transformation = Transformation(
run_id=run_id,
xchem_fragment_id=xchem_fragment_id,
initial_microstate=CompoundMicrostate(
compound_id=get_compound_id(reference_microstate.microstate_id),
microstate_id=reference_microstate.microstate_id),
final_microstate=CompoundMicrostate(
compound_id=get_compound_id(microstate.microstate_id),
microstate_id=microstate.microstate_id))
transformations.append(transformation)
run_id += 1
# Compile compound series
from fah_xchem.schema import CompoundSeries
compound_series = CompoundSeries(metadata=series_metadata,
compounds=list(compounds.values()),
transformations=transformations)
# Write JSON
print(f'Writing JSON to {json_filename}')
if '.bz2' in json_filename:
import bz2
with bz2.open(json_filename, "wt") as f:
f.write(compound_series.json())
elif '.gz' in json_filename:
import gzip
with gzip.open(json_filename, "wt") as f:
f.write(compound_series.json())
else:
with open(json_filename, "wt") as f:
f.write(compound_series.json())
nmols_with_assay_data += 1
print(
f'Found assay data for {nmols_with_assay_data} / {len(oemols)} molecules'
)
if __name__ == '__main__':
new_json_filename = 'json/sprint-5-x12073-monomer-neutral-simulated.json'
from fah_xchem.schema import CompoundSeries, Transformation, CompoundMicrostate, CompoundMetadata, Microstate, Compound
# Read source JSON
source_json_filename = 'json/sprint-5-x12073-monomer-neutral.json'
import json
with open(source_json_filename, "r") as infile:
compound_series = CompoundSeries.parse_obj(json.loads(infile.read()))
# Read all microstates
import os
project = '13429'
from glob import glob
runs = glob(f'{project}/RUNS/RUN*')
nruns = len(runs)
from rich.progress import track
runs = list()
oemols = dict()
for run in track(
range(nruns),
description='Reading initial and final molecules from all RUNs...'
):
try:
return
# work out which ligand pair to run
def load_json(filename):
"""
Load a JSON file that may be .bz2 or .gz compressed
"""
if '.bz2' in filename:
import bz2
with bz2.open(filename, 'rt') as infile:
return json.load(infile)
elif '.gz' in filename:
import gzip
with gzip.open(filename, 'rt') as infile:
return json.load(infile)
else:
with open(filename, 'rt') as infile:
return json.load(infile)
# Sprint 5
json_filename = 'json/sprint-5-x12073-monomer-neutral.json'
microstate_sdf_filename = 'docked/sprint-5-microstates-x12073-sorted.sdf' # TODO: Encapsulate this in JSON file as source_sdf_filename and source_molecule_index?
json_data = load_json(json_filename)
from fah_xchem.schema import CompoundSeries
compound_series = CompoundSeries.parse_obj(json_data)
# Process 0-indexed transformation index
transformation_index = int(sys.argv[1])
run_relative_perturbation(compound_series, transformation_index, microstate_sdf_filename)
mol = oechem.OEGraphMol()
oechem.OESmilesToMol(mol, smiles)
return oechem.OEMolToSmiles(mol)
if __name__ == '__main__':
from fah_xchem.schema import CompoundSeries, Transformation, CompoundMicrostate
new_json_filename = 'json/sprint-5-x12073-monomer-neutral-corrected.json'
from fah_xchem.schema import CompoundSeries, Transformation, CompoundMicrostate
# Read source JSON
source_json_filename = 'json/sprint-5-x12073-monomer-neutral.json'
import json
with open(source_json_filename, "r") as infile:
compound_series = CompoundSeries.parse_obj(json.loads(infile.read()))
# Index microstates
compound_microstates = dict()
for compound in compound_series.compounds:
for microstate in compound.microstates:
compound_microstate = CompoundMicrostate(
compound_id=compound.metadata.compound_id,
microstate_id=microstate.microstate_id)
compound_microstates[canonical_smiles(
microstate.smiles)] = compound_microstate
xchem_fragment_id = compound_series.transformations[0].xchem_fragment_id
# Scan RUNs
print('Scanning RUNs...')