Double peaks of ChIP-Seq library

[hejin1]

Hi,

I used NEBnext ultra II DNA library construct kit for illumina to make ChIP-seq library and bioanalyzer showed there are two peaks: one is around 330bp and the other is 570bp(see the attached file). Input DNA is 1ng and amplification cycle is 15. I used Ampure beads to remove the adaptor after ligation and beads to clean the PCR products. Could anyone give me some insights what causes this pattern. Thanks!

[jdk787]

A trace of your fragmented input DNA would help determine what is causing the two peaks. It could just be a result of the size distribution of your input material.

Also, it looks like your BA software called your upper peak incorrectly so you will need to manually set that to get an accurate idea of what the actual size of your library is.

[pmiguel]

Looks like it could be "bubble products" in the apparently large peak. But like jdk787 says, you may need to manually set your upper size standard marker to get accurate peak sizes.

If the the second peak are bubble products, your library is probably fine.

--
Phillip

[hejin1]

Hi Phillip and jdk787,

Thanks!

Are the bubble products caused by PCR over-amplification and lack of enough primers? Any experiments to prove the 2nd peak are bubble products? Could I increase primers to reduce bubble products?

Your help is highly appreciated!

[hejin1]

Input profile

The input profile is attached.

[jdk787]

Hi Phillip and jdk787,

Thanks!

Are the bubble products caused by PCR over-amplification and lack of enough primers? Any experiments to prove the 2nd peak are bubble products? Could I increase primers to reduce bubble products?

Your help is highly appreciated!

You can run some of your final product on a denaturing gel and see if the upper material goes away. Or amplifying a portion of your library for a few cycles with new primers and running that on a BA would tell you if it is bubble product.

The profile of your library looks a lot like the profile of your input material, so if no size selection was performed your final library looks as expected.

[pmiguel]

Yes, those are probably bubble products, then.
For bubble products, you just ignore them and do your titration using qPCR.
Yes, they are the result of too many cycles of amplification/not enough primer. At least that is the accepted theory.

--
Phillip