Getting error with PDB 1A6W about propka

[amashaly]

Any solution to this error yet?

[dspencer12]

My testing indicates that this issue occurs when atoms are missing in the PDB file (i.e. not resolved by the crystallography experiment). Currently I am working around this issue by using Modeller to fill the missing atom positions.

[FloLangenfeld]

Had the same issue with pdb 1zgl (chains P, R, T and V). Using Modeller solved the issue, indeed. In my case, only OXT atoms at the C-ter residues were missing (which is a quite common feature in X-ray structures).

IMO, the end-user should not be forced to use Modeller, but just be informed that the input structure has something wrong. And it would avoid to add a dependency to propka. A message to the stdout / log should be enough to warn the user that atoms may be missing in the input structure.