RNA TruSeq humps *pre-amplification*

[pmiguel]

TruSeq RNA library Agilent Bioanalyzer High Sensitivity DNA Chip. In green, the ds cDNA prior to ligation. In red after ligation but before any amplification. (Called "pre-amp final library" after this point.)



Notice that the pre-amp final library plot is trimodal. The first (left-most) peak would just be non-ligated cDNA. The middle peak looks reasonable for cDNA + both adapters. (150 bp insert + 120 bp of adapters).

What is that 3rd peak? Again, no amplification at this point. The samples were not heated enough to allow strand denaturation, at least not that I know of.

--
Phillip

This is one of 12 samples. All show the same phenomenon.

[pmiguel]

No one?

Here is what all 12 non-amped libraries, pooled and subjected to 6 cycles of PCR followed by the final RNA TruSeq Ampure clean up:



Or an overlay of all 3:



Blue: is double stranded cDNA prior to ligation.
Red: the same sample after ligation to TruSeq adapters (no amplification.)
Green: sample + 11 of its siblings pooled basically using Ethan's method and amplified 6 cycles + normal Ampure

Part of it is basically what one expects. At the ligated, non-amplified stage you see a trimodal distribution. Leftmost is the remnants of the non-ligated ds cDNA. Middle: fragments + 120 bp of adapter. Right: ???

After amplification: the leftward peak is gone -- did not amplify. The rightward peak? Still there.

I am concerned that the rightward peak is dimer inserts: chimerics? We will size select, just in case.

--
Phillip

[pbluescript]

Is there a difference in ligation efficiency depending on the fragment size? I haven't noticed anything like that, but I haven't specifically tested for it with something like a DNA ladder.

[pmiguel]

Is there a difference in ligation efficiency depending on the fragment size? I haven't noticed anything like that, but I haven't specifically tested for it with something like a DNA ladder.

Not that we have seen. We have successfully constructed 454 libraries with the RNA TruSeq kit. Actually the cDNA was longer than we wanted, so we had to sonicate it. Of course those were Roche Y-adapters.

But we also did a 1 second fragmentation time library construction. If I remember correctly, we got 2+ kb libraries. Not useful for anything, but I was impressed.

--
Phillip