Repository for the paper Genetic algorithm-based re-optimization of the Schrock catalyst for dinitrogen fixation. The GA implementation is based on the GB_GA.
This codebase is slightly outdated for the work we do on catalyst design with GAs. For the interested, this is a more robust codebase we are currently using for catalyst design: catalystGA. And for other updated work on the Schrock catalyst see: genetic_algorithm_for_nitrogen_fixation
For simple use of the GA install with conda install the env file.
conda env create --file environment.yml
To run the GA activate the relevant environment and run the following for a quick run on a local installation:
python GA_schrock.py --supress_amines --debug --average_size 10 --size_stdev 2 --scoring_func rdkit_embed_scoring_NH3toN2 --cpus_per_task 2 --population_size 3 --mating_pool_size 4
A list of possible commandline arguments.
| Arg | Description |
|---|---|
-h or --help |
Prints help message. |
--population_size |
Sets the size of the population pool. |
--mating_pool_size |
Sets the size of the mating pool. |
--n_confs |
Sets how many conformers to generate for each molecule. |
--n_tries |
Sets how many times the GA is restarted. Can be used to run multiple GA runs in a single submission. |
--cpus_per_task |
How many cores to use for each scoring job. |
--RMS_thresh |
RMS cutoff value for RDKit conformer embedding. |
--generations |
How many evolution cycles of the population is performed. |
--mutation_rate |
Decides the probability of performing a mutation operation instead of crossover. |
--sa_screening |
Decides if synthetic accessibility score is enabled. Highly recommended to turn this on. |
--file_name |
Path to the database extract to create starting population. |
--output_dir |
Sets output directory for all files generated during generations. |
--timeout |
How many minutes each slurm job is allowed to run. |
--debug |
If set the starting population is a set of 4 small molecules that can run fast locally. Used for debugging. |
--ga_scoring |
If set, removes all higher energy conformers in GA. |
--supress_amines |
Supress amine heavy molecules by converting any primary amines to hydrogen in generations. |
--method |
Which gfn method to use. |
--energy_cutoff |
Sets energy cutoff on the conformer filtering. |
--bond_opt |
Decides if a final Mo-N bond optimization is performed during scoring. |
--cleanup |
If enabled, all scoring files are removed after scoring. Only the optimized structures and their energies are saved. |
--scoring_func |
Which scoring function to use. |
--opt |
Set optimization convergence criteria for xTB. |
--gbsa |
Which type of solvent to use for xTB. |
--input |
Name of input control file created for xTB |
--average_size |
Average number of atoms in molecules resulting from crossover. |
--size-stdev |
STD of crossover size distribution for molecules. |
Magnus Strandgaard1 Julius Seumer1 Jan H. Jensen1
1 Department of Chemistry, University of Copenhagen, 2100 Copenhagen Ø, Denmark.
For any questions regarding the code, please do not hesitate to contact me at : [email protected].