Double peak on Truseq RNA-seq library

[Andersen]

Hi All,

I have prepared some Truseq RNA-seq libraries and i have 2 peaks.

I have been reading some previous threads that it could be bead carry-over, but I do not really see any beads in my library and I think some of the libraries look a bit different. Do you have any suggestions to what the problem could be?

I used another protocol than usual to remove rRNA but I do not think it should be the problem (Epicentre - Ribo Zero : http://www.illumina.com/products/rib...-kit-faqs.html)

[nucacidhunter]

Possible reasons:
1-Too many PCR cycles
2- Too much input

I wonder what was the difference between samples 4 and 5 with others and also if samples have been run on BA without dilution.

[pmiguel]

Bubble products.
The good news is that they usually sequence fine.
TruSeq kits ask for 15 cycles of amplification, which is usually crazy high. Later cycles begin to kinetically favor products re-annealing to other over primers annealing/new strand polymerization. Since library molecules are a complex mixture only the adapters anneal leaving a floppy single stranded section in the middle of the heteroduplex amplicons that causes them to migrate slowly during electrophoresis. Or so the lore goes.

--
Phillip

[Andersen]

Possible reasons:
1-Too many PCR cycles
2- Too much input

I wonder what was the difference between samples 4 and 5 with others and also if samples have been run on BA without dilution.

Not a whole lot was done differently, between 4-5 and the others. I believe that the only difference was that 4-5 had 8 cycles and 1-3 had 10 cycles and 6 had 9 cycles. But this was due to the lower amplification test score before amplifying the libraries.

We use 1000 ng as starting material, is this too high, or is it a too much input for the bioanalyzer you think of?

Bubble products.
The good news is that they usually sequence fine.
TruSeq kits ask for 15 cycles of amplification, which is usually crazy high. Later cycles begin to kinetically favor products re-annealing to other over primers annealing/new strand polymerization. Since library molecules are a complex mixture only the adapters anneal leaving a floppy single stranded section in the middle of the heteroduplex amplicons that causes them to migrate slowly during electrophoresis. Or so the lore goes.

--
Phillip

I like your phrasing - Or so the lore goes. Is there any way to control if this is the case, or can I avoid this by any means?


THANKS TO BOTH OF YOU FOR YOUR HELP!

[nucacidhunter]

1000 ng total RNA as input is OK but as you can see the size of larger peak is proportional to the PCR cycle number indicating over cycling. I would think that 7 cycles would have been enough.

Using Good quality RNA results in high PCR yields and I would suggest using BA DNA1000 kit or diluting the libraries to below 1 ng/ul for running on HS Chip.

Keeping PCR cycles the same for the samples that are going to be compared for differential expression is also a good practice to avoid PCR induced biases. This is less important if gene expression analysis is not the aim.