kundajelab / 3DChromatin_ReplicateQC

Software to compute reproducibility and quality scores for Hi-C data
MIT License
43 stars 16 forks source link

Potential issue in wrappers/QuASAR/data_to_hifive.py #13

Open henrybzhang opened 5 years ago

henrybzhang commented 5 years ago

This may be a issue specific to me, but I had float values in my contact map which caused similar issues in https://github.com/bxlab/hifive/issues/12 where there were no valid read data.

I approximately fixed it by modifying line 83 of wrappers/QuASAR/data_to_hifive.py to account for float values: count = int(temp[4]) -> count = int(float(temp[4]))

Additionally, perhaps exceptions could be printed for easier debugging purposes?

stelmanAM commented 4 years ago

I had the same issue and this code change worked for me also. I had generated contact maps from hic files using juicer_tools dump, and the resulting scores are float values.

gbonilla18 commented 4 years ago

I had the same issue, since I generated the contact maps using juicer_tools. I wish it was better documented, but to be fair, the example provided by the authors shows counts as integer. I just wish I had discovered this post sooner! Thank you

Devalock commented 1 year ago

Much aprreciated! I sincerely hope @oursu could modify this part, because most HiC contact normalization generate floats, so that future users won't have to spend hours struggling with where the problem is.