about HiC data without restriction cutting and abnormally high outward pairs

[Jiang Xu]

Hey guys,

   I've been developing a variant HiC protocol using samples generated from random shearing of DNA fragment and then proximity ligation. I used hicexplorer to generate the contact matrix. For one sample, I found that the fracion of pair used for building the matrix is very low. After consulting one of you guys, I checked the log file and found that the outward pairs is significantly higher than standard high C data. So, could you guys explain the mechanism how hicBuildMatrix judge whether a pair belong to outward pair or not? 

Below shows how I did the experiment:

1. shear the DNA randomly in the nuclear(so there's no fixed breaking point like in a restriction digestion)

2.repair the ends to generate a blunt end.

3. add an "A" to the 3' end of the DNA

4. ligate an adaptor to the A-added ends. (The adaptor has a "T" overhang at one end and a adhesive 5'-overhang on the other)

5.do proximity ligation(using the adhesive 5'-overhang), after ligation, there will form a full BamHI sequence,which was formed from the adaptor.

Thanks a lot!

Best!

Jiang Xu

Lin Chen's Lab

Department of Biological Sciences, Computational and Molecular Biology

University of Southern California

[Vivek Bhardwaj]

Hi Jiang

Outward pairs are the ones where the R1 and R2 are not oriented towards each other.

Expected configuration of R1 and R2 (inward pair) : R1------->>> <restriction site/breakpoint>  <<<----------- R2 

Outward pair:   R1 <<--------- <> ------------->> R2

here arrows mean direction of orientation (5' to 3')

Best Wishes

Vivek

--------

Vivek Bhardwaj

Post-doc Researcher

Hubrecht Institute

3584CT Utrecht 

The Netherlands

--
You received this message because you are subscribed to the Google Groups "deepTools" group.
To unsubscribe from this group and stop receiving emails from it, send an email to deeptools+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/deeptools/CACAVt%3DMHhv%3DMp1BnBT%3DvF3BAri51TFyW_qR-5qwk3MoT-8h4HA%40mail.gmail.com.

[jiang]

Hi Vivek,
Thanks foryour explanation. Does hicBuildMatrix take account of distance between the outward facing pair when judging whether to discard it or not? Or all outward facing pairs are discarded irrespective of their distance? Because I think outward facing pair is still useful when their distance on the reference map is long enough(say, 1000bp). So, does hicBuildMatrix has such parameter of this distance that the user can change?
Best!
Jiang

Sent from my iPhone

> On Oct 24, 2019, at 12:52 AM, Vivek Bhardwaj <vbhard...@gmail.com> wrote:
>

[Vivek Bhardwaj]

Hi Jiang

While the hicBuildMarix by defaults discards all outward-facing pairs with a < 25kb distance in between them (we refer to them as "self-circles"), while the rest are kept. But if you want to reduce this threshold you can first try keeping all the outward-facing reads using the option "--keepSelfCircles". The length however, is hard-coded in the program currently, so you would have to modify the code to change that.

Best,

Vivek

--------

Vivek Bhardwaj

Post-doc Researcher

Hubrecht Institute

3584CT Utrecht 

The Netherlands

[Jiang Xu]

Hi Vivek,
1. Do you have a systematic principle on what pairs will be
discarded, in addtion to like "outward-facing pairs < 25 kb", e.g.
does 25 kb rule still apply for inward facing pairs?
2. Since I don't use restriction enzyme to do the nuclear
digestion(i.e. the genome is sheared by machenical force), can I still
use HiC explorer?
3. Is there any easy way that I can modify the code, on which
part should I modify? I personally know very little about python code.
If you can help it will be great!

Thanks a lot!
Best!
Jiang