search

openmm / openmm

OpenMM is a toolkit for molecular simulation using high performance GPU code.
1.42k stars 505 forks source link

Better way to save structure #4570

Open asiomchen opened 5 days ago

asiomchen commented 5 days ago

I'm using OpenMM to build and simulate a protein-ligand system constructed from protein PDB and ligand SDF files.

The results are saved as minimised solvated structure in PDB file and trajectory in XTC format. However I have some problems with this approach, when loaded (for example in MDAnalysis) no connectivity information exists and - when ligand bonds are guessed leading to clashes and bond order issues.

Does some better combination of topology/trajectory format exists to preserve all the information about the connectivity?

peastman commented 5 days ago

The PDB file contains the bonds. MDAnalysis should be loading them.

First check to make sure your PDB actually specifies the bonds in the ligand. Search through it for CONECT records. (Yes, that's spelled correctly. Only one N.) If you don't find any, that suggests a problem in how you created the file.

The other possibility is a problem in how you're loading the files. Can you show your code?