HiC Ligation

[Julie]

Hi, 

We are currently working with the in situ HiC protocol and we are having problem with the ligation verification after biotin fill in. We check with gel after every steps. The digestion step was fine which we see a wide smear. We expect to see a shift of the DNA to a higher molecular weight after ligation, but so far, in all our attempts, we have never seen it. We tried to sequence one of the library, but the analysis results shows that majority of our reads are dangling ends or single fragment. We suspect that the ligation is not working. Any suggestions?

[Elena Stamenova]

Hi Julie,

I am sorry that the ligation may not be working for you. If you are following strictly the in situ Hi-C protocol as outlined in the Extended Experimental Procedures of Rao & Huntley et al, Cell 2014, the only suggestion I have is to make sure that you start with high quality material e.g. cells with low viability at harvest would not produce good libraries.  If there is DNA fragmentation that is not result of restriction, biotin will get incorporated non-specifically and pulled-down material may not be chimeric.

I also would note that not all restriction buffers are compatible with the fill-in step in case you are not using MboI  (e.g. DpnII restriction should be performed in NEB3 or NEB3.1  buffer and  not in the DpnII buffer supplied with the enzyme).  If fill-in reaction is incomplete, ligation will not be successful.

Hope that this may help.

Best regards,

Elena

[Julie]

Hi Elena,

Thanks for the reply. We are using MboI in this case. Regarding the starting material, we will look into see to see if the DNA fragmentation is due to degradation. However, we did compare undigested samples and digestion samples and only the digested one give smear at smaller DNA size in our previous trials. On top of that, is there any way to determine the efficiency of ligation apart from the gel shift? The weird thing with the ligation is that when we do bioanalyzer analysis, the post ligation sample seems to have smaller average fragment size than the post digestion which we find it hard to comprehend. Have anyone seen this at any point? Thanks.

[Elena Stamenova]

Hi Julie,

You mentioned that you have sequenced one of your libraries.  You should be able to see the ligation junction (GATCGATC for MboI)  in your data. We normally look at this to judge if ligation was successful and do not run gels.

Best regards,

Elena

[Ola]

Hi Julie,

just saw your post fro June. Did you succeed? 

If I saw this in June I would ask you:

1. Did you definitely inactivate MboI at 65'C after restriction digest?

2. What ligase did you use and how much? (I'm assuming you're going blunt end Hi-C ligation?). Did you use a freshly bought enzyme?

3. Is there any chance your sample got contaminated with EDTA or salt?

Best

Ola

[Julie]

Hi Ola,

Apparently, I only saw this in July.

Yes and No, we succeeded the HiC by switching to a low cell number protocol, but is still working very hard on the in situ HiC with 1 million cells as starting material. We still hope to get this working as the low cell number protocol fails to provide high resolution.

1. Yes, we heat inactivate the MboI.

2. Amount of ligase used is 5ul (400U/ul)

3. It's rare that we get contamination in every batch of experiments, we repeated the experiments with fresh reagents when they fail.

Best,

Julie