def csConsistencyCheck(self):
"""Perform a consistency check of this record against chemspider. Raise a ValidationError on error."""
if not self.custom:
errorList = []
cs = ChemSpider(settings.CHEMSPIDER_TOKEN)
if self.CSID is None or self.CSID is '':
raise ValidationError('No CSID set', 'no_csid')
else:
csCompound = cs.get_compound(self.CSID)
if self.name not in ('', None):
nameResults = cs.simple_search(self.name)
if csCompound not in nameResults:
errorList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid name', code='invalid_inchi'))
else:
self.name = csCompound.common_name
if self.INCHI == '':
self.INCHI = csCompound.stdinchi
elif self.INCHI != csCompound.stdinchi:
errorList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid InChi', code='invalid_inchi'))
if self.smiles == '':
self.smiles = csCompound.smiles
elif self.smiles != csCompound.smiles:
errorList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid smiles string', code='invalid_smiles'))
if self.formula == '':
self.formula = csCompound.molecular_formula
elif self.formula != csCompound.molecular_formula:
errorsList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid formula', code="invalid_formula"))
if len(errorList) > 0:
raise ValidationError(errorList)
def csConsistencyCheck(self):
"""Perform a consistency check of this record against chemspider. Raise a ValidationError on error."""
if not self.custom:
errorList = []
cs = ChemSpider(settings.CHEMSPIDER_TOKEN)
if self.CSID is None or self.CSID is '':
raise ValidationError('No CSID set', 'no_csid')
else:
csCompound = cs.get_compound(self.CSID)
if self.name not in ('', None):
nameResults = cs.simple_search(self.name)
if csCompound not in nameResults:
errorList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid name', code='invalid_inchi'))
else:
self.name = csCompound.common_name
if self.INCHI == '':
self.INCHI = csCompound.stdinchi
elif self.INCHI != csCompound.stdinchi:
errorList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid InChi', code='invalid_inchi'))
if self.smiles == '':
self.smiles = csCompound.smiles
elif self.smiles != csCompound.smiles:
errorList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid smiles string', code='invalid_smiles'))
if self.formula == '':
self.formula = csCompound.molecular_formula
elif self.formula != csCompound.molecular_formula:
errorsList.append(ValidationError(
'A compound was consistency checked and was found to have an invalid formula', code="invalid_formula"))
if len(errorList) > 0:
raise ValidationError(errorList)
def find_matches(matched_in_ChemSpider, massFile_Name):
from chemspipy import ChemSpider
cs = ChemSpider('dfdc677d-e7d3-435b-a74e-bfe6167a3899')
for i in matched_in_ChemSpider.keys():
print i
# intialiaztion
matched_compounds = []
matches = {}
# load mol file info of the product
product_molFile = read_product_molFile(massFile_Name, i)
# for each compound in data base with almost the same mass
for CSID in matched_in_ChemSpider[i]:
# extract the compound's mol file
c = cs.get_compound(CSID)
ChemSpider_compound_mol_info = c.mol_2d
# compare the product's and compound's mol files
is_the_same = compare_two_molFiles(product_molFile,
ChemSpider_compound_mol_info)
# add the compound to the list if it's molfile is the same as the product's
if is_the_same:
matched_compounds.append(CSID)
# if at least one compound found as a match
if matched_compounds != []:
matches.update({i: matched_compounds})
# return the whole matches for products
return matches
def getLongNames(molsDict, pref=4, onlyLettersDigits=False, token="2228d430-a955-416b-b920-14547d28df9e"):
cs = ChemSpider(token)
names = {}
for mol in molsDict:
comp = cs.get_compound(mol[pref:])
nName = comp.common_name.encode("ascii", "ignore")
if onlyLettersDigits:
nName = leftOnlyLettersDigits(nName)
names[mol] = nName
return names
class ChemSp(object):
def __init__(self):
sett = SettingsConstants()
self.key = sett.get('CHEMSPI_KEY')
self.url = sett.get('CHEMSPI_API_URL')
self.cs = ChemSpider(self.key, api_url=self.url)
def get_cmpd(self, csid):
return self.cs.get_compound(csid)
def search(self, query):
print('Connected to ChemSpider API')
print("Searching started")
print("Searching for: " + query)
i = 0
results = []
for result in self.cs.search(query):
if i > 5:
break
print("Compound " + str(i))
formula = str(result.molecular_formula)
csid = str(result.csid)
inchi = result.inchi
name = result.common_name
cas = cirpy.resolve(inchi, 'cas')
iupac_name = cirpy.resolve(inchi, 'iupac_name')
if type(cas) is list:
c_cas = query
sim_cas = difflib.get_close_matches(str(c_cas), cas, 3, 0)
print(sim_cas)
cas_ = sim_cas[0]
else:
cas_ = cas
image = result.image_url
print(image)
i = i + 1
result_line = {'csid': csid, 'name': name, 'iupac_name': iupac_name, 'cas': cas_, 'inchi': inchi, \
'formula': formula, 'image': image}
results.append(result_line)
print("Searching finished")
print(results)
return results
def render_image(self, csid, image_id):
image_png = self.get_cmpd(csid).image
temp_image = '/home/marcin/Dokumenty/projekty/production/Chem/chembase/static/chembase/temp/temp' + image_id + '.png'
with open(temp_image, 'wb+') as destination:
destination.write(image_png)
image_path = '/static/chembase/temp/temp' + image_id + '.png?timestamp=' + str(
datetime.datetime.now())
return image_path
def database_setup():
"""
Download 2D & 3D molecule structure
from ChemSpider server to create a database
"""
from chemspipy import ChemSpider
# compile id list for calling molecules
id_list = get_id()
directory = DATABASE
# make directory database_chemspider/ if needed
if os.path.isdir(directory):
print('Database folder already existed! Aborting... \n '
'Please remove the folder and rerun')
exit()
else:
os.mkdir(directory)
print('downloading..')
os.chdir(directory) # change dir to database_chemspider/
# access API key
cs = ChemSpider('text')
# go through each id
for id_chemspider in id_list:
if os.path.exists(str(id_chemspider) + '_2d.txt'):
# pass if id already exist
print('ID ' + str(id_chemspider) + ' already existed')
continue
# access molecule data
c = cs.get_compound(id_chemspider)
# write 2d coord and bond data
f = open(str(id_chemspider) + '_2d.txt', 'w')
f.write(c.mol_2d)
f.close()
# write 3d coord and bond data
f = open(str(id_chemspider) + '_3d.txt', 'w')
f.write(c.mol_3d)
f.close()
os.chdir('../')
def get_chemspider_structure(csid):
"""
Get a molecular structure from ChemSpider, generate a PDB file of the
structure, and return the name of the PDB file
"""
pdbpath = '{}.pdb'.format(csid)
token = 'a03b1636-afc3-4204-9a2c-ede27680577c' # XXX
cs = ChemSpider(token)
cmpd = cs.get_compound(csid)
conv = ob.OBConversion()
conv.SetInAndOutFormats('mol', 'pdb')
mol = ob.OBMol()
conv.ReadString(mol, cmpd.mol_3d)
mol.AddHydrogens()
with open(pdbpath, 'w') as f:
f.write(conv.WriteString(mol))
return pdbpath
def getChemspiderCompounds(token, list, pref, delim="_", longNames=True, onlyLettersDigits=False):
cs = ChemSpider(token)
names = []
molecules = []
for chsId in list:
comp = cs.get_compound(chsId)
name = pref + delim + str(chsId)
if longNames:
name += delim
sn = comp.common_name.encode("ascii", "ignore")
if onlyLettersDigits:
sn = leftOnlyLettersDigits(sn)
name += sn
# .replace('(', '_').replace(')', '_').replace('[', '_').replace(']', '_').replace(',', '_').replace(' ', '_').replace(';', '_')[:25]
print(name)
smiles = comp.smiles.encode("ascii", "ignore")
mol = Chem.MolFromSmiles(smiles)
mol2 = Chem.AddHs(mol)
molecules.append(mol2)
names.append(name)
return molecules, names
#read back in final list which is the two chem lists combined
#then extract SMILE and MW information using the csid
#boom
df = pd.read_csv("chem_cidList_total.csv")
#make columns for new data
df['MW'] = 0
df['SMILE'] = 0
#write a loop that fills in these variables
i = 0
for index, row in df.iterrows():
csid = row[7]
if csid != 0:
compound = cs.get_compound(csid)
MW = compound.molecular_weight
SMILE = compound.smiles
df.ix[i, 'MW'] = MW
df.ix[i, 'SMILE'] = SMILE
i = i + 1
else:
i = i + 1
#write to csv
df.to_csv('ChemID_MW_SMILES_everything.csv')
class CompoundForm(forms.ModelForm):
"""
A form for users to add compounds to the compound guide.
