ChiP-seq - ThermoScientific Pierce Kit - possibly overfixed DNA

Elsie · 05-18-2014, 08:33 PM

Hi, newbie to ChiP-seq. I have two libraries prepared with the Thermo Scientific Pierce Magnetic ChIP kit, micrococcal nuclease digestion and sonication. QC from an Agilent Tapestation shows several low peaks at ~180bp, 363, 537 and 713bp respectively. Normal practice I'm told is to pick at 180bp and generate libraries from that. With such a tiny peak there, I could either:
- stick with 180bp, but not have much material to generate my library from
- pick from several peaks but then end up with fragments of different sizes, which could make the bioinformatics hard.

So would I:
- do the size selection and end up with a small amount of material
- skip the size selection (or rather have a broad size selection) and deal with the consequences later

Any advice appreciated, this experiment is unlikely to be repeated so I'll need to work with what I've got.
thanks.

nucacidhunter · 05-19-2014, 06:08 AM

If you prepare libraries with DNA samples described there, there will be two downstream issues. Firstly, accurate quantification of libraries for clustering with multiple peaks is difficult and inaccurate quantification can result in over or under clustering which will affect read number or quality or both, although that might be considered sequencing service provider problem. Secondly, small fragments are more efficient in cluster formation in comparison with large fragments and more of your reads will be from sequencing smaller fragments.

Elsie · 05-19-2014, 03:09 PM

Thanks Nucacidhunter, your comments are exactly what I suspected would happen downstream. Unfortunately these libraries are not going to be remade so I will just have to work with what I've got. Thank you.

nucacidhunter · 05-19-2014, 04:00 PM

One suggestion I have reluctantly because I have not tried it myself, is shearing pool of your DNA with a Covaris set to 150-200 bp if you have access to one. In this case smaller fragment would not shear any further but larger fragments should shear. You can try shearing a cheap source of DNA to your current peaks. Pooling them to emulate your current ChIP DNA pool and trying shearing on them first to ensure that it works and that also will allow you to optimize if things does not work out at first go.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
New NEXTflex Rapid DNA-Seq Kit Offers Two-Hour NGS Library Construction from 1 ng DNA	Bioo Scientific	Vendor Forum	3	11-10-2014 07:54 AM
NEBNext ChIP-Seq library prep kit vs Illumina TruSeq kit?	ASL	Sample Prep / Library Generation	1	04-10-2013 08:35 PM
Can I prepare ChIP-seq library by using TruSeq DNA sample preparation kit	lyxsdu2006	Sample Prep / Library Generation	8	05-08-2012 07:41 PM
NEXTflex ChIP-Seq Kit for Improved Genome Wide Analysis of DNA-Protein Interactions	Bioo Scientific	Vendor Forum	0	11-07-2011 09:25 AM
ChIP-seq of possibly high repeated sequences	dawe	Epigenetics	8	09-10-2009 12:04 PM

05-18-2014, 08:33 PM	#1
Elsie Member Location: Australia Join Date: Mar 2011 Posts: 85	ChiP-seq - ThermoScientific Pierce Kit - possibly overfixed DNA Hi, newbie to ChiP-seq. I have two libraries prepared with the Thermo Scientific Pierce Magnetic ChIP kit, micrococcal nuclease digestion and sonication. QC from an Agilent Tapestation shows several low peaks at ~180bp, 363, 537 and 713bp respectively. Normal practice I'm told is to pick at 180bp and generate libraries from that. With such a tiny peak there, I could either: - stick with 180bp, but not have much material to generate my library from - pick from several peaks but then end up with fragments of different sizes, which could make the bioinformatics hard. So would I: - do the size selection and end up with a small amount of material - skip the size selection (or rather have a broad size selection) and deal with the consequences later Any advice appreciated, this experiment is unlikely to be repeated so I'll need to work with what I've got. thanks. Last edited by Elsie; 05-18-2014 at 08:49 PM. Reason: Clarify question.

05-19-2014, 06:08 AM	#2
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	If you prepare libraries with DNA samples described there, there will be two downstream issues. Firstly, accurate quantification of libraries for clustering with multiple peaks is difficult and inaccurate quantification can result in over or under clustering which will affect read number or quality or both, although that might be considered sequencing service provider problem. Secondly, small fragments are more efficient in cluster formation in comparison with large fragments and more of your reads will be from sequencing smaller fragments.

05-19-2014, 03:09 PM	#3
Elsie Member Location: Australia Join Date: Mar 2011 Posts: 85	Thanks Nucacidhunter, your comments are exactly what I suspected would happen downstream. Unfortunately these libraries are not going to be remade so I will just have to work with what I've got. Thank you.

05-19-2014, 04:00 PM	#4
nucacidhunter Jafar Jabbari Location: Melbourne Join Date: Jan 2013 Posts: 1,238	One suggestion I have reluctantly because I have not tried it myself, is shearing pool of your DNA with a Covaris set to 150-200 bp if you have access to one. In this case smaller fragment would not shear any further but larger fragments should shear. You can try shearing a cheap source of DNA to your current peaks. Pooling them to emulate your current ChIP DNA pool and trying shearing on them first to ensure that it works and that also will allow you to optimize if things does not work out at first go.