def __init__(self):
"""
Initializes all the object variables
"""
# Reaction Dataframe
self.reactions_dataframe = None
# Reactant Dataframe
self.species_df = None
# Unique Reactants Dictionary
self.unique_species_dict = None
# Creating a transator for cleaning individual reactants off non-familiar characters
self.translator = str.maketrans("Î", "α", "±€™") # Argument style
# (# intab,outtab,character string that should be mapped to None)
# Autheticating ChemSpider API using the token
self.security_token = "99c9f388-12be-4b22-8f83-00b6f1e2d7d0" # Maneet's token
self.cs = ChemSpider(
self.security_token,
user_agent="StudentResearcher, ChemSpiPy 1.0.5, Python 3.6")
print('--Populator Initialized--')
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 get_image_url(self):
md = jsonpickle.decode(self.metadata)
if 'csid' in md:
# If this doc already has a csid, make the url
return 'http://www.chemspider.com/ImagesHandler.ashx?id=' + str(
self.csid)
elif 'InChIKey' in md or 'inchikey' in md:
# If it doesnt but it does have an InChIKey get the csid and make the image url
# this code doesn't work...due to an upgrade in chemspider
# if you want images, get the mol
from chemspipy import ChemSpider
cs = ChemSpider(settings.CHEMSPIDER_APIKEY)
ikey = md.get('InChIKey', md.get('inchikey'))
results = cs.search(ikey)
if results:
# Return the image_url and also save the csid
csid = results[0].csid
md['csid'] = csid
self.metadata = jsonpickle.encode(md)
self.save()
return results[0].image_url
else:
return None
else:
# If it has neither, no image!
return None
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 clean_name(self):
chemSpider = ChemSpider(settings.CHEMSPIDER_TOKEN)
nameResults = chemSpider.simple_search(self.cleaned_data['name'])
if self.instance.CSID not in (nameResult.csid for nameResult in nameResults):
raise ValidationError("That name is not a known synonym for this compound")
else:
return self.cleaned_data['name']
def set_and_initialize_token(self, input_token):
"""
Stores you ChemSpider security token as an object attribute and Associate your token to the ChemSpider api
:param input_token: your security token (for ChemSpider)
:return: None
"""
self.security_token = input_token
self.cs = ChemSpider(self.security_token)
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_name(self):
"""Check the name is a valid synonym."""
chemSpider = ChemSpider(settings.CHEMSPIDER_TOKEN)
nameResults = chemSpider.simple_search(self.cleaned_data['name'])
if self.instance.CSID not in (nameResult.csid
for nameResult in nameResults):
raise ValidationError(
"That name is not a known synonym for this compound")
else:
return self.cleaned_data['name']
def search_by_mass(mass, margine):
# pip install chemspipy
from chemspipy import ChemSpider
# register to generate a sequrity code
cs = ChemSpider('dfdc677d-e7d3-435b-a74e-bfe6167a3899')
# search the data base
CSIDs = []
for result in cs.simple_search_by_mass(mass, margine):
CSIDs.append(result.csid)
return CSIDs
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 fromCsv(self, fileName, labGroup=None):
"""Read a CSV into the creating objects, returning a list of compounds which have not yet been saved.
This assumes that the uploaded csv will have headers which map to the names of the fields and that compound classes are
stored as comma separated lists of the chemicalClass LABEL only.
Each compound will perform a chemspider-based consistency check on the information it has been created with to ensure
information is consistent- this throws an ValidationError if it is not.
"""
if labGroup is None and hasattr(self, 'instance'):
# we presume that if this is being called without a labgroup that's because this manager belongs to a lab group
labGroup = self.instance
compoundsList = []
cs = ChemSpider(settings.CHEMSPIDER_TOKEN)
with open(fileName) as f:
reader = csv.DictