Additional flags for pkaani to fine tune computation

[lpravda]

Hi guys again,

I've noticed that the pkaani modifies the structure (minimization, atom addition, ligand removal), I wonder if you could add a mode to make pkaani computation more configurable and flexible. What I'm thinking of is:

• default - present behaviour
• interpret structure as is (no modifications to the structure)
• just run structure minimisation.
• perhaps any other combination that makes sense for you?

Would you consider a PR with the modified behaviour?

[HGokcan]

Hello

If you would like to, you are welcome to make a pull request.

Just to be clear on the modifications you mentioned; they are performed to make sure the input structure for pkaani models are correct and have the same configuration as the training. Because we are using ANI, the missing atoms (like H atoms) should be added to structure. The missing atoms will cause generation of wrong descriptors and wrong predictions. Additionally, protonation states should be same as in the training data (for example ASP should be negatively charged not neutral). As for the removal of HETATM entries, pkaani was not trained (or tested) with organic molecules. If organic molecules are wanted to be kept regardless, the protonation state of this molecule should be correct (H atoms should have been added correctly), and if minimization is going to be performed (which is to ensure no clash exists after missing atoms are added) it will require organic molecule parametrization prior to minimization.

[lpravda]

Hey, yeah that makes perfect sense. I've got data coming from a pipeline with all the above-mentioned issues sorted (protonation, incompleteness etc.). So the only thing I really need to do is to remove all HET atoms before the computation. But I'll bear all your points in mind. Thank you for that!