Jwindler/AutoHiC

AutoHiC seems to be making changes, but results look same.

kiratalreja3 opened this issue · 2 comments

Hi, I have tested AutoHiC at least 5 times in the "onehic.py" mode, and every single time it seems to be making adjustments (As indicated by the infer_results folder) - but as soon as I use 3d-dna post review script to generate .hic map on the adjusted assembly, it looks exactly the same as what was supplied to AutoHiC. Am I missing something?

Before AutoHiC :
Screenshot 2023-10-05 at 10 18 25 am

After AutoHiC:
Screenshot 2023-10-06 at 10 19 47 am

{"Raw error number": {"translocation": 3, "inversion": 0, "debris": 573}, "Score filtered error number": {"translocation": 3, "inversion": 0, "debris": 561}, "Length filtered error number": {"translocation": 1, "inversion": 0, "debris": 559}, "Length removed error number": {"translocation": 2, "inversion": 0, "debris": 2}, "Overlap filtered error number": {"debris": 365, "translocation": 1}, "Chromosome real length filtered error number": {"translocation": {"normal": 0, "abnormal": 1}, "inversion": {"normal": 0, "abnormal": 0}, "debris": {"normal": 45, "abnormal": 320}}}

This is the error summary.

Can you please help me out here?

Thanks

autohic.e97171355.txt
Here is the log, and the .assembly files before and after

Before:
base.ctg_info.assembly.txt

After:
adjusted.assembly.txt

These are normal operating conditions. There are many debris errors in the genome. These errors are small and usually require many iterations to remove. My suggestion is to stop running onehic once the translocation and inversion errors have been removed from the genome.