search

natedey / RINRUS

Residue Interaction Network ResidUe Selector
MIT License
3 stars 0 forks source link

PDB files from MD that are processed through VMD will have water residue IDs written in hexadecimal - crashes RINRUS #38

Closed natedey closed 11 months ago

natedey commented 11 months ago

Hi all, The res id# columns in the pdb format only go up to 4 digits (9999). If you import an AMBER trajectory into VMD and write the pdb, it will start writing water IDs past 9999 as hex notation (x_2710) then we will have res id clashes for 10 straight values, then becomes x_271a. This will crash RINRUS since it is expecting an integer.

This messed up "standard" has been discussed here: https://github.com/openmm/openmm/issues/3501

I don't know if this is a bug that can be fixed. It may be that we have to specify workflows for printing PDBs from MD trajectories that avoid VMD.

natedey commented 11 months ago

Same as issue #36 , Bachega has written a script that can fix common problem of MD waters exceeding 9999 res id#