Révision f3004731 modules/calculation.py
| b/modules/calculation.py | ||
|---|---|---|
| 1 | 1 |
import os |
| 2 | 2 |
import logging |
| 3 | 3 |
|
| 4 |
from rdkit.Chem import AllChem as Chem |
|
| 5 |
|
|
| 6 | 4 |
logger = logging.getLogger('DockOnSurf')
|
| 7 | 5 |
|
| 8 | 6 |
|
| 9 |
def prep_iso(inp_file, mol: Chem.rdchem.Mol, exemplars): |
|
| 10 |
"""Prepares the directories to run the calculations |
|
| 11 |
|
|
| 12 |
@param inp_file: Input file of the code to run the calculations with. |
|
| 13 |
@param mol: rdkit mol object of the relevant molecule. |
|
| 14 |
@param exemplars: list of conformer indices selected as representative. |
|
| 15 |
@return: None |
|
| 7 |
def create_bak_calc_dir(run_type): |
|
| 8 |
"""Checks if calculations directory already exists, backs it up if so and |
|
| 9 |
creates an empty one. |
|
| 10 |
@param run_type: Type of calculation. 'isolated', 'screening' or |
|
| 11 |
'refinement' |
|
| 16 | 12 |
""" |
| 17 |
from shutil import copy |
|
| 18 |
|
|
| 19 |
# Checking if 'isolated' directory already exists and backing it up if so. |
|
| 20 |
dir_name = 'isolated' |
|
| 13 |
dir_name = run_type |
|
| 21 | 14 |
bak_num = 0 |
| 22 | 15 |
while dir_name in os.listdir("."):
|
| 23 | 16 |
bak_num += 1 |
| 24 | 17 |
dir_name = dir_name.split(".")[0] + f".bak{bak_num}"
|
| 25 | 18 |
if bak_num > 0: |
| 26 |
os.rename('isolated', dir_name)
|
|
| 27 |
logger.warning("'isolated' directory already present. Moved former "
|
|
| 19 |
os.rename(run_type, dir_name)
|
|
| 20 |
logger.warning(f"'{run_type}' directory already present. Moved former "
|
|
| 28 | 21 |
f"directory to {dir_name}")
|
| 29 |
os.mkdir('isolated')
|
|
| 22 |
os.mkdir(run_type)
|
|
| 30 | 23 |
|
| 31 |
# Creating and setting up directories for every conformer calculation. |
|
| 32 |
for i, conf in enumerate(exemplars): |
|
| 33 |
os.mkdir(f'isolated/conf_{i}')
|
|
| 34 |
copy(inp_file, f'isolated/conf_{i}/')
|
|
| 35 |
Chem.MolToXYZFile(mol, f'isolated/conf_{i}/coord.xyz', confId=int(conf))
|
|
| 36 | 24 |
|
| 25 |
def prep_cp2k(inp_file, run_type, atms_list): |
|
| 26 |
"""Prepares the directories to run isolated calculation with CP2K. |
|
| 37 | 27 |
|
| 38 |
def run_calc(run_type, inp_vars, **kwargs): |
|
| 28 |
@param inp_file: CP2K Input file to run the calculations with. |
|
| 29 |
@param run_type: Type of calculation. 'isolated', 'screening' or |
|
| 30 |
'refinement' |
|
| 31 |
@param atms_list: list of ase.Atoms objects to run the calculation of. |
|
| 32 |
@return: None |
|
| 33 |
""" |
|
| 34 |
from shutil import copy |
|
| 35 |
import ase.io |
|
| 36 |
from pycp2k import CP2K |
|
| 37 |
cp2k = CP2K() |
|
| 38 |
cp2k.parse(inp_file) |
|
| 39 |
force_eval = cp2k.CP2K_INPUT.FORCE_EVAL_list[0] |
|
| 40 |
coord_file = force_eval.SUBSYS.TOPOLOGY.Coord_file_name |
|
| 41 |
|
|
| 42 |
# Creating and setting up directories for every atoms configuration. |
|
| 43 |
for i, conf in enumerate(atms_list): |
|
| 44 |
os.mkdir(f'{run_type}/conf_{i}')
|
|
| 45 |
copy(inp_file, f'{run_type}/conf_{i}/')
|
|
| 46 |
ase.io.write(f'{run_type}/conf_{i}/{coord_file}', conf)
|
|
| 47 |
|
|
| 48 |
|
|
| 49 |
def run_calc(run_type, inp_vars, atms_list): |
|
| 39 | 50 |
"""Directs the calculation run according to the provided arguments. |
| 40 | 51 |
|
| 41 |
@param run_type: Type of calculation. Isolated, screening or refinement |
|
| 52 |
@param run_type: Type of calculation. 'isolated', 'screening' or |
|
| 53 |
'refinement' |
|
| 42 | 54 |
@param inp_vars: Calculation parameters from input file. |
| 43 |
@type kwargs: keyword arguments relevant to the specified run_type: |
|
| 44 |
for isolated: |
|
| 45 |
confs: rdkit mol object of the relevant molecule containing the |
|
| 46 |
conformers |
|
| 47 |
exemplars: list of conformer indices that are exemplars for every |
|
| 48 |
cluster. |
|
| 49 |
for screening: |
|
| 55 |
@param atms_list: List of ase.Atoms objects containing the sets of atoms |
|
| 56 |
aimed to run the calculations of. |
|
| 50 | 57 |
""" |
| 51 | 58 |
run_types = ['isolated', 'screening', 'refinement'] |
| 52 | 59 |
run_type_err = f"'run_type' should be one of the following: {run_types}"
|
| ... | ... | |
| 56 | 63 |
|
| 57 | 64 |
logger.info(f"Running {run_type} calculation with {inp_vars['code']} on "
|
| 58 | 65 |
f"{inp_vars['batch_q_sys']}")
|
| 59 |
|
|
| 66 |
create_bak_calc_dir(run_type) |
|
| 60 | 67 |
if run_type == 'isolated': |
| 61 |
key_err = "If 'run_type' is 'isolated', 'run_calc' needs the following"\ |
|
| 62 |
"arguments: 'inp_file', 'confs' and 'exemplars'." |
|
| 63 |
err = False |
|
| 64 |
try: |
|
| 65 |
confs = kwargs['confs'] |
|
| 66 |
exemplars = kwargs['exemplars'] |
|
| 67 |
except KeyError as e: |
|
| 68 |
logger.error(key_err) |
|
| 69 |
err = e |
|
| 70 |
else: |
|
| 71 |
err = False |
|
| 72 |
finally: |
|
| 73 |
if isinstance(err, BaseException): |
|
| 74 |
raise err |
|
| 75 |
|
|
| 76 |
prep_iso(inp_vars['isol_inp_file'], confs, exemplars) |
|
| 68 |
if inp_vars['code'] == 'cp2k': |
|
| 69 |
prep_cp2k(inp_vars['isol_inp_file'], run_type, atms_list) |
|
Formats disponibles : Unified diff