cmuyehara
commented
23 hours ago @xyljoslin - I ran into this same issue. This seems to happen when you run hicpro2higlass.sh using a chromosome-sizes files sorted differently than the one that HiC-Pro uses. It seems to be some combination of something happening in that script, along w/ the fact that Hi-Glass is sensitive to chromosome order.
To fix it, I had to first run the script using the file packaged with HiC-Pro (under HiC-Pro/annotation). Then I resorted the chromosomes to match the order Hi-Glass uses. You can find this file here: https://github.com/pkerpedjiev/negspy/blob/master/negspy/data/.
The command I ran to resort was -
cooler dump --join {cool_file} | cooler load --format bg2 {chrom_sizes_file}:{binsize} - {output_cool}
See here for a discussion of how to re-sort... https://github.com/open2c/cooler/discussions/367
Dear HiC-Pro group,
I used HiC-Pro (mm10) to generate a .mool file and visualize it using HiGlass. The chromatin contact heatmap was generated successfully, but the order of the chromatin in the heatmap is completely wrong. Would you know what is wrong here and how to fix it? The chromatin order should be chr1,2,3,4...X,YM; however, after checking the square size in the heatmap, the chromatin order of my heatmap is chr1, 10, 11, 12, 13,...
Thank you so much!