Forces a check against the chemspider database to ensure no spurious compounds make their way into the compound guide.
"""
CAS_ID = forms.CharField(label='CAS ID', required=False)
"""Adding this field, not in the database, allows users to match compounds to a CAS_ID without us incuring issues for storing them."""
CSID = forms.IntegerField(
label='Chemspider ID',
min_value=1,
error_messages={'required': 'This value must be set or selected'})
"""If the user already knows the right value for this it allows them to skip a step."""
abbrev = forms.CharField(label="Abbreviation", max_length=100)
class Meta:
fields = ('labGroups', 'CSID', 'abbrev', 'name', 'CAS_ID',
'chemicalClasses')
model = Compound
help_texts = {
'name':
'A common or IUPAC name for the compound.',
'CAS_ID':
'The CAS number for the compound. Optional.',
'CSID':
'The Chemspider ID for the compound. If this is not included, a list will be provided for you to choose from.'
}
def __init__(self, user, *args, **kwargs):
"""Overridden version of the init method allows us to place the user's lab groups as a restricted set."""
super(CompoundForm, self).__init__(*args, **kwargs)
self.compound = None
self.chemSpider = ChemSpider(settings.CHEMSPIDER_TOKEN)
self.fields['labGroups'].queryset = user.labgroup_set.all()
if user.labgroup_set.all().exists():
self.fields['labGroups'].empty_label = None
def clean_CSID(self):
"""Check that the CSID is actually a valid id from chemspider."""
searchResults = self.chemSpider.simple_search(
self.cleaned_data['CSID'])
if (len(searchResults) < 1):
raise ValidationError('The CSID you have provided is invalid',
code='invalid_csid')
else:
self.compound = searchResults[0]
return self.cleaned_data['CSID']
def clean(self):
"""Verify that the CSID, CAS_ID (where supplied) and name are consistent."""
self.cleaned_data = super(CompoundForm, self).clean()
if 'labGroups' in self.cleaned_data:
for labGroup in self.cleaned_data.get('labGroups'):
if CompoundGuideEntry.objects.filter(
abbrev=self.cleaned_data.get('abbrev'),
labGroup=labGroup).exclude(
compound=self.instance).exists():
self.add_error(
'abbrev',
'A compound with this abbreviation already exists for the selected labgroup.'
)
if self.cleaned_data.get('name'):
nameResults = self.chemSpider.simple_search(
self.cleaned_data['name'])
if self.cleaned_data.get('CAS_ID') != '':
CAS_IDResults = self.chemSpider.simple_search(
self.cleaned_data['CAS_ID'])
compoundChoices = [
compound for compound in nameResults
if compound in CAS_IDResults
][0:10]
# the CAS_ID always generates a more restrictive set
else:
compoundChoices = nameResults[0:10]
# if the CAS_ID is not supplied, then we just create a subset
# based on the name search alone
if self.compound is None and len(compoundChoices) > 0:
self.fields['CSID'] = forms.ChoiceField(
choices=((choice.csid, choice.common_name)
for choice in compoundChoices),
widget=forms.widgets.RadioSelect)
