Correctly sort gmx xvg files for analysis

[fjclark]

This pull request fixes issue #383

• I confirm that I have merged the latest version of devel into this branch before issuing this pull request (e.g. by running git pull origin devel): y
• I confirm that I have permission to release this code under the GPL3 license: y

Suggested reviewers:

@lohedges

[fjclark]

Sorry, I forgot to blacken again.

[lohedges]

Thanks, @fjclark. Is this only an issue for GROMACS, or are any of the other engines affected?

[fjclark]

Looking at

function_glob_dict = {
            "SOMD": (Relative._analyse_somd, "**/simfile.dat"),
            "SOMD2": (Relative._analyse_somd2, "**/*.parquet"),
            "GROMACS": (Relative._analyse_gromacs, "**/[!bar]*.xvg"),
            "AMBER": (Relative._analyse_amber, "**/*.out"),
        }

I think the code:

• Should currently be fine for SOMD as directories are named by lambda values
• Is fine for SOMD2 as the the parquet file is named by the lambda value
• I'm not sure about amber as I don't know how the output is named

An alternative fix, which is probably simpler, would be to ensure a consistent number of digits so that the string-based sorting works, e.g. the lower example:

In [26]: sorted(["lambda_19", "lambda_2"])
Out[26]: ['lambda_19', 'lambda_2']

In [27]: sorted(["lambda_19", "lambda_02"])
Out[27]: ['lambda_02', 'lambda_19']

Would you prefer this? Thanks.

[lohedges]

I guess I'm confused. By default, when setting things up via BioSimSpace.FreeEnergy.Relative, then output directories are always namged by the lambda value not index for all of the engines. (I know that Exs changed this in their sandpit.) For example:

In [1]: import BioSimSpace as BSS

In [2]: import sire as sr

In [3]: mols = sr.load_test_files("merged_molecule.s3")

In [4]: system = BSS._SireWrappers.System(mols._system)

In [5]: freenrg_gmx = BSS.FreeEnergy.Relative(system, BSS.Protocol.FreeEnergy(), engine="gromacs", work_dir="fep_gmx")

In [6]: ls -l fep_gmx/
total 44
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.0000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.1000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.2000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.3000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.4000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.5000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.6000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.7000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.8000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_0.9000/
drwxr-xr-x 2 lester lester 4096 Jan 24 18:13 lambda_1.0000/

So, if each directory contains a gromacs.xvg file, the sorted glob gives:

import pathlib

glob_path = pathlib.Path("fep_gmx")
files = sorted(glob_path.glob("**/[!bar]*.xvg"))

for file in file:
    print(file)

gives:

fep_gmx/lambda_0.0000/gromacs.xvg
fep_gmx/lambda_0.1000/gromacs.xvg
fep_gmx/lambda_0.2000/gromacs.xvg
fep_gmx/lambda_0.3000/gromacs.xvg
fep_gmx/lambda_0.4000/gromacs.xvg
fep_gmx/lambda_0.5000/gromacs.xvg
fep_gmx/lambda_0.6000/gromacs.xvg
fep_gmx/lambda_0.7000/gromacs.xvg
fep_gmx/lambda_0.8000/gromacs.xvg
fep_gmx/lambda_0.9000/gromacs.xvg
fep_gmx/lambda_1.0000/gromacs.xvg

which is the correct order. Are you setting things up in a different way so that the directory naming structure is different for GROMACS? If so, then I agree that you would need to use a consistent number of digits in your naming scheme.

[fjclark]

I know that Exs changed this in their sandpit.

Sorry, this was exactly the issue - I was running gmx ABFE with the sandpit, but analysing with standard BSS for consistency with SOMD2. The sandpit analysis works fine for the gmx output.

[lohedges]

Ah, that makes sense. No problem. I think we'll eventually move over to naming things by index, so I'll make sure to use an appropriate number of digits when I do.