ucm-lbqc/polypharm

A Python-based library to perform IFD and MM/GBSA calculations on different targets using a polypharmacological approach.

polypharm's Issues

UnicodeEncodeError from schrödinger library

If the working directory contains a special character, such as an accent (´) the following error will occur:

[1/1] Failed ifd/proteinx/ligandx with output:
      Traceback (most recent call last):
        File "/data/software/schrodinger/schrodinger2023-4/psp-v7.4/bin/Linux-x86_64/ifd_startup.py", line 289, in <module>
          jobc.launch()
        File "/data/software/schrodinger/schrodinger2023-4/internal/lib/python3.8/site-packages/schrodinger/job/launcher.py", line 915, in launch
          cmd_args = self.getCommandArgs()
        File "/data/software/schrodinger/schrodinger2023-4/internal/lib/python3.8/site-packages/schrodinger/job/launcher.py", line 798, in getCommandArgs
          cmd.encode('ascii')).decode('ascii')
      UnicodeEncodeError: 'ascii' codec can't encode character '\xed' in position 96: ordinal not in range(128)
WARNING: Some commands have failed. Please check the output.

The error also occurs with older versions of schrödinger.

One possible implementation/enhancement is to add a check for the working directory path to warn about special characters found.

The function cross_dock produces a TemplateNotFound error

The module helpers.py at line 14:

    loader=jinja2.FileSystemLoader(ROOT_DIR / "templates")

Produces an error of type TemplateNotFound because the library name is missing in the path.

The function cross_dock produces a KeyError.

In the polypharm.py module, line 355:

resids=",".join(bs_residues[prot_file.name])

A KeyError occurs because it is receiving the prot_file.name variable instead of the basename variable that corresponds to prot_name.