Locate TS based on RMSD-PP method
To run a TS search
- Create dataframes and save them as .pkl to store the results of succesful runs (
success.pkl
) and failed runs (fail.pkl
) - Create folder to run the calculations (e.g.
mkdir test_run
) - The following files should be present in the created folder
path.inp
containing parameters defining the RMSD-PP runreaction.xyz
containing the geometry of the reactantproduct.xyz
containing the geometry of the product
The run can now be initiated from within the test_run
folder by:
rmsd_pp_ts.py reactant.xyz product.xyz success.pkl fail.pkl
The procedure relies on external programs to do the quantum chemical calculations. In particular, there are calls to four different submit scripts (submitting xTB and Gaussian16 calculations) within rmsd_pp_ts.py:
xyz2mol
is used to calculate adjacancy matrices and SMILES from .xyz files
https://github.com/jensengroup/xyz2mol
Submits an RMSD-PP GFN2-xTB calculation for
- reactant structure: reactant.xyz
- product structure: product.xyz
- push value: k_push
- pull value: k_pull
- alpha value (width of Gaussian biasing potential): alp
The file path.inp
must be present in the directory
When called separately, the usage is as follows:
submit_xtb_path reactant.xyz product.xyz k_push k_pull alp
Submits all .xyz files in the current directory to single point energy calculations using the xTB program
The jobs are submitted to slurm in batches
When called separately, the usage is as follows:
Go to the directory with the .xyz files to be submitted and call
submit_batches_xtb
Works as submit_batches_xtb
except that the single point energy calculations are done using Gaussian 16, and instead of .xyz files it submits all gaussian input files (.com file) in the present directory.
Submits a Gaussian16 calculation with the type of calculation, structure and method stated in the input .com file
When used separately, the usage is as follows:
submit_gaus16 input.com
The four submit scripts used can be found in submit_scripts