[monsterbargain]

Pmiguel,

two reasons i can think of lower than expected cluster density for unamplified high GC bacteria:

1. missing adaptor on the fragment. amplified library has already filtered out those fragments without adp on both ends.
2. High GC fragment doesn't amplify effectively during cluster generation even if it has perfect adp on both ends. the very high GC fragment probably has dropped out in the amplified library already.

ECO,

Do you see discrepancy between the qPCR quantitation and bioanalyzer for amplified and unamplified libraries?

[SeqR&D]

adapters

Think about the adapter system. There are either 2 or 4 oligos that make up the Y-adapter system. Next...how does cluster generation happen (how does the FC grab the library?), and what do you need on your library element to make this happen...AND, with a Y-adapter system without at least 1 cycle of PCR, what are you expecting to get (compared to an amplified library)? Also, your qPCR quant (Ct) might also be off a cycle with an unamplified library. the answer lies within

Out of curiosity, and with the TruSeq v2 Y-adaptor chemistry:

If you have sufficient amounts of adapted DNA, how important is it to do the PCR enrichment? In principle, my expectation is that cluster generation will work without enrichment since the adaptors should be ligated on both strands. In not a few instances, I've been getting yields upwards to 0.5 nM of adapted DNA, which is more than enough for putting on the Miseq.

Thoughts?

[ECO]

ECO,

Do you see discrepancy between the qPCR quantitation and bioanalyzer for amplified and unamplified libraries?

I refuse to use bioanalyzer for quantitating libraries. IME it's an exercise in futility/variability as you aren't measuring cluster-capable molecules. If you amplify a LOT that may be ok, but I need very reproducible clustering.

[james hadfield]

I'd agree with Eco about preferring qPCR for quantification. However run correctly, within its parameters the Bioanalyser can work very well and it gives qualitative information as well. This makes it very useful for some applications.

[austinso]

My limited experience has been that for good libraries, any method is good enough once you figure out the "offset" for that particular method.

qPCR/ddPCR give you more consistency when there are background issues to contend with.

[austinso]

Think about the adapter system. There are either 2 or 4 oligos that make up the Y-adapter system. Next...how does cluster generation happen (how does the FC grab the library?), and what do you need on your library element to make this happen...AND, with a Y-adapter system without at least 1 cycle of PCR, what are you expecting to get (compared to an amplified library)? Also, your qPCR quant (Ct) might also be off a cycle with an unamplified library. the answer lies within

You know, if you had simply stated that you believed that the Y-adaptor sequence is in the same sense as the flow cell oligos, that would have been sufficient.

But the problem with that is that in fact (according to the info I have), is that the Y-adaptor sequences are actually the reverse complement, and can actually bind to the FC oligos as is.

[perilla]

I am curious if anyone ever figured out whether unamplified TruSeq libraries cluster successfully on either the MiSeq or the HiSeq. I see in this strand that unamplified Bioo libraries cluster efficiently but that unamplified TruSeq libraries may have problems? I have prepared an unamplified TruSeq amplicon library that is ready (I hope) for MiSeq sequencing. I used the NEBNext blunt end and A-tail kits and then used TruSeq adapters from the DNA kit in a home-brewed ligation reaction. I quantified the library using qPCR and have plenty of correctly ligated product. I just want to make sure that I don’t need to do anything else to the library before the run. Is a 5 minute 95C denaturation step necessary to guarantee denaturation of the adapters before clustering? Or as the Illumina tech support response states (referenced earlier in this strand), is one PCR cycle absolutely necessary to get good clustering results? I am just trying to play it safe now rather than being sorry later. I’d appreciate any thoughts or suggestions that may be offered.