# in essence, if a CSID was not supplied, but the chemspider
# search returned chemspider results, then we offer those
# results to the user to make a selection.
return self.cleaned_data
elif self.compound is None:
raise ValidationError(
'Your search terms failed to validate against the Chemspider database. Please contact a local administrator.',
code='no_compounds')
else:
if self.compound not in nameResults:
raise ValidationError(
'The name provided was not valid for the CSID provided. Please change the entry, or contact your local administrator.',
code='name_csid_conflict')
elif self.cleaned_data.get(
'CAS_ID') and self.compound not in CAS_IDResults:
raise ValidationError(
'The CAS ID provided is not valid for the CSID provided. Remove, replace, or contact your local administrator.',
'name_cas_id_conflict')
else:
return self.cleaned_data
else:
if self.compound is not None:
# this is probably some of the most horrible code I have
# written, but it is the only way to get this to work - Phil.
data = self.data.copy(
) # because otherwise the query dict is immutable
# replace the data directly, as bad as that is...
data['name'] = self.compound.common_name
# manually input an error message which is less demanding (this
# is actually canonical method)
self._errors['name'] = self.error_class(
['Please review this suggestion'])
self.data = data # override the old data
return self.cleaned_data
def save(self, commit=True):
"""Create (and if appropriate, saves) the compound instance, and adds Inchi and smiles from chemspider."""
try:
self.instance = Compound.objects.get(
CSID=self.cleaned_data['CSID'])
except Compound.DoesNotExist:
pass # Hakuna Matata
compound = super(CompoundForm, self).save(commit=False)
csCompound = self.chemSpider.get_compound(compound.CSID)
compound.INCHI = csCompound.inchi
compound.smiles = csCompound.smiles
compound.formula = csCompound.molecular_formula
if commit:
compound.save()
if 'labGroups' in self.cleaned_data:
for labGroup in self.cleaned_data['labGroups']:
try:
cgEntry = CompoundGuideEntry.objects.get(
labGroup=labGroup,
abbrev=self.cleaned_data['abbrev'])
cgEntry.compound = compound
cgEntry.save()
except CompoundGuideEntry.DoesNotExist:
CompoundGuideEntry.objects.create(
labGroup=labGroup,
abbrev=self.cleaned_data['abbrev'],
compound=compound)
return compound
backup_map = {
'Propyne': 6095,
'R236EA': 71342,
'R245ca': 62827,
'trans-2-Butene': 56442,
'Oxygen': 952,
'Fluorine': 22932,
'Hydrogen': 762,
'Deuterium': 22931,
'HFE143m': 66577,
'SulfurHexafluoride': 16425,
'R114': 13853215
}
# Make sure the key works
c = cs.get_compound(2157)
assert(c.inchikey == 'BSYNRYMUTXBXSQ-UHFFFAOYAW')
for fname in glob.glob('../fluids/*.json'):
with open(fname,'r') as fp:
jj = json.load(fp)
fluid = jj['INFO']['NAME']
def doset(result):
jj['INFO']['INCHI_STRING'] = result.inchi
jj['INFO']['INCHI_KEY'] = result.inchikey
jj['INFO']['CHEMSPIDER_ID'] = result.csid
jj['INFO']['2DPNG_URL'] = result.image_url
jj['INFO']['SMILES'] = result.smiles
backup_map = {
'Propyne': 6095,
'R236EA': 71342,
'R245ca': 62827,
'trans-2-Butene': 56442,
'Oxygen': 952,
'Fluorine': 22932,
'Hydrogen': 762,
'Deuterium': 22931,
'HFE143m': 66577,
'SulfurHexafluoride': 16425,
'R114': 13853215
}
# Make sure the key works
c = cs.get_compound(2157)
assert (c.inchikey == 'BSYNRYMUTXBXSQ-UHFFFAOYAW')
for fname in glob.glob('../fluids/*.json'):
with open(fname, 'r') as fp:
jj = json.load(fp)
fluid = jj['INFO']['NAME']
def doset(result):
jj['INFO']['INCHI_STRING'] = result.inchi
jj['INFO']['INCHI_KEY'] = result.inchikey
jj['INFO']['CHEMSPIDER_ID'] = result.csid
jj['INFO']['2DPNG_URL'] = result.image_url
jj['INFO']['SMILES'] = result.smiles
# read the API key from file and instantiate Chem Spider api client
with open("apikey", "r") as key:
cs = ChemSpider(key.read().strip())
index = 0
# the "startfrom" file persists the ID of the molecule we last requested, check if that exists first
try:
with open("startfrom", "r") as f:
index = int(f.read().strip())
except:
pass
while (True):
try:
compound = cs.get_compound(index)
# if (compound.common_name):
# print(compound.common_name)
# save the image with the ID as the name
with open("images/" + str(index) + ".png", "wb") as f:
f.write(compound.image)
print(".", end="", flush=True)
except ChemSpiPyServerError as err:
# skip over invalid IDs
if ("Invalid ID" in err.args[0]):
print("x", end="", flush=True)
else:
print("ERROR", index)
class CompoundForm(forms.ModelForm):
"""
A form for users to add compounds to the compound guide.
Forces a check against the chemspider database to ensure no spurious compounds make their way into the compound guide.
"""
CAS_ID = forms.CharField(label='CAS ID', required=False)
"""Adding this field, not in the database, allows users to match compounds to a CAS_ID without us incuring issues for storing them."""
CSID = forms.IntegerField(label='Chemspider ID', min_value=1, error_messages={
'required': 'This value must be set or selected'})
"""If the user already knows the right value for this it allows them to skip a step."""
class Meta:
fields = ('labGroup', 'abbrev', 'CSID',
'name', 'CAS_ID', 'chemicalClasses')
model = Compound
help_texts = {
'abbrev': 'A local abbreviation by which the compound is known.',
'name': 'A common or IUPAC name for the compound.',
'CAS_ID': 'The CAS number for the compound. Optional.',
'CSID': 'The Chemspider ID for the compound. If this is not included, a list will be provided for you to choose from.'
}
def __init__(self, user, *args, **kwargs):
"""Overridden version of the init method allows us to place the user's lab groups as a restricted set."""
super(CompoundForm, self).__init__(*args, **kwargs)
self.compound = None
self.chemSpider = ChemSpider(settings.CHEMSPIDER_TOKEN)
self.fields['labGroup'].queryset = user.labgroup_set.all()
if user.labgroup_set.all().exists():
self.fields['labGroup'].empty_label = None
def clean_CSID(self):
"""Check that the CSID is actually a valid id from chemspider."""
searchResults = self.chemSpider.simple_search(
self.cleaned_data['CSID'])
if(len(searchResults) < 1):
raise ValidationError(
'The CSID you have provided is invalid', code='invalid_csid')
else:
self.compound = searchResults[0]
return self.cleaned_data['CSID']
def clean(self):
"""Verify that the CSID, CAS_ID (where supplied) and name are consistent."""
self.cleaned_data = super(CompoundForm, self).clean()
if self.cleaned_data.get('name'):
nameResults = self.chemSpider.simple_search(
self.cleaned_data['name'])
if self.cleaned_data.get('CAS_ID') != '':
CAS_IDResults = self.chemSpider.simple_search(
self.cleaned_data['CAS_ID'])
compoundChoices = [
compound for compound in nameResults if compound in CAS_IDResults][0:10]
# the CAS_ID always generates a more restrictive set
else:
compoundChoices = nameResults[0:10]
# if the CAS_ID is not supplied, then we just create a subset
# based on the name search alone
if self.compound is None and len(compoundChoices) > 0:
self.fields['CSID'] = forms.ChoiceField(choices=(
(choice.csid, choice.common_name) for choice in compoundChoices), widget=forms.widgets.RadioSelect)
# in essence, if a CSID was not supplied, but the chemspider
# search returned chemspider results, then we offer those
# results to the user to make a selection.
return self.cleaned_data
elif self.compound is None:
raise ValidationError(
'Your search terms failed to validate against the Chemspider database. Please contact a local administrator.', code='no_compounds')
else:
if self.compound not in nameResults:
raise ValidationError(
'The name provided was not valid for the CSID provided. Please change the entry, or contact your local administrator.', code='name_csid_conflict')
elif self.cleaned_data.get('CAS_ID') and self.compound not in CAS_IDResults:
raise ValidationError(
'The CAS ID provided is not valid for the CSID provided. Remove, replace, or contact your local administrator.', 'name_cas_id_conflict')
else:
return self.cleaned_data
else:
if self.compound is not None:
# this is probably some of the most horrible code I have
# written, but it is the only way to get this to work - Phil.
data = self.data.copy() # because otherwise the query dict is immutable
# replace the data directly, as bad as that is...
data['name'] = self.compound.common_name
# manually input an error message which is less demanding (this
# is actually canonical method)
self._errors['name'] = self.error_class(
['Please review this suggestion'])
self.data = data # override the old data
return self.cleaned_data
def save(self, commit=True):
"""Create (and if appropriate, saves) the compound instance, and adds Inchi and smiles from chemspider."""
compound = super(CompoundForm, self).save(commit=False)
csCompound = self.chemSpider.get_compound(compound.CSID)
compound.INCHI = csCompound.inchi
compound.smiles = csCompound.smiles
compound.formula = csCompound.molecular_formula
if commit:
compound.save()
self.save_m2m()
return compound
# print(tokenchoice())
if os.path.isfile('chemspiderdb.json'):
spiderjsonfileid = []
with open('chemspiderdb.json', 'r') as jsonfile:
for f in jsonfile.readlines():
the_dict = json.loads(f)
spiderjsonfileid.append(the_dict['_id'])
# print(spiderjsonfileid)
for csid in csids:
# cskey = random.choice(cs_security_key)
cs = ChemSpider(tokenchoice())
if csid in spiderjsonfileid:
print('{0} has been in the file'.format(str(csid)))
continue
compound = cs.get_compound(csid)
try:
doc = {'_id': int(compound.csid), 'common_name': compound.common_name}
sleep(random.uniform(0.2, 0.5))
doc['molecular_weight'] = compound.molecular_weight
sleep(random.uniform(0, 0.5))
doc['molecular_formula'] = compound.molecular_formula
doc['stdinchi'] = compound.stdinchi
sleep(random.uniform(0.1, 0.5))
doc['stdinchikey'] = compound.stdinchikey
doc['smiles'] = compound.smiles
# sleep(random.uniform(1, 1.1))
with open('chemspiderdb.json', 'a') as jsonfile:
json.dump(doc, jsonfile)
jsonfile.write('\n')
print("{0} Thanks! ".format(str(csid)))
cid = pcpCmpd.cid
items["pubchem_cid"] = cid
else:
cid = items["pubchem_cid"]
pcpCmpd = pcp.Compound.from_cid(cid)
# ChemSpider
if not "chemspider_id" in items:
results = cs.simple_search(inchikey)
results.sort()
csCmpd = results[0]
csid = csCmpd.csid
items["chemspider_id"] = csid
items["name"] = csCmpd.common_name.lower()
else:
csid = items["chemspider_id"]
csCmpd = cs.get_compound(csid)
# NIH resolver
nihCmpd = cirpy.Molecule(inchikey, ['stdinchikey'])
if not "iupac_name" in items:
name = nihCmpd.iupac_name # try NIH resolver
if name is None: # try PubChem
name = pcpCmpd.iupac_name
if type(name) is ListType:
name = name[0]
if name is None:
name = items["name"] # chemspider common name
items["iupac_name"] = name.lower()
if not "molwt" in items:
items["molwt"] = pcpCmpd.molecular_weight
def getMolecularWeight(self):
cs = ChemSpider(settings.CHEMSPIDER_TOKEN)
if self.CSID:
csCompound = cs.get_compound(self.CSID)
return Decimal(csCompound.molecular_weight)
