MiSeq v3 2x300 run, Low PF and Q30

SarahAurora · 03-26-2018, 05:27 AM

I recently ran a ~200-400bp library prepped with NexteraXT on the MiSeq using reagents kit v3 for a 2x300 run, paired end, dual index reads: 10pM pooled library with 15% 10pM phiX.

I got a high cluster density of 1185 k/mm2 with only 27% clusters PF and >=Q30 14%.

Does anyone have any ideas for troubleshooting? Could this come from old kits or bubbles on the flowcell, would this even be visible? I did a qPCR (KAPA library quantification) before dilution all samples to 4nM. The qPCR worked fine, so there must be an error afterwards or with the kit.
Could it be that there was a DNase destroying the amplicons?

I'm curious about your ideas and answers.

Thanks in advance
Sarah

GenoMax · 03-26-2018, 05:41 AM

Have you checked the actual insert size of your library? It is possible that you have inserts significantly shorter than you expect them to be. That can explain the precipitous drop in Q score on R2.

You can check post #2 in this thread to see how you can calculate that information.

This run can also be borderline overloaded. That cluster density number is not very accurate once you get next the upper spec limit. Have you had Illumina tech support look at this run?

SarahAurora · 03-26-2018, 05:52 AM

Thank you for your reply. We have done the sequencing of this amplicon multiplex for many times and know the actual sizes. But are the moment the demultiplexing is problematic because it's too less reads.
The run could be overloaded but we have also done the sequencing with equal samples and even more input. It was very overloaded but we did get a lot of clusters passing filter (but a bad Q30). Now we have a very low clusters passing filter. So I think that overloading could not be the main issue.
No, we have not (yet) been in touch with tech support. I'm still hoping that the clues here could lead to an explanation.

GenoMax · 03-26-2018, 06:03 AM

If you have done similar sequencing successfully in the past then I suggest that you contact Illumina tech support and have them look at the run. This may turn out to be a hardware related problem on the sequencer.

thermophile · 03-26-2018, 06:26 AM

have you looked at the thumbnails? I'd guess your cluster density is higher than reported.

SarahAurora · 03-26-2018, 10:25 AM

Thumbnails of the tiles look ok, or let's say, they don't look as bad as the very overclustered run before. But they don' look balanced.

GW_OK · 03-26-2018, 10:48 AM

Can you post the %base graph?

SarahAurora · 03-26-2018, 10:59 AM

This one? I am not sure which one you want

SarahAurora · 03-26-2018, 11:07 AM

It's strange that we can read the first bases of R1, but no R4 and just a few Indices (R2+R3). Looks like the sequencing could not get read out like something intereferred with the optics.

I am very concerned about the run and my mind starts to melt, so please excuse this question: If I would have had trouble with the Index-PCR, then the adapters never would have gotten fixed on the flowcell, right? I want to exclude errors while preparing the run.

GW_OK · 03-26-2018, 11:09 AM

That's intensity. I was talking about the graph that is from % Base, which is selectable in the menu from the gray button under Intensity.

Still, though, in looking at this graph it appears you don't have a short insert but you do have sort've an odd 3-steps down to 0 intensity.

It looks to me like a hardware issue.

SarahAurora · 03-26-2018, 11:48 AM

Ok, thank you for your opinion. The longer I check all circumstances, the more it gets clear that the problem was not the DNA. Because at first the bases looked fine.

GW_OK · 03-26-2018, 11:58 AM

What's troubling to me is that you don't have any intensity in read 2, which should always show if you have at least a read 1, regardless of index.

Have you run any of the onboard diagnostic tests?

GenoMax · 03-26-2018, 12:00 PM

@Sarah: Do you not have a maintenance contract on this MiSeq with Illumina? If you do then you should go ahead and let tech support have a look at this run.

SarahAurora · 03-26-2018, 12:15 PM

Quote:

Originally Posted by GW_OK

What's troubling to me is that you don't have any intensity in read 2, which should always show if you have at least a read 1, regardless of index.

Have you run any of the onboard diagnostic tests?

Our bioinformatician told me we have 300 bp read 1 and read 2 but he main part is not Q30 or even Q20. So probably the DNA-bases part worked finde but the optics couldn't recognize anything...

No I did not know that those tests even exist...

SarahAurora · 03-26-2018, 12:20 PM

Quote:

Originally Posted by GenoMax

@Sarah: Do you not have a maintenance contract on this MiSeq with Illumina? If you do then you should go ahead and let tech support have a look at this run.

It's not our MiSeq, we just use it from our cooperation partner. I'll check if they have a maintenance contract with Illumina. Nevertheless, I wrote tech support an email with all details.

GenoMax · 03-26-2018, 12:21 PM

Quote:

No I did not know that those tests even exist...

I strongly suggest not running those tests, if you are not familiar with them (and before talking with Illumina tech support).

GenoMax · 03-26-2018, 12:23 PM

Quote:

Originally Posted by SarahAurora

It's not our MiSeq, we just use it from our cooperation partner. I'll check if they have a maintenance contract with Illumina. Nevertheless, I wrote tech support an email with all details.

Illumina tech support will entertain questions from end-users (so you have done the right thing) but you would need your partner to be involved if they need remote access to the sequencer to check this run/require the partner to run the diagnostic tests.

