MiSeq run failure - inhibition?

[firefly2280]

Hi all,

I recently did a MiSeq run that failed before it had really begun. I am doing 16S custom sequencing with the barcodes.

Everything was properly calculated, diluted and Qubited, so I am sure the right concentrations were loaded (went for 8pM) and 10% Phi X.

The only thing we can think it might have been is that some chlorine or other inhibitors may have got through and ruined the run.... Does anyone have any ideas about this or any experience with inhibitory samples?

thanks

[barrmur]

What do you mean by failed? No clusters formed, low intensity data, fluidics check failed??

[firefly2280]

The error message

"Object reference not set to an instance of an object”

came up and that was it, so didnt even get as far as cluster densities.

Illumina thought it was the concentration of my library but It was definately at 2nM to start with then denatured and diluted correctly as per their protocol

[AStretton]

When we had that error come up, it was because we had a pinched line. Try doing a fluidics check to see if there is a blockage.

[firefly2280]

Thanks. Other people are doing successful runs, so I think it is a sample issue sadly!

[mcnelson.phd]

The error message

"Object reference not set to an instance of an object”

came up and that was it, so didnt even get as far as cluster densities.

Illumina thought it was the concentration of my library but It was definately at 2nM to start with then denatured and diluted correctly as per their protocol

Unfortunately, that error message comes up for a large number of reasons on the MiSeq, and it's the most uninformative error ever because you can never figure out what actually caused a problem.

We've had that specific message show up for a number of run failures, and only two or three could be traced back to the same cause. In that case, our optical table wasn't fully secured, so over the course of a run the images became progressively more out of focus until the system just gave up and quit. Other reasons for that error have been traced to bad reagents, bad libraries, and a few other random problems.

Since you said people are successfully using the MiSeq now, best thing to do is get Illumina to replace your kits and try the samples again.

[firefly2280]

thanks for the input. Sadly Illumina are not going to replace the kits and I cant afford to trial and error things, so I dont really know what to do now. The only thing any of us can think of for failure of my library is some kind of inhibition...

[AStretton]

I'm surprised that Illumina are not going to replace the kit considering you had 10% PhiX in with your sample. The PhiX should have clustered even if your library didn't, I would argue that indicates a reagent or flow cell problem.

[mcnelson.phd]

I agree, Illumina should be replacing those reagents. Every time we've had tat object reference error Illumina has replaced the kits. I would pretty much demand that they send replacements unless they can prove that this was your fault, which given the nature of the error they most likely can't.

[microgirl123]

How do you think inhibitors could have gotten into your run? From your original sample DNA? I would think that over the library creation process, you would have done enough clean up steps to get rid of anything!

[rwinegar]

Unfortunately, we have encountered some similar situations recently where there seems to be inhibitory factors in our samples. We always get terrific results from pure samples isolated with DNeasy, but for environmental samples isolated with MOBio kits, we usually have to perform a follow-up MinElute clean-up. Since the MinElute produces losses and cuts out high MW DNA we are currently evaluating other alternatives. The insight we've gained is that removal of PCR inhibitors does not mean you won't have inhibition of Nextera tagmentation.

[fozrun]

Did you put the custom primers into the appropriate wells with the Illumina primers and make sure you didn't indicate in your worksheet that you used custom primers?

[firefly2280]

How do you think inhibitors could have gotten into your run? From your original sample DNA? I would think that over the library creation process, you would have done enough clean up steps to get rid of anything!

Thats what I think too but maybe its something that associates with the DNA and co-purifies?

[firefly2280]

Unfortunately, we have encountered some similar situations recently where there seems to be inhibitory factors in our samples. We always get terrific results from pure samples isolated with DNeasy, but for environmental samples isolated with MOBio kits, we usually have to perform a follow-up MinElute clean-up. Since the MinElute produces losses and cuts out high MW DNA we are currently evaluating other alternatives. The insight we've gained is that removal of PCR inhibitors does not mean you won't have inhibition of Nextera tagmentation.

Thanks for this. Im not using Nextera but I suppose its 'good' to know that this is what might be causing a problem!

[firefly2280]

I ran some samples on a collegues run which worked but my samples only produced 4, 10 and 43 reads...!! so something is for sure going on!