Using pKa-ANI to compute titration states at a given pH?

[sastrys1]

In a recent paper by Gokcan and Isayev, they developed a new model using representation learning built on ANI (a learned way to quickly compute the environment around a given atom) and was able to outperform propka.

I was wondering if you would be interested in me adding pka-ANI as a command line option to assign titration states? This would be useful for me personally so I will probably do it anyway but I wanted to know if it would be useful to the community.

[sastrys1]

For clarity, pka-ANI operates on a PDB file with atom coordinates just like propka does so it seems like plugging in pka-ANI to the existing propka code and adding that as a command line option is not super hard. pka-ANI probably would take more time though.

[sobolevnrm]

Yes, this sounds very useful. Please submit a PR!