Load .hic file for large genome & chromosomes

[tbrown91]

Hi,

I am trying to load a .hic file into jbrowse2 v1.6.5 desktop and am getting nothing showing up. This genome is 20Gb in size and the smallest chromosome is about 700Mb. I was able load genome and gff (with .csi indexing) fairly ok, but the hic track is just blank.

I first started with the whole-genome hic file (17Gb) and then one for only the smallest chromosome (224Mb) but both show nothing. It's difficult for me to say anything more as there is no error message given.

Here is the example juicebox image for smallest chromosome:

Is there a way to check .hic file formatting is correct, or to potentially see what the issue may be?

Thanks,

Tom

[cmdcolin]

thanks for the feedback :) definitely curious...are you able to load the data in juicebox.js (https://aidenlab.org/juicebox/)? we use the same library for parsing the .hic files as them. the other thing that can sometimes cause tracks to not show up is "refname aliasing" e.g. using chr1 vs 1, in which case the refNameAliases can be configured, or the files can be standardized on the same naming scheme

[cmdcolin]

the upper case vs lower case chromosome names could be a factor....if you want to send me the file (chrom sizes and hic) i can check it out also

[tbrown91]

Do you need the genome as well, or are the sizes and hic file enough?

Do you have an upload link in case everything is too big?

[cmdcolin]

chrom.sizes and hic should be enough (as long as the chrom.sizes names match the fasta to make sure though)

I don't have an upload link myself but google drive or something like that might work

[tbrown91]

You were right! I changed the chromosome names in the fast to capitals and it loaded. It's a bit of a pain that juicer_tools pre did this, but at least I have a workaround now.

Thanks for your help.

[cmdcolin]

excellent...jbrowse 2 is pretty strict about names matching while other tools might smooth this over. glad this worked out though :) let us know if anything else comes up