Issue with using "-t" flag

[EnriqueDoster]

Hello!

A few weeks ago, we changed the AMR++ code to also pass the "threads" variable with the "-t" flag, into the command for calling the SNP_Verification.py script. I didn't notice at first, but it seems like the SNP_Verification.py script cannot parse the "-t" flag. This command fails:

python3 SNP_Verification.py -c config.ini -t 4 -a true -i S1_test.bam -o S1_test.AMR_SNPs --count_matrix AMR_analytic_matrix.csv
ERROR - this is the list of arguments recognized by the program:
        -a: amrplusplus; is either 'true' or 'false'
        -c: config file; if this argument is used, must be the first listed
        -h: help
        -i: BAM input file
        -l: license disclaimer
        -o: main output folder
        -r: conditions for redistribution
        -t: threads for multiprocessing

        --mt_and_wt:            true by default, used in case of insertion leading to presence of both mt and wt; if true, mark as resistant; if false, mark as susceptible
        --detailed_output:      false by default, determines whether a more detailed output will be given; can be either 'false', 'all', or include a list of accessions seperated by commas
        --count_matrix:         count matrix that will be updated if amrplusplus is true
        --count_matrix_final:   the file where the updated count matrix will be found if amrplusplus is true

But, when I run it without the flag it works fine:

python3 SNP_Verification.py -c config.ini -a true -i S1_test.bam -o S1_test.AMR_SNPs --count_matrix AMR_analytic_matrix.csv

Do you think this is something we can fix on your end, or should AMR++ modify the config file if we want to use a different number of threads?

Thank you! Enrique

[Isabella136]

The newest commit should have fixed the problem. Let me know if there is something else

[EnriqueDoster]

Thanks for the quick response! I tried the latest commit and got the following error:

Traceback (most recent call last):
  File "/scratch/user/enriquedoster/test_AMR/AMRplusplus/work/8a/4c2a648a404c41aa74a4d41e757234/SNP_Verification.py", line 349, in <module>
    main()
  File "/scratch/user/enriquedoster/test_AMR/AMRplusplus/work/8a/4c2a648a404c41aa74a4d41e757234/SNP_Verification.py", line 338, in main
    config, argList = parse_config()
  File "/scratch/user/enriquedoster/test_AMR/AMRplusplus/work/8a/4c2a648a404c41aa74a4d41e757234/SNP_Verification.py", line 123, in parse_config
    config['SETTINGS']['THREADS'] = int(arg)
  File "/scratch/user/enriquedoster/conda_envs/AMR++_env/lib/python3.9/configparser.py", line 1261, in __setitem__
    self._parser._validate_value_types(option=key, value=value)
  File "/scratch/user/enriquedoster/conda_envs/AMR++_env/lib/python3.9/configparser.py", line 1188, in _validate_value_types
    raise TypeError("option values must be strings")
TypeError: option values must be strings

I tried it with these two commands and got the same error.

python3 SNP_Verification.py -c config.ini -a true -t 4 -i S1_test.bam -o S1_test.AMR_SNPs --count_matrix AMR_analytic_matrix.csv

# And with quotes
python3 SNP_Verification.py -c config.ini -a true -t "4" -i S1_test.bam -o S1_test.AMR_SNPs --count_matrix AMR_analytic_matrix.csv

[Isabella136]

Let me know if the newest commit still causes the same problems

[EnriqueDoster]

Beautiful, that fixed it. Thanks again!