Using ffid flag to change the charge set

[FrederikGoetz]

I tried using the ffID flag to change the charge set to CHARMM ('ffID' : 'CHARMM36m'). After I did this the console print out just showed Start Preprocessing for each of the input PDBs and did not give any pKa results.

[pedrishi]

Hi! The same conformations worked without setting ffID? Could you share one these conformation and your parameters so that I try to replicate the problem?

[FrederikGoetz]

Hello it worked with those parameters:

params = { 'structure' : file_path, 'ncpus' : 8, 'epsin' : 15, 'ionicstr' : 0.1, 'pbc_dimensions': 0, 'ffID' : 'G54A7', #'ffID' : 'CHARMM36m', #Set the ffinput when using PDB files from simulations #'ffinput': 'GROMOS' # options: GROMOS, AMBER, CHARMM }

It also worked without setting the ffID.

But it did not work with those parameters:

params = { 'structure' : file_path, 'ncpus' : 8, 'epsin' : 15, 'ionicstr' : 0.1, 'pbc_dimensions': 0, 'ffID' : 'CHARMM36m', #'ffID' : 'G54A7', #Set the ffinput when using PDB files from simulations #'ffinput': 'GROMOS' # options: GROMOS, AMBER, CHARMM }

I used the PDB 1huc to test it.

Weirdly when I tested it today it worked one time with the CHARMM parameter and then stopped working again.

[pedrishi]

sorry for the delay in the reply, but I had to fix the bug and unfortunately I'm struggled to find the time to address it.

anyways, the bug is fixed. Somewhere along the many versions since CHARMM was introduced in the code, I must have introduced a bug on the internal nomenclature of free CYS in CHARMM. For disclosure this part of the code is not as tested as the GROMOS counterpart. It is in my to-do list though. In fact, in the near future I will recalibrate the pK values for the model compounds for both the CHARMM and GROMOS charge and radii sets. I'll keep this issue open so that I'll notify you once that is done.

Using GROMOS or CHARMM charge and radii sets should yield very similar pKa values. I've calculated both without the titration of THR and SER (ser_thr_titration: False) and you should be able to reproduce this results with the latest version I've just pushed (version 2.2.0). pKas_gromos.txt pKas_charmm.txt

As a final note, unless you have a very specific reason for using CHARMM charges and radii, I'd use PypKa for now with GROMOS.