def search_cocrystals(filter_solvents=True):
'''
Search the whole CSD for structures that contain two different molecules
with the specific settings
'''
start_time = time.clock()
csd = MoleculeReader('CSD')
entry_reader = EntryReader('CSD')
settings = search.Search.Settings()
settings.only_organic = True
settings.not_polymeric = True
settings.has_3d_coordinates = True
settings.no_disorder = True
settings.no_errors = True
settings.no_ions = True
settings.no_metals = True
pairs=[]
for entry in csd:
#if len(pairs)==100:
# break
if settings.test(entry):
mol = csd.molecule(entry.identifier)
mol.normalise_labels()
smi= mol.smiles
if smi != None:
smi = smi.split('.')
# We make sure that the structure consist of two different molecules
if len(Remove(smi)) == 2:
pairs.append(mol.identifier)
# clean the list from solvents
if filter_solvents:
print('Solvates and hydrates will be removed')
solvates=[]
name_dict={}
for mol1 in pairs:
mol = csd.molecule(mol1)
e=entry_reader.entry(mol1)
name_dict[mol1]=e.chemical_name
for i in range(0, (len(mol.components))):
if mol.components[i].smiles in clean_smiles.SOLVENT_SMILES:
solvates.append(mol.identifier)
solvates = Remove(solvates)
final_cocrystals = [x for x in pairs if x not in solvates]
#print(name_dict)
else:
final_cocrystals=pairs
# Clean the list from polymorphs
cocrystals = remove_polymorphs(final_cocrystals)
#print the time
end_time = time.clock()
name=[]
name= [name_dict[i] for i in cocrystals]
cocrystals_data= pd.concat([pd.DataFrame(cocrystals, columns=['csd_id']), pd.DataFrame(name, columns=['name'])], axis=1)
cocrystals_data=cocrystals_data.dropna(axis=0)
dataset_cocrystals = cocrystals_data[~cocrystals_data.name.str.contains("solvate")]
dataset_cocrystals = dataset_cocrystals[~dataset_cocrystals.name.str.contains("clathrate")]
print(end_time-start_time)
dataset_cocrystals.to_csv('new_all_cocrystals.csv',index=False)
return cocrystals
def get_entry(identifier, database="CSD"):
"""
input an identifier as a string and get the
ccdc.entry.Entry object
"""
csd_reader = EntryReader(database)
entry = csd_reader.entry(identifier)
return entry
entries = list(df.refcode)
from ccdc.search import TextNumericSearch
data = []
# for e in entries:
# query = TextNumericSearch()
# query.add_all_identifiers(e)
# hits = query.search()
# data.append(hits[0].entry.publication.doi)
# from pprint import pprint
#
# print len(data)
# print len(set(data))
from ccdc.diagram import DiagramGenerator
from ccdc.io import EntryReader
diagram_generator = DiagramGenerator()
diagram_generator.settings.font_size = 12
diagram_generator.settings.line_width = 1.6
diagram_generator.settings.image_width = 500
diagram_generator.settings.image_height = 500
csd_reader = EntryReader('CSD')
mols = set([csd_reader.entry(m) for m in entries])
for i, e in enumerate(mols):
img = diagram_generator.image(e)
img.save("hit{}.png".format(i))
