msuruzhon
commented
4 years ago Hi there,
Regular topologies are a special case of morphed ones when both of your ligands are the same. So if you pass pairs of your ligands of interest in your morphs, it will write out a valid topology (two topologies in fact, but that doesn't make any difference).
This is the quickest way to do it practically. Alternatively, you can create a separate Protein and use its native parametrise() method, which will result in a BioSimSpace object, called complex_template, and then parametrise() another Ligand and load its parametrised_files into BioSimSpace as well. Then you can combine them and solvate using solvate() and saving that should get you what you want. That's a more roundabout way (still only a couple of lines of code), but it's still valid (and it's in fact what the method in Ensemble uses anyway). For more information look at https://protocaller.readthedocs.io. This method doesn't perform ligand alignment to a reference though, so the easier method is probably your best bet .
In any case, you will likely need ligand restraints if you run ABFE, which ProtoCaller doesn't currently support - you will have to add these using some other means.
Hope that helps.
Hello,
I would like to use your (amazing) tool to prepare (remove alternative conformation, protonate, add missing residues, and atoms) and parameterized my protein-ligand system for simple unbiased simulations and later ABFE.
However, it does not seem possible to generate the files if morphs are not supplied. Since I am not aiming (yet) to perform relative calculations, I was wondering if it would be possible to use protocaller without supplying morphs.
Many thanks in advance,