paramx.py expects a rotabase.txt file that is not included with FoldX 5.0

[kalestew]

How can I use paramx.py to properly parameterize an arbitrary ligand from a PDB structure?

I am trying to follow the included notebooks to parameterize a structure of interest but I am running into some issues with the newMol = parameterize( glcLines, atomMappingsDict) parameterize() function from paramx.py.

Just following the included example but it is clear that this method is expecting a rotabase.txt file from FoldX which is not present in 5.0 distributions. I have attempted using the rotabase.txt file and 4.0 binaries but this is not appropriate as the old rotabase.txt file is not json formatted as this package is expecting.

Maybe I am missing something but I have used FoldX 5.0/5.1 a fair amount and have never seen a rotabase.txt file used; there is the /molecules directory which contains the .json definition files for their prepared ligands but those do not contain the mappings for standard residue types.
Am I just missing the right rotabase.txt file from somewhere or is there a way to force FoldX5.0 to generate one?

Running: newMol = parameterize( glcLines, atomMappingsDict) from paramX_AtomTemplates.ipynb:

Mappings loaded:
Atom C1 mapped to atom ('CG', 'PRO')
Atom C2 mapped to atom ('CG', 'PRO')
Atom O1 mapped to atom ('OG', 'SER')
Atom O2 mapped to atom ('OG', 'SER')
Atom C3 mapped to atom ('CG', 'PRO')
Atom O3 mapped to atom ('OG', 'SER')
Atom C4 mapped to atom ('CG', 'PRO')
Atom O4 mapped to atom ('OG', 'SER')
Atom C5 mapped to atom ('CG', 'PRO')
Atom C6 mapped to atom ('CG', 'ARG')
Atom O5 mapped to atom ('O4*', 'A')
Atom O6 mapped to atom ('OG', 'SER')

---------------------------------------------------------------------------
JSONDecodeError                           Traceback (most recent call last)
Cell In[11], line 1
----> 1 newMol = parameterize( glcLines, atomMappingsDict)

File ~/nuProtean/pyfoldx/pyfoldx-env/lib/python3.12/site-packages/pyfoldx/paramx/paramx.py:475, in parameterize(mol_string, mappings_dict)
    472 for atom in mappings_dict:
    473     for table in ["H_Pos", "solvenergy", "hbondinfo", "xyz_water"]:
    474         # Get the records of the table for the selected molecule and the target atom
--> 475         records = JSonParameterLoader.getRecordsInTable(table,mappings_dict[atom][1],mappings_dict[atom][0])
    477         for record in records:
    478             # Re-define the molecule name in the record
    479             record[JSonParameterLoader.getAAFieldName(table)] = JsonMol.threeLetterCode

File ~/nuProtean/pyfoldx/pyfoldx-env/lib/python3.12/site-packages/pyfoldx/paramx/paramx.py:183, in JSonParameterLoader.getRecordsInTable(table, molecule, atom)
    181 @staticmethod
    182 def getRecordsInTable(table,molecule, atom):
--> 183     loaded_dict = JSonParameterLoader.loadTable(TABLE_ALIAS[table])
    186     records = []
    187     for record in loaded_dict[TABLE_ALIAS[table]]:

File ~/nuProtean/pyfoldx/pyfoldx-env/lib/python3.12/site-packages/pyfoldx/paramx/paramx.py:139, in JSonParameterLoader.loadTable(table)
    136     if line == "JSonDataStart %s" % table or line == "JSonDataStart %s" % TABLE_ALIAS[table]:
    137         read = True
--> 139 loaded_dict = simplejson.loads("\n".join(json_lines))
    141 return loaded_dict

File ~/nuProtean/pyfoldx/pyfoldx-env/lib/python3.12/site-packages/simplejson/__init__.py:514, in loads(s, encoding, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, use_decimal, allow_nan, **kw)
    462 """Deserialize ``s`` (a ``str`` or ``unicode`` instance containing a JSON
    463 document) to a Python object.
    464 
   (...)    508 
    509 """
    510 if (cls is None and encoding is None and object_hook is None and
    511         parse_int is None and parse_float is None and
    512         parse_constant is None and object_pairs_hook is None
    513         and not use_decimal and not allow_nan and not kw):
--> 514     return _default_decoder.decode(s)
    515 if cls is None:
    516     cls = JSONDecoder

