Library Quantification Confusion!

[peromhc]

I have just finished making some Illumina mRNA libraries using the v2 TruSeq Kit. I used the QUBIT to quantify samples, and got high concentrations (200-600ng/uL). Bioanalyzer traces suggest that there is a lot of material there, as well, and of the appropriate size.

I contrast, I used the Kapa Bio library quantification kit, and got very different numbers concentrations, ranging from .01-90ng/uL. At the end of library prep, after enrichment PCR, the library should be enriched for dsDNA fragments that have adaptors on both ends, and thus are PCR competent. There should not be a lot of dsDNA floating around I don't think..

So, why such big differences?

Attached you'll see a bioanalyzer trace (1:10 dilution) of a library. QUBIT told me concentration was 321ng/uL, but Kapa qPCR says only .12ng/uL..

Can anybody suggest a reason why this may be?

[niceday]

do you have quantitation for pre and postPCR enrichment?
DEpending on the number of cycles you should see an increase in the amount of library.
If the increase is low then the amount of properly ligated library is low before PCR or the PCR hasn't worked and you haven't amplified your correctly ligated material over your incorrectly ligated material.

[Eric@Kapa]

There are many possible explanations for this, but assuming there was nothing wrong with the qPCR assay, I would suggest that you proceed with caution -- the KAPA qPCR primers correspond to the flow cell oligo sequences, so anything that does not amplify during qPCR will not produce clusters on your flow cell.

Without seeing the details of your qPCR assay, it's impossible for me to comment on potential problems. Please feel free to send full details to [email protected]:

* details of how the assay was set up, which qPCR instrument was used, etc.
* screen-captures of amplification profiles for standards and unknowns
* screen-captures of standard curves
* spreadsheet containing data including Ct scores of replicates, calculated concentrations, etc.

Hopefully we can help to shed some light on your problem!

Best regards,

Eric

[pmiguel]

The simple explanation would be that you generated a large amount of cDNA and that relatively little of it was converted to library molecules. Presuming there was no issue with blunting/A-tailing your ds-cDNAs and ligation went as expected, then the limiting factor would be the amount of adapter available. In another thread the relative concentrations of adapter in the DNA vs RNA kits was determined to be 60:1.

Once the adapter:insert molar conc. ratio drops below 2 you get into problematic territory. That is, if there is only 1 adapter available for each insert molecule, chances of getting a molecule with an adapter on each end drops to 25% (0.5^2).

That said, the 10x difference in concentration between your Agilent chip (30 ng/ul -- taking into account the 1:10 dilution) and your QBIT is mysterious.

--
Phillip

[peromhc]

perhaps more enlightening (see image)

this is a different sample than the one I reference above, but the problem is the same. Note that between the top image (sample before PCR enrichment) and below (after PCR enrichment) that there is obvious peak enhancement.. Despite this, this the qPCR conc. value is very low.

[HESmith]

Did you include a positive control sample (e.g., phiX) in your qPCR? If that value is off, it would indicate a problem with your instrument or reagents.

[NextGenSeq]

Note the small RNA adapters have a mismatch at the 3' end for the KAPA QPCR reatltime PCR primers. This gave us a huge difference in quantification of a small RNA library.

Anyway, double check that the KAPA primers are correct for your library prep.

[pmiguel]

Also, check the calibration on your QBIT.

--
Phillip

[sehrrot]

Yes recalibration for QuBit I would recommend. Even though it's done, QuBit gives not too accurate value and somtimes differs by a type of QuBit.

[pmiguel]

The OP is getting 100x higher numbers on the QBIT than on the BioAnalyzer. That suggests the a decimal place slip and/or the standards are 100x off.

--
Phillip