GW_OK · 03-26-2018, 12:25 PM

Onboard diagnostics are in the "System Check" icon under "Manage Instrument" from the main screen on the Miseq Control Software. The optics checks are fairly fast to run, the fluidics will take some time and babysitting (raising and lowering sippers as prompted).

Has your partner had any successful runs since your failed run?

Edit: Asking me to run these tests is usually what Tech Support has me do first when I call them with an issue. Enough so that I'll sometimes run them prior to calling...

SarahAurora · 03-26-2018, 12:25 PM

Ok that's all very good advice, thanks a lot! I hope tech support can help me!

GenoMax · 03-26-2018, 12:29 PM

If this failure is due to hardware/reagent problem (and the partner has a maintenance contract with Illumina) then Illumina generally will replace the reagents at no cost, so you should be able to re-run your samples for free (or at least no charge for the reagents).

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
low Q30 in index, miseq v3 600	cnano	Illumina/Solexa	3	06-05-2019 01:43 PM
2x300 MiSeq read 2 failures due to reagent?	ksawatzki	Illumina/Solexa	68	01-19-2017 08:40 AM
MiSeq run with low Q30 and high cluster PF	haroldnunez	Illumina/Solexa	12	01-06-2017 07:34 AM
MiSeq Run with low Cluster PF Help	dross11	Illumina/Solexa	13	03-23-2015 09:01 AM
MiSeq v3 2x300 run, Low Q30	ReGenMC	Illumina/Solexa	16	02-13-2015 04:51 PM

03-26-2018, 06:26 AM	#5
thermophile Senior Member Location: CT Join Date: Apr 2015 Posts: 243	have you looked at the thumbnails? I'd guess your cluster density is higher than reported. __________________ Microbial ecologist, running a sequencing core. I have lots of strong opinions on how to survey communities, pretty sure some are even correct.

03-26-2018, 05:41 AM	#2
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,080	Have you checked the actual insert size of your library? It is possible that you have inserts significantly shorter than you expect them to be. That can explain the precipitous drop in Q score on R2. You can check post #2 in this thread to see how you can calculate that information. This run can also be borderline overloaded. That cluster density number is not very accurate once you get next the upper spec limit. Have you had Illumina tech support look at this run?

03-26-2018, 05:52 AM	#3
SarahAurora Member Location: Bonn Germany Join Date: Jun 2017 Posts: 17	Thank you for your reply. We have done the sequencing of this amplicon multiplex for many times and know the actual sizes. But are the moment the demultiplexing is problematic because it's too less reads. The run could be overloaded but we have also done the sequencing with equal samples and even more input. It was very overloaded but we did get a lot of clusters passing filter (but a bad Q30). Now we have a very low clusters passing filter. So I think that overloading could not be the main issue. No, we have not (yet) been in touch with tech support. I'm still hoping that the clues here could lead to an explanation.

03-26-2018, 06:03 AM	#4
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,080	If you have done similar sequencing successfully in the past then I suggest that you contact Illumina tech support and have them look at the run. This may turn out to be a hardware related problem on the sequencer.

03-26-2018, 10:48 AM	#7
GW_OK Senior Member Location: Oklahoma Join Date: Sep 2009 Posts: 411	Can you post the %base graph?

03-26-2018, 11:07 AM	#9
SarahAurora Member Location: Bonn Germany Join Date: Jun 2017 Posts: 17	It's strange that we can read the first bases of R1, but no R4 and just a few Indices (R2+R3). Looks like the sequencing could not get read out like something intereferred with the optics. I am very concerned about the run and my mind starts to melt, so please excuse this question: If I would have had trouble with the Index-PCR, then the adapters never would have gotten fixed on the flowcell, right? I want to exclude errors while preparing the run.

03-26-2018, 11:09 AM	#10
GW_OK Senior Member Location: Oklahoma Join Date: Sep 2009 Posts: 411	That's intensity. I was talking about the graph that is from % Base, which is selectable in the menu from the gray button under Intensity. Still, though, in looking at this graph it appears you don't have a short insert but you do have sort've an odd 3-steps down to 0 intensity. It looks to me like a hardware issue.

03-26-2018, 11:58 AM	#12
GW_OK Senior Member Location: Oklahoma Join Date: Sep 2009 Posts: 411	What's troubling to me is that you don't have any intensity in read 2, which should always show if you have at least a read 1, regardless of index. Have you run any of the onboard diagnostic tests?

03-26-2018, 12:00 PM	#13
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,080	@Sarah: Do you not have a maintenance contract on this MiSeq with Illumina? If you do then you should go ahead and let tech support have a look at this run.

03-26-2018, 12:25 PM	#18
GW_OK Senior Member Location: Oklahoma Join Date: Sep 2009 Posts: 411	Onboard diagnostics are in the "System Check" icon under "Manage Instrument" from the main screen on the Miseq Control Software. The optics checks are fairly fast to run, the fluidics will take some time and babysitting (raising and lowering sippers as prompted). Has your partner had any successful runs since your failed run? Edit: Asking me to run these tests is usually what Tech Support has me do first when I call them with an issue. Enough so that I'll sometimes run them prior to calling...

03-26-2018, 12:29 PM	#20
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,080	If this failure is due to hardware/reagent problem (and the partner has a maintenance contract with Illumina) then Illumina generally will replace the reagents at no cost, so you should be able to re-run your samples for free (or at least no charge for the reagents).