1. Obtain snapshots from the trajectory by using ~/mopac/traj2xyzs.py 100.
  2. Go to the snapshot folder, follow the instruction to optimize all snapshots using the PM6 method.
  3. Go to the “optresult” folder created during the last step.
  4. Run ~/mopac/sort_conformers.py.
  5. Edit if needed and submit “submission.sh”: sbatch submission.py
  6. The task can finish within 2 minutes. Check the group id for each snapshots in “selected*conformers.csv”. And the selected distinct conformers are copied to the folder named with “selected”.

Note

  • GaussView without non-exchangeable hydrogen. The non-exchangeable hydrogen number is in the slurm file. This is useful to view the structure in GaussView without these hydrogens. Open the structure in GaussView. Right click -> Tools -> Atom Group. Copy the hydrogen number to the first row of Atom Tags, change Display from Show to Hide, uncheck the other rows, click Persistent Visuals and click ok.
  • Sort multiple trajectories for one molecule. If you have two or more trajectories for one molecule and want to pick conformers from these two trajectories together:
    • Either after step 1 and before step 2,  copy all snapshots to the same folder and then perform pm6 optimization.
    • Or after step 3 and before step 4, copy all gjf files in all “optresult” folders to the same folder, and have a file named with “pm6energy.csv” that include all gjf files and their energies similar to the original “optresult” folder.