TruSeq RNA PCR Cycle Number

[little_henry]

Hello,

I'm seeing two size populations in my final RNAseq libraries, one at ~280 bp which is the expected library size and a larger one at ~700 bp which isn't expected.

I start with 500ng of intact total RNA and follow the TruSeq protocol with 15 cycles of PCR. When I back the number of cycles down to ten the larger sized population disappears (see picture).

Has anyone else experienced this?

Thanks!

[torben]

Hi,

the extra peak is probably due to overamplification which can give you ssDNA that migrates slower than dsDNA. See e.g. slide 30 here: http://www.mbl.edu/jbpc/files/2014/0..._slideshow.pdf

[pmiguel]

Well, probably not completely ssDNA.

If the PCR reaction runs low on primers and high on products, then amplicons annealing back to each other becomes kinetically favorable. But since this is a library then the likelihood is that the inserts of the two annealling amplicons will be unrelated. So you end up with "bubble" products. Double-stranded at the ends (where the adapters are) but with little or no annealing in the middle.

These floppy bubble products apparently migrate slowly and so appear to be larger than they actually are. Happily you can pretty much ignore them as they cluster and sequence fine. (Although they can play havoc with your cluster titrations if you don't understand what they are.)

This probably the single most common artifact posted to this forum! But in recent years one only sees new posts about it a couple of times per year.

--
Phillip

[luc]

The Illumina Truseq protocol must be written horribly - almost all of our customers using the Truseq RNA-seq kit for the first time totally over-amplify their libraries (15 cycles).