Changing the MiSeq annealing temperature

[giderk]

Does anyone have experience modifying the MiSeq to run at an annealing temperature that is less than 65 C? I am following the Kozich et al. 2013 protocol to sequence 18S rRNA amplicons and my custom sequencing primers fall short of a Tm of 65. Had I realized this issue sooner, I might have attempted to change the pad to increase the Tm, but now >300 libraries have been made. It would be expensive and time consuming to redo all that work.

I have the option of extending the length of my sequencing primers into the beginning of my amplicon to increase the Tm, but this will result in losing some diversity that I wanted to detect.

Alternatively, an Illumina technician said that they could change the MiSeq's annealing temperature for my run, but they had no data on whether that might be successful or not.

The latter option would be ideal, but I'm hesitant to try given the cost of each run. Has anyone out there attempted something like this before with or without success?

Thank you

[thermophile]

Whoa? they offered to change the annealing temp? I've talked to several FAS and FAE about my issues with amplicon runs and never had anyone suggest that you could change the annealing temp for a specific run.

I'm increading the primer spike in (was spiking 3.4ul 100mM sequencing primer, now I'm adding 6 ul). this little tweak seems to be helping.

[thermophile]

my saga of annealing temp

http://seqanswers.com/forums/showthread.php?t=65825

[giderk]

Thanks for sharing that experience. Do you have any new information after the temps on the machines were inspected? Did it turn out that they were operating at different temperatures, and does it seem the likely cause of all the read 2 problems?

[thermophile]

what do your fails look like? qscore heatmaps and % base

Since I had access to another MiSeq, I was able to run my problem library on that machine (something about that library just made the annealing temp issue worse). After replacing my TCM, the FAE matched my temps to the other machine on campus.

[kerplunk412]

Can't you just extend the sequencing primers in the other direction, into the P5/P7? Another option would be to add an LNA base or two, but that would be more expensive.

[giderk]

I haven't run the MiSeq yet, but I will soon. My funding for re-runs in the event of failures is really limited, so I am trying to weigh my options.

In response to kerplunk412, I didn't think that extending primers into the P5 and P7 was an option because I have 16 different P5 indices and 24 different P7s. Wouldn't the sites be too variable to be good priming sites?

After speaking with Illumina support some more, they seem least confident in lowering the annealing temperature. It seems my best option is to extend the sequencing primers into my amplicon to increase the Tm, but there is a loss of coverage when I do this. I am now thinking that I will create a couple different sequencing primers and mix them. This won't recover all the diversity I was hoping to, but it will catch some taxonomic groups that I really didn't want to miss. I'm now wondering what the best way to combine sequencing primers is. One of my sequencing primers has a decent amount of degeneracy, the other does not. It makes sense to me that I should calculate the proportion of each sequencing primer variant (resulting from a degenerate primer) and combine my non-degenerate primer in equal proportion with all the variants. Does that make sense to anyone out there?

Thanks already for all the response to this thread.

[HESmith]

If you're going to use a mixture of sequencing primers, why not extend the 5' side (per @kerplunk412) by adding the index sequences (plus upstream P5/P7 bases, if necessary to obtain Tm = 65)? Yes, it's 40 oligos, but those are cheap. And it has the advantage of sequencing all of the diversity that you want to capture.

[giderk]

Hello HESmith, I was worried that with so many variants of the sequencing primers that there would be competitive binding and a negative dilution effect. Is that not what you would expect in this situation? Thank you.

[thermophile]

Are you doing bacteria using exactly the Kozich primers? That is what I do. I wouldn't extend the sequencing primers because you will lose diversity.

Have you had any failures? Or you're just worried you will?

Sorry, you already answered these questions. What is your tm? Kozich bac are 62, those work. Had someone design some at 59, those don't work. For the 59 primers, we're going to order sequencing primers with LNA from IDT to raise the TM without changing the sequence. Do you have someone who will run a few of your samples to see if they work?

[giderk]

Sorry for the delayed response @thermophile, have been at a conference for past week. I am doing 18SrRNA sequencing and the annealing temperatures of the degenerate primers I am using ranges from 60-63.5 and 58-62.5, respectively. I haven't done any runs yet, I have just been worried that the run would fail because I didn't realize the Tm mistake before prepping the libraries. I am scheduled to run the MiSeq tomorrow though.

Although I lose a bit of diversity I decided to extend the sequencing primers into the amplicon enough to raise the Tm to an average of 65. For some specific groups of organisms that I am missing, I have designed additional sequencing primers that I am going to spike in. To test that the sequencing primers work, I am running a MiSeq Nano V2 kit. If the sequencing primer cocktails work, I'll run the MiSeq 2x300 kit later in the week.

Will share the outcome when I know what it is.