Library QC using RNA 6000 Nano chip

[GraceM]

Hello,

I am new to this forum and I would love your opinion on some peculiar size profiles that we observe after our home-made library prep.

We digest the DNA, ligate adapters, perform 2 times 1xSPRI to remove the adapter dimers, then perform a PCR of 7-10 cycles and finally do a BluePippin size selection (in the range of 390 to 650 bps). After the size selection, we check the library using the fragment analyzer (Advanced Analytical) and often observe additional peaks outside of the selected BluePippin size range (see attach, sample 1 and 5). Reducing the number of PCR cycles seems to reduce the occurrence of these additional peaks (attach, sample 2). When we sequence the library, it turns out that the additional peak can be mapped to the genome and corresponds to genuine restriction digest fragments that are smaller than the selected BluePippin size range.

Question 1: What do you think may cause the small peak of short restriction digest fragments that are co-eluted in another size-range in the BluePippin?

Strolling through this forum, I encountered an interesting theory about single stranded fragments or smaller fragments or adapter dimers co-migrating anonymously with bigger fragments. Since this kind of fits most with our library size profiles, we decided to perform a QC of the library via denaturation and analysis on an RNA chip (as suggested in http://seqanswers.com/forums/showthread.php?t=12523). We analysed both non-denatured samples (red) as well as the denatured sample (2 min @ 95 and snap-cool on ice, blue) on an RNA Nano chip; the results are shown in attach.

Question 2: We observed that the Bioanalyzer with the RNA Nano chip does not detect the additional smaller peaks which we see on the fragment analyzer with the DNA chip. What could be the reason for this?

Question 3: We also observe that the denatured library gets a larger size profile than the non-denatured library, and double humps start to appear. What could be the reason for this?

Many thanks for your suggestions!

[kerplunk412]

1. If this smaller product was very abundant in the pre-selection sample it might make sense to see some of it in the size-selected sample. If another product is very abundant and similar in size to the desired product it can be difficult to completely remove it with any size selection method.

2. This is strange, but maybe the shorter product is re-annealing into bubble products and thus runs with the larger products you are seeing.

3. I suspect that the larger hump in the denatured traces is probably some bubble product that reformed after the heating. I would think the RNA chip would be denaturing, but does anyone know if this is actually true? Maybe you could go old-school and run it on a TBE-Urea PAGE gel to check.

[nucacidhunter]

I do not have experimental proof, but following seems to explain your results:

1- It indicates issues with size selection and probably overloading Pippin cassettes due to increased DNA quantity with more cycles because the small fragments as you mention are real.

2- Every platform has different detection threshold. Even with some fragment range the observed shape or fragment distribution looks different on Agilent TapeStation and Bioanalyzer and also on different Bioanalyser DNA chips.

3- Migration of ssDNA and dsDNA fragments in the Chip is different and therefore they will appear in different sizes. Double hump is indicator of pool of dsDNA and ssDNA.


One approach to overcome this issue would be to clean up with bead (high sample number) or column (low sample numbers) after ligation followed by size selection and then PCR.