File ~/nuProtean/pyfoldx/pyfoldx-env/lib/python3.12/site-packages/simplejson/decoder.py:386, in JSONDecoder.decode(self, s, _w, _PY3)
    384 if _PY3 and isinstance(s, bytes):
    385     s = str(s, self.encoding)
--> 386 obj, end = self.raw_decode(s)
    387 end = _w(s, end).end()
    388 if end != len(s):

File ~/nuProtean/pyfoldx/pyfoldx-env/lib/python3.12/site-packages/simplejson/decoder.py:416, in JSONDecoder.raw_decode(self, s, idx, _w, _PY3)
    414     elif ord0 == 0xef and s[idx:idx + 3] == '\xef\xbb\xbf':
    415         idx += 3
--> 416 return self.scan_once(s, idx=_w(s, idx).end())

JSONDecodeError: Expecting value: line 1 column 1 (char 0)

[laurensvdwiel]

Hi @kalestew , I fixed this temporarily by using adding the rotabase.txt file from FoldX 4.1.

[kalestew]

Thanks @laurensvdwiel.

I am assuming you were instead using the 4.1 executable as well too then.
I had actually been trying FoldX 5.0/5.1 with the 4.1 rotabase.txt. Trying now with 4.1 and still getting some errors but I think this time it might just be my path. I'm going to try and define some things manually and just copy the 4.1 rotabase.txt everywhere. - if you were actually still using FoldX5 then it must still be some issue with my paths.

I was concerned though because it seemed like paramx.py is expecting a rotabase.txt file that contains JSON formatted data and looking at the text file I have from 4, that is clearly the old illegible format from FoldX.

[laurensvdwiel]

@kalestew I used the rotabase.txt from 4.1 for FoldX5.1 on MacOs with YASARA without any errors.
In all honesty, I stumbled on this repository yesterday as I was looking to scale up my ddg calculations outside of YASARA.
My fix was thus based on getting FoldX5.1 to run and not directly related to this repository.

I have not yet got this repository/code to work/run yet due to other tasks on my priority list

[kalestew]

Gotcha, @laurensvdwiel. I too just stumbled upon this repo after attempting to parameterize with YASARA and immediately flinched with the task for my large ligand.

I was able to get the YASARA plugin working just fine with FoldX5.1 and I never specified any rotabase.txt file, maybe it was hidden and I missed it? I don't see one included with the plugin itself either.
This did give me the idea to peek into the YASARA plugin files and take a look at how the PM_xxx.py files handle all of this. That is a project lower on my priority list right now though.

Love to hear how it goes when you get a chance to take a look again at this project. I am going to finish up the Rosetta portion of my own project before diving back into this repo (or buckling down for tedious hours in YASARA). Will report back here.

[Kpmurshid]

I faced an issue while running the PyFoldX script. According to the documentation for FoldX5 and FoldX5.1, the rotabase.txt file is not required. However, in the paramx.py script of PyFoldX, there is a hardcoded path for rotabase.txt:

# @todo: make this an option to specify!
ROTABASE_PATH = "rotabase.txt"

As a result, when I run the following line of code:

newMol = parameterize(glcLines, atomMappingsDict)

I get the following error:

JSONDecodeError: Expecting value: line 1 column 1 (char 0)

the script expects the rotabase.txt file, but there is no such file in the PyFoldX directory or in the foldx5 directory other than molecules (json files).

Could you please update the script? if the rotabase.txt file is still required, could you please share it with us? It seems to be missing from the current distribution.

Thanks!

Solved ! :)

[kalestew]

@Kpmurshid How did you solve this issue?

[Kpmurshid]

@kalestew

I resolved the issue by using the rotabase.txt file suitable for FoldX 5/5.1. You can download the compatible rotabase.txt from my GitHub repository, where I’ve also included step by step instructions in the README section.

To make PyFoldX work with this rotabase.txt, you’ll need to update the file path inside the paramx.py file. You can find it at:

pyfoldx -> paramx -> paramx.py

Once you've set the correct path, everything should work smoothly.

If you still need any help, feel free to mail me.