[kcchan]

If you're worried that the NaOH concentration is too high, I would recommend neutralizing the NaOH using a bit of HCl. You can add it just after the first dilution with HT1.

[ECO]

If you're worried that the NaOH concentration is too high, I would recommend neutralizing the NaOH using a bit of HCl. You can add it just after the first dilution with HT1.

We do this, but pH isn't the only factor here, I suspect library complexity and temp (effecting renaturation) are also important here.

Out of curiousity I just went and ran ILMN's standard protocol (as I posted above) and measured pHs...

10ul TrisHCl pH 8.5 (sample)
10ul 0.1N NaOH (diluted fresh)
980ul HT1 (initially pH 7)

Final pH was ~12.5, but this protocol works great. So pH isn't the only issue.

We've been using a neutralization protocol which brings the final pH back to 7, and haven't noticed any real increase in template requirements for v2 MiSeq kits. It's really funny/sad that people are finding they need higher template concentrations for v2 kits...when their final NaOH concentrations are 100% higher if they are following the "new/revD" manuals.

[pmiguel]

We do this, but pH isn't the only factor here, I suspect library complexity and temp (effecting renaturation) are also important here.

Out of curiousity I just went and ran ILMN's standard protocol (as I posted above) and measured pHs...

10ul TrisHCl pH 8.5 (sample)
10ul 0.1N NaOH (diluted fresh)
980ul HT1 (initially pH 7)

Final pH was ~12.5, but this protocol works great. So pH isn't the only issue.

We've been using a neutralization protocol which brings the final pH back to 7, and haven't noticed any real increase in template requirements for v2 MiSeq kits. It's really funny/sad that people are finding they need higher template concentrations for v2 kits...when their final NaOH concentrations are 100% higher if they are following the "new/revD" manuals.

Hi ECO,
Tris has a 0.3 pH unit/10 degrees Celsius pKa delta, though. So if HT1 is tris-buffered, you might have some major systematic errors in your pH calculations if everything is not being measured at the same temp.

Now that we are using a v2 upgraded instrument with v2 chemistry, the issues seem to be gone. Although I did switch to using higher concentration initial library to avoid higher final NaOH concentration problems. (If any.)

--
Phillip

[LVAndrews]

We've been having a hell of a time getting the desired cluster density on our system. The initial run with our FAS (PhiX only) loaded at 12pM and yielded 1068k/mm.

I ran a couple of genome runs subsequently. The first I loaded at 10pM, but I think the lab that prepared it fouled their library quant and it overclustered at around 1700k/mm. Still got 8.8Gb usable data from that run (67% PF).

The next run (genome) I loaded at 7pM at the advice of our FAS. The same lab prepared the library using the same method (Kapa qPCR), and it clustered at only 391k/mm.

I ran a few amplicon runs next (bacterial community amplicons) spiking PhiX between 25 and 40%, but all runs clustered around 400k despite loading between 6 and 8pM each time.

Then another genome run (for the same lab), loading at 10pM and still at 400k. I was checking this lab's quant by now, and it seemed pretty accurate this time. I have been bugging my FAS about this and she sent me a 50cycle kit I am running right now to check the system. Using PhiX only, loading at 10pM, I still only clustered at 600k/mm.

This is really frustrating and we need to be able to get our clustering right every time. I'm ready to start increasing the loading concentration, but those first two runs still have me a little gun shy.

Our instrument was delivered with the upgrade and we have only ever run v2 kits. Ideas?

AK

[kcchan]

Illumina just posted a new technical bulletin regarding sample loading. They now use 4nM of DNA and 0.2N NaOH at 5uL each for the denaturation. This would result in a lower final NaOH concentrations than previous v2 protocols and more in line with the earliest MiSeq and cBot protocols.

[genomeseeker]

Hi AKrohn -
At install (Oct 2012) our FAS said we should aim for cluster densities between 850 and 1200K/mm2 and load 12.5pM. This generally works for us when running balanced samples; low diversity samples are another issue.

Illumina's website indicates you should be able to get max instrument output with cluster densities between 880-965k/mm2 that pass filtering (based on PhiX data, of course).

Definitely follow the new denaturing/dilution protocol kcchan referenced to ensure the final NaOH conc is not more than 1mM in your final dilution. Higher than this can inhibit cluster formation...

[LVAndrews]

Hi AKrohn -
At install (Oct 2012) our FAS said we should aim for cluster densities between 850 and 1200K/mm2 and load 12.5pM. This generally works for us when running balanced samples; low diversity samples are another issue.

Illumina's website indicates you should be able to get max instrument output with cluster densities between 880-965k/mm2 that pass filtering (based on PhiX data, of course).

Definitely follow the new denaturing/dilution protocol kcchan referenced to ensure the final NaOH conc is not more than 1mM in your final dilution. Higher than this can inhibit cluster formation...

This was communicated to us as well. A recent PhiX only run suggests maybe something else is going on with our instrument (10pM clustered at 600k/mm). A field engineer will arrive tomorrow and we will do another PhiX 50 cycle to see if things are clustering appropriately and to get fresh phasing/matrix data for the config.xml file.

According to my math, even with the "old" denature protocol, NaOH still should have been lower than 1mM in the final library. Also, our first PhiX (12pM) clustered at almost 1100k/mm. But yes, I have downloaded the new protocol and will follow the revisions therein.

[matth431]

Interesting that folk are seeing a difference pre and post-upgrade. We've just run our first v2 kit (though not first v1) since upgrade at the usual 8pM and obtained broadly the same density we typically see.