search

tanlongzhi / dip-c

Tools to analyze Dip-C (or other 3C/Hi-C) data
61 stars 18 forks source link

How to pileup interchromosomal contact maps? #31

Closed tarak77 closed 5 years ago

tarak77 commented 5 years ago

Hey @tanlongzhi , Hope you are doing good! In your analysis of interchromosomal and long-range intrachromosomal contacts, how did you pileup the inter chromosomal contact maps to generate the super elliptical neighbourhood region? Because the dimensions for each chromosomes are different, I don't know how to overlay the haplotype resolved inter chromosomal contact maps?

Any help will be great!

tanlongzhi commented 5 years ago

Hey @tarak77, hope all is well for you too.

Pileup of contact maps is done with dip-c ard. I think a typical pileup contact "map" (2D histogram of contacts, as in Fig. S7A) can be generated with something like:

dip-c ard -d10000000 -h200000 in.con > in.inter.d10M_h200k.ard

I'll write more details on the README page.

tarak77 commented 5 years ago

That would be awesome @tanlongzhi !!! Also a quick question, I think I am confused a bit.

For one of my sample cell, my haplotype-resolved .hic file looks like this

Screenshot 2019-04-13 at 5 57 32 PM

What I want to do is to pile up all the inter chromosomal boxes(just in upper triangle or both) and see what the pattern looks like. I followed your command dip-c ard -d10000000 -h200000 impute3.round4.con.gz > in.inter.d10M_h200k.ard . The output matrix looks like this

Screenshot 2019-04-13 at 5 57 49 PM

I was confused because looking at the genome wide contact matrix we just see some inter chromosomal contacts along the diagonal. And therefore shouldn't the pileup matrix should also show a diagonal bias instead of the super elliptical region??

P.S. I have attached the .con file used here impute3.round4.con.gz