qPCR quantification of mRNA seq libraries: strange amplicon sizes

[Martin_Ranik]

We tried to do qPCR quantification of some Illumina paired-end mRNA seq libraries (+-325bp in length) using the following primers:

qPCR primer 1.1 5’ AATGATACGGCGACCACCGAGAT 3’
qPCR primer 2.1 5’ CAAGCAGAAGACGGCATACGA 3’

The primers anneal to the 5'ends of the adapters and should thus amplify the entire fragment (the qPCR product size should be exactly the same as the original library product)

After doing a serial dilution of a library (with known cluster numbers) we performed qPCR on a Roche LC480.
We used several other libraries (with known sizes) as the unknowns. The standard curve generated was good with PCR efficiency of 93% and very low error. All of the amplifications looked good (Cp<15) and no template controls produced no product at all after 45 cycles.

So on first impression, it seemed that the qPCR worked well - that is until we ran the products on agarose to see the product sizes. It turned out that the products were quite mixed but the prominent product which should have been the same size as the library turned out to be about 200bp larger than the library products.
Pictures below show original library product and qPCR product (arrow). Other libraries (unmarked lanes) also feature the same pattern - product +-200bp larger than the library band.



Any suggestions as to why the qPCR product was so much larger?
Anyone out there doing routine qPCR quant of libraries - your input would be much appreciated!

[james hadfield]

I could not see your initial library on the gel, unlikely I know but running it alongside as a positive control might be helpful. Were all the other libraries the same 350bp size and did you run the PhiX as another control?

It looks like you are using a SYBR based approach which will be confounded by the average insert size when trying to determine the number of molecules present as opposed to the amount of DNA. A TaqMan assay would be more reliable and I think this was proposed in the Quail et al publication.

We are almost there on getting TaqMan QC ito our protocol, but it has been challenging to make it as robust as possible.

[Martin_Ranik]

thanks for your insight James,

Library and qPCR product samples run on agarose side-by-side indicate the size difference as well.

Unfortunately we do not have a phiX library to PCR as control. However, most of the libraries we used in the qPCR have been sequenced on the GA already.

One question: Did you run your PCR products from your qPCR assay on a gel and get the same length as the library? Even when using a taqman assay, with PCR primers annealing to the ends of your library amplicons, you'd expect to amplify products of the same length as the library. If so, what primers have you been using? My primers are designed to anneal to the termini of the adapters.

Any suggestions about how to go about the taqman assay would welcome.