You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Find a summary of my experiments with xtal minimization here .
Description
Preparation
This is for subset of 293 COD molecules (see here). I attempted to built xtal supercells by replicating the unitcell in each cell direction so that any cell axis is at least 2.5 times the real space cutoff (.9 nm). This was succesfull for 232 of those molecules. Note that for each successful supercell, there is also valid rdkit object. For each supercell, I attempted to build an openmm system with Sage and Parsley FFs. This worked for 178 systems.
Minimization
Minimization worked (i.e. didn't crash for any reason and generated physically meaningful output) fine for 165 xtals. I experimented with three minimization approaches:
1.) Minimize box and position vectors simultaneously
2.) Minimize box and position vectors in alternating fashion
3.) First 1.), then 2.)
Code to reproduce these things is here. I did not upload CIF files, they can be found here). Pdb files as well as system/state xml files are too large. If there is interest in them, I can provide them through bitbucket or something similar.
Find a summary of my experiments with xtal minimization here .
Description
Preparation
This is for subset of 293 COD molecules (see here). I attempted to built xtal supercells by replicating the unitcell in each cell direction so that any cell axis is at least 2.5 times the real space cutoff (.9 nm). This was succesfull for 232 of those molecules. Note that for each successful supercell, there is also valid rdkit object. For each supercell, I attempted to build an openmm system with Sage and Parsley FFs. This worked for 178 systems.
Minimization
Minimization worked (i.e. didn't crash for any reason and generated physically meaningful output) fine for 165 xtals. I experimented with three minimization approaches:
1.) Minimize box and position vectors simultaneously
2.) Minimize box and position vectors in alternating fashion
3.) First 1.), then 2.)
Code to reproduce these things is here. I did not upload CIF files, they can be found here). Pdb files as well as system/state xml files are too large. If there is interest in them, I can provide them through bitbucket or something similar.
@mrshirts
The text was updated successfully, but these errors were encountered: