Rework `filter_response` response method in `lib/prediction_class.py`

Hi,
a few days ago, i have installed Pvactools and got test run that Pvactools homepage guide "Getting started". https://pvactools.readthedocs.io/en/latest/pvacseq/getting_started.html
i referred this code
/pvacseq_example_data/input.vcf \ Test \ HLA-A*02:01,HLA-B*35:01,DRB1*11:01 \ MHCflurry MHCnuggetsI MHCnuggetsII NNalign NetMHC PickPocket SMM SMMPMBEC SMMalign \ /pvacseq_output_data \ -e 8,9,10 --iedb-install-directory /opt/iedb

but it gave me this error.

Traceback (most recent call last):
File "/opt/rh/rh-python36/root/usr/lib/python3.6/site-packages/lib/pipeline.py", line 343, in
call_iedb
lib.call_iedb.main(arguments)
File "/opt/rh/rh-python36/root/usr/lib/python3.6/site-packages/lib/call_iedb.py", line 62, in main
tmp_output_filehandle.write(response_text)
TypeError: a bytes-like object is required, not 'NoneType'

so i checked call_iedb.py script and prediction_class.py, i find something.
line 45 of prediction_class.py, there is if line.startswith(b"allele"): code.
but the input file contents are like below

b'HLA-DRB1*11:01\t34\t802\t816\tPPCAVCIGH\tSPEPSQPPCAVCIGH\t29283.0\t99.21' 
b'HLA-DRB1*11:01\t34\t843\t857\tPESEPPHHE\tPESEPPHHEPPYIPP\t29636.2\t99.29'

so i think that lines were skipped to else and return nothing.

i wonder whether i have installed Pvactools and iedb as wrong way or something else?

Best Regards
Jeongmin

Start with lib/prediction_class.py and the filter_response method, then trace how its result is used by lib/call_iedb.py at the response_text write. Reproduce the documented pvacseq command and inspect the reported NoneType path. Done means the IEDB response is handled correctly without the shown write failure.