Hi Yiftah,

Better enrichment is not always in the same way of good specificity... What about the specificity of your antibody? I got the same once, but I knew that my antibody was not great (and it was the only one available for my protein of interest :/).

How do you do your lysis, whole or nucleus lysis? I got better results when I was using nuclear lysis. 

For your beads, are they blocked with salmon sperm? This is not good for ChIPseq. I blocked my beads with BSA...

I hope I could help you a little bit...

I am afraid to tell you that " excellent enrichement vs the IgG beads"..does not mean your ChIP works. You need more controls, normalizations...not possible to explain it all here.

In fact, your ChIP seq tell you the ChIP does not work you just see random DNA.

Try to improve ChIP protocol/detection/antobody.....if you are postive about your ChIP results then go for ChIP seq.

Good luck!

Hi Yiftah,

I think if you used different primer sets for same genomic region and got inconsistent results, it might be caused by poor specificity of primer sets you used.

To confirm your ChIP, I suggest running the ChIP-qPCR for both positive and negative control in your experiment. If you cannot see any peaks in ChIP-seq even in genomic region of positive control, then the enrichment vs IgG control may only be simply caused by different amount of DNA between experimental sample and control sample.

Besides, selection of antibody is primarily essential, I prefer polyclonal antibody rather than monoclonal one because usually polyclonal antibody is more sensitive. Especially in your case, although as you claimed your target protein is abundant in nuclear, it does not mean that abundant target protein molecules bind to the chromosome. Protein without forming complex with DNA will weakens the efficiency of Protein-DNA complex immunoprecipitation.

DNA fragmentation is also important, sonication and enzyme digestion (micrococcal nuclease) are commonly used. I prefer enzyme digestion because it is more gentle and thorough for DNA fragmentation without affecting protein-DNA combination.

I hope this helps and good luck with your experiment!

Thank you all. Your answers are very helpful. Can you think of a reason for such intense percipitation of dna besides non-specific binding? i am talkig about 50-60 fold more dna in ip vs control. Is it posdible that I cross link all the dna to my protein of interest and percipitate it?

Yes, too much crosslinking is bad. Another posibility is that your antobody precipitate lots of non-specific DNA , or (less likely) that the antobody is contaminanted with DNA - this happened to me XD.

I agree with Pedro. I think that your antibody precipitates non-specific DNA. You can try less time for antibody incubation instead overnight (I don't know how much time you use in your protocol...), or you can try binding the antibody to the beads before add your sample. 

Hi Yiftah! 

As it has been mentioned before an enrichment over IgGs does not mean that your IP has worked.

As for regular immunoprecipitation, your IP MUST be enriched over the input in the positive control to detect a signal in ChIPseq. 

What I do before sending samples to sequence is:

-Measure the concentration of input and IP by Qubit (do not use Nanodrop, it's not sensible enough)

-Run a qPCR loading 2ng/well of INPUT or IP DNA (if you don't have enough, you can try less, I've succesfully use as few as 0.2ng/well) using primers of your positive or negative control.

-Compare the Cts obtained, if your ChIP has worked: 

In the positive control: Ctinput > Ctip 

In the negative control: Ctinput < =Ctip

Hope it helps! 

I agree crosslinking could represent a problem. I suggest to perform some tests changing crosslinking time, or trying to crosslink at 4°C. If protein is abundant, it is definitely better to incubate for short time (few minutes at RT). This said, there are several other conditions to consider. For example, some magnetic beads could work better than others. In my experience, Dynabeads protein A work much better than Dynabeads protein G. I suggest to incubate beads with antibody in PBS/BSA overnight at 4°C. Consider also to increase stringency of the washing steps.

About peaks, consider that ChIP-seq is not quantitative as ChIP-qPCR and this might reflect lower peaks than what expected. This is because sequencing provides IP values relative to input, while qPCR returns absolute values. Comparing IP to a massive amount of input in CHIP-seq could therefore reduce the IP peaks height in the final ChIP map. Good luck!

 Hi Davide, thank you for your answer. Was there any case that you saw peaks in one system of magnetic beads but not in the other? Is it possible that a+g are causing problems? I would like to hear if there are known problems with magnetic beads and getting no peaks. Thanks.