Reference chromosome(s) with no reads mapped to them are not in dictionary keys

[dehui333]

Hi,

Thanks for making this tool, it is useful for my work. I have a question about something:

When I ran hifieval with the -h option, I got the error message "Chromosomes don't match for Homopolymer Evaluation.". After looking at the code and some testing, I think this is because sequences in the reference that are not mapped to by any of the reads (thus not appearing in paf) will not be in correction_dict.keys() but will be in hp_dict.keys(). Is this intended?

hifieval/hifieval.py

Line 463 in 0cdc47e

if correction_dict.keys() != hp_dict.keys():

[magspho]

Hello @dehui333 , sorry for the late reply and thank you for raising this issue! Did you use hifiasm for error correction? (That is the only tool I tested when I run a homopolymer evaluation) I will take a look into this issue as soon as I can and get back you.

[dehui333]

Hi, no worries. I did not use hifiasm for error correction. In fact, the issue can be reproduced by generating a random sequence to use as the "reference", then generating perfect reads from it to use as the "corrected reads" and modifying these reads somewhat (even if just 1bp) to use as the "reads before correction". If I add an extra sequence to the "reference" which does not appear in the pafs as a target sequence, the above-mentioned issue occurs.

Although if I removed that extra sequence, another issue occurs - an error:
ValueError: max() arg is an empty sequence with reference to

hifieval/hifieval.py

Line 510 in 0cdc47e

max_len = max(hp_len_range)