multi model analysis

[bilbibalou]

Hello, I did dockings with smina and therefore I have PDBQT files containing 50 poses each. I followed the PDB tutorial for interaction analysis, but I saw that it only considers a single frame. From what I saw in the Docking tutorial, if I want to analyze all the poses I need to split them into separate PDBQT files, each one, with sorted and then pdbqt_supplier? And does each PDBQT need to have explicit hydrogens? Since I have 50 poses across 10 iterations for 300 ligands, in both flexible and rigid modes, creating one file for every pose would be slow and heavy. Isn’t there a way to run the analysis without splitting everything into separate files?

[cbouy]

Hi @bilbibalou,

There's a whole section on PDBQT files in the docking section of the tutorials explaining the different ways of working with this specific file format.
TL;DR: use Meeko during the PDBQT preparation process to make it easy to accurately export to more robust file formats later on. If you didn't use Meeko, you'll have to split the multi-file PDBQTs into single files, as MDAnalysis only supports single-pose files.

The PDBQT files don't require explicit hydrogens on all atoms, they should be fine as is (i.e. hydrogens on polar atoms only).

Feel free to reopen the issue if you still have related questions

[bilbibalou]

This is also the first thing I checked when I got the error. Everything is in the same folder, I also tried using ./ and the absolute path. That's why I don't understand why only the sdf with 1 model works and I have these errors.

[cbouy]

The fact that for the mol2 file it's raising an OSError: [Errno 22] Invalid argument instead of a FileNotFoundError: [Errno 2] No such file or directory means that the path does indeed exist.
I didn't realize that this was on Windows, could you have a look at this and see if it helps?

[bilbibalou]

I write my paths correctly with "/" or more rarely "\ \" or r"" and I have no probleme in the others scripts of my pipeline. And in this example I don't even use a path since I directly give "1_h.mol2" which is in the script folder, so it can't be a /\ problem.

[cbouy]

Fair enough, what's the output of this?

with open(poses_path) as fh:
    print(fh.read(200))

There isn't really anything fancy going on under the hood of the prolif supplier functions so I'm surprised there'd be an issue with multimodel vs single

[bilbibalou]

Well, I just added your part to my cell to test and I don't know why today everything is working with sdf and mol2. I think it's fix for some random reason

with open(poses_path) as fh:
    print(fh.read(200))
pose_iterable = plf.sdf_supplier(poses_path)```

1
  PyMOL3.1          3D                             0

 44 47  0  0  0  0  0  0  0  0999 V2000
    1.5190    6.3860    7.3570 C   0  0  0  0  0  0  0  0  0  0  0  0
    2.2230    7.7540    7.2620 C