[pmiguel]

I am curious if anyone ever figured out whether unamplified TruSeq libraries cluster successfully on either the MiSeq or the HiSeq. I see in this strand that unamplified Bioo libraries cluster efficiently but that unamplified TruSeq libraries may have problems? I have prepared an unamplified TruSeq amplicon library that is ready (I hope) for MiSeq sequencing. I used the NEBNext blunt end and A-tail kits and then used TruSeq adapters from the DNA kit in a home-brewed ligation reaction. I quantified the library using qPCR and have plenty of correctly ligated product. I just want to make sure that I don’t need to do anything else to the library before the run. Is a 5 minute 95C denaturation step necessary to guarantee denaturation of the adapters before clustering? Or as the Illumina tech support response states (referenced earlier in this strand), is one PCR cycle absolutely necessary to get good clustering results? I am just trying to play it safe now rather than being sorry later. I’d appreciate any thoughts or suggestions that may be offered.

If you are using Illumina Y-adapters from a normal TruSeq DNA kit, then doing one cycle of PCR would be "playing it safe". But this is really based on my experience with a just a couple of no-amp TruSeq libraries. Nothing else. (They clustered at around 1% the efficiency that they did after amplification.)

Now Illumina has a couple of kits out that specify no amp conditions (their new ChIP seq and PCR free DNA kits). The adapters in those likely don't have whatever the modification that prevents the normal TruSeq adapters from clustering well.

But again, this is all guess work based on only a little data.

--
Phillip

[perilla]

Thank you so much for your response, Phillip. I was still curious because I compared the product number on the adapter tubes from the TruSeq DNA Kit to those listed in the TruSeq DNA PCR-Free Kit and found them to be the same. To confirm, I just called Illumina technical support this morning and directly asked whether the adapters from the normal TruSeq DNA Kit are exactly the same as the TruSeq DNA PCR-Free Kit, and I was told "yes, they are exactly the same." I then said that I have prepared a PCR-free library using the adapters from the TruSeq DNA Kit and asked if I would encounter problems with clustering efficiency if I do not perform a single round of PCR and if I do nothing else to the library before running it. I was told that it would cluster just fine and that I don't have to do anything more to the library. Finally, I asked if this specific protocol has already been performed and tested internally and was told "Yes it has. Don't worry. It will work!"

Phillip, I am not sure whether this was a contributing factor to why you observed 1% clustering efficiency, but when I ran my final library out on a gel, I saw three very distinct bands: one at my starting amplicon size, one ~ 60 bp longer than my amplicon, and one ~120 bp longer than my amplicon. The latter two presumably are amplicon + 1 adapter and amplicon + 2 adapter ligated products. The interesting thing is that they were present in about equal proportion. The unligated amplicon will not bind to the flow cell, but the single adapter products will compete with the double adapter products for binding. Since the single adapter products cannot bridge amplify, they may effectively reduce the observed clustering efficiency. In my case, it would probably reduce it to 50% of expected, but I will be size selecting my library (since I have the distinct size differences) to make sure I don’t include the single adapter products. I realize that you can't really do this if you start with sheared DNA and thus a range of fragment lengths. Depending on the relative proportion of single and double adapter products in your final no-amp library, this might help (but not completely) explain clustering problems?

[pmiguel]

Thank you so much for your response, Phillip. I was still curious because I compared the product number on the adapter tubes from the TruSeq DNA Kit to those listed in the TruSeq DNA PCR-Free Kit and found them to be the same. To confirm, I just called Illumina technical support this morning and directly asked whether the adapters from the normal TruSeq DNA Kit are exactly the same as the TruSeq DNA PCR-Free Kit, and I was told "yes, they are exactly the same." I then said that I have prepared a PCR-free library using the adapters from the TruSeq DNA Kit and asked if I would encounter problems with clustering efficiency if I do not perform a single round of PCR and if I do nothing else to the library before running it. I was told that it would cluster just fine and that I don't have to do anything more to the library. Finally, I asked if this specific protocol has already been performed and tested internally and was told "Yes it has. Don't worry. It will work!"

Phillip, I am not sure whether this was a contributing factor to why you observed 1% clustering efficiency, but when I ran my final library out on a gel, I saw three very distinct bands: one at my starting amplicon size, one ~ 60 bp longer than my amplicon, and one ~120 bp longer than my amplicon. The latter two presumably are amplicon + 1 adapter and amplicon + 2 adapter ligated products. The interesting thing is that they were present in about equal proportion. The unligated amplicon will not bind to the flow cell, but the single adapter products will compete with the double adapter products for binding. Since the single adapter products cannot bridge amplify, they may effectively reduce the observed clustering efficiency. In my case, it would probably reduce it to 50% of expected, but I will be size selecting my library (since I have the distinct size differences) to make sure I don’t include the single adapter products. I realize that you can't really do this if you start with sheared DNA and thus a range of fragment lengths. Depending on the relative proportion of single and double adapter products in your final no-amp library, this might help (but not completely) explain clustering problems?

Sure it could be that both libraries just had issues that were papered over by PCR. Only did 2 that way, so I don't know.
However, at the time I emailed tech support and we had the following email exchange.

First I wrote the following and was assigned a case#:

Sent: 10/14/2011 07:00 AM PDT To: [email protected] CC: Subject: Question about PCR-free TruSeq
Dear Illumina Tech Support,
I see here<http://www.illumina.com/support/sequ...uestions.ilmn:

http://www.illumina.com/support/sequencing/sequencing_kits/truseq_dna_sample_prep_kit_v2/questions.ilmn

The following:

Is there a PCR-free TruSeq Sample Prep protocol available?
The new index adapter design enables PCR-free protocols. (A single cycle of synthesis is required to separate the forked adapter.)

Since samples are strand-denatured (NaOH treatment) would not the cluster PCR serve as the "single cycle of synthesis"?
Regards, Phillip
-----<Email Received-----
Subject: RE: SFC#00341984 - Question about PCR-free TruSeq
Hi Phillip,
Thank you for contacting Illumina Technical Support. Your email has been logged as Case Number 341984. Let me look into your question and see if I can confirm your assumption.

Then I got the following response from Illumina:

On 10/15/2011 11:10 AM, Illumina Technical Support wrote: Hi Philip,
Sorry for the delayed response, but I wanted to clarify a couple of things first.
While you can, technically, use the cluster reactions as a "single cycle of synthesis", it is not that simple. Because of the forked adapter, if it is not properly separated you will have a final product that will be partially single and partially double stranded. That's going to be harder to store stably and quantify accurately.
You can probably get clusters by directly loading ligation product, but the results will be inconsistent and will significantly impact sequencing quality. Hence the need to perform a single cycle of synthesis.
I hope this information is helpful. Please don't hesitate to contact Illumina Tech Support with any additional questions or concerns.
-----<Email Received-----

To which I responded:

Subject: Re: SFC#00341984 - Question about PCR-free TruSeq
Hi, I don't understand what you mean by "properly separated". Why would a single cycle of PCR separate the two strands of the adapter better than the NaOH treatment prior to clustering?
This is of concern because even a single cycle of PCR could cause bias in some cases. Phillip

And then Illumina responds:

-----<Email Received-----
Hi Phillip,
We have found that if you simply take the annealed fragment that has the forked adapters on each end and add it to the clustering reaction that even with NaOH treatment before clustering, there is poor initial hybridization and the resultant clustering is poor. By allowing at least one PCR cycle this separates the strands and the forked adapters, allows a single amplification of the entire strands, resulting in a clean, double-stranded species that gives cleaner clustering results.
The new v3 chemistry and optimized procedures in the TruSeq kits have been designed and tested to significantly reduce bias in the subsequent sequencing reactions. Even with the recommended PCR cycles given in the TruSeq protocol, there should not be significant introduction of bias if you are sequencing a genome or region that is fairly consistent in base composition.
With that said, there is nothing preventing you from going straight from the ligation of adapters to the clustering steps (obviously after you have quantified and validated the library), but our experience has shown us that you may not be as successful with the clustering.
I hope this information is helpful. Please don't hesitate to contact Illumina Tech Support with any additional questions or concerns.

So, anyway, you may be completely correct that the two libraries we tried were just freaky and then we never gave it another shot because of the exchange above.

Or there may still be an issue that the rep you spoke to does not know about.

If you do go ahead, please post your results here.

--
Phillip