but I certainly never got it so now we're stuck with 4 kits that aren't useable and no idea when we'll be getting replacements.

[microgirl123]

Without sharing lot numbers, the email states that it's for 500 cycle kits received after November 27, 2012. They are working to resolve the issue and plan to be shipping again by Feb 4, 2013. Quoting from the letter, "The affected 500 cycle kits may have a shorter than expected shelf life and use of these kits can produce higher than expected pre-phasing values, and decreased data quality. "

To add to this, the letter also states "You are receiving this notice because our records indicate you received the affected product." Possibly, if Illumina doesn't have you listed as receiving one of these kits, you won't receive the notice?

[TonyBrooks]

I'd done three runs that looked poor. Initially I attributed it to overclustering, then we had a run with 881k/mm2 and the Q-scores seem to fall off a cliff after 200 cycles.
Our phasing/prephasing % estimates actually look OK, though.

[microgirl123]

And. . . I've just received notice that our replacement reagent cartridge should arrive by February 8.

[mcnelson.phd]

Well, I never got an email from Illumina QA, but our FAS has been pretty good at providing us with all of the information and getting us our replacements.

Apparently it was a contaminant in the incorporation mix that could lead to poor run quality, which certainly helps explain some previous runs. Just finished running a "good" kit this morning and the data quality is much improved, although I don't know how much of that is the kit versus the new MCS software that was recently put out as well.

Hopefully Illumina steps up the sampling for the QA checks so they don't have these issues in the future.

[Bucky]

I'd done three runs that looked poor. Initially I attributed it to overclustering, then we had a run with 881k/mm2 and the Q-scores seem to fall off a cliff after 200 cycles.
Our phasing/prephasing % estimates actually look OK, though.

Same with our runs! We had several kits from "bad" lots in our freezers and the ones we used had really poor quality, especially towards the end of the run (as exemplified by a "tower" at q10

[gwilkie]

Yep we had several 500 cycle runs all clustering between 800 - 950k/mm2 that failed. The quality degraded after about 170 cycles, reverse reads were only 60% >Q30. Some of the thumbnails were very blurry but phasing/prephasing wasn't terribly bad IIRC.

I still have six kits from the dodgy batch in the freezer! Currently waiting on PhiX results from a new 500 cycle kit to make sure everything really is OK with the instrument, as it also had some parts replaced recently. Illumina were starting to scratch their heads over what was wrong with our MiSeq then came up with this... hope its the answer.

The MiSeq is a great machine, when it works

[mcnelson.phd]

The MiSeq is a great machine, when it works

Truer statements have rarely been said on this forum...

[matth431]

We had one of the affected kits - no advisory (noticed the bulletin on MyIllumina when going to order replacement). Didn't even get as far as sequencing - read 1 focus failed and the MiSeq threw up an orange light. Unfortunately this was our first test of a v2 kit since we had the upgrade so we were initially uncertain if it was something we'd done wrong (loading, etc).

[MLog]

We had one of the affected kits - no advisory (noticed the bulletin on MyIllumina when going to order replacement). Didn't even get as far as sequencing - read 1 focus failed and the MiSeq threw up an orange light. Unfortunately this was our first test of a v2 kit since we had the upgrade so we were initially uncertain if it was something we'd done wrong (loading, etc).

We tried to use the kit from the bad lot and got almost the same result today... The sequencing seemed to progress well until 174 cycle, but after this MCS reported an error on autofocusing ("best focus is too near the edge of range") and stopped the run.

[Vinz]

It is hard to say if all lots of the bad lots are the same. Maybe some are a lot worse than others. All I can say is that we had quite a couple of different lots from the bad batch and the effects were mild. I would bet that your focusing issues are not related to the bad lot. These are most likely instrument issues (matth might be handling).

[tymek666]

The same here, two reagent kits 500v2 shipped in December went to trash. kits shipped in January semms to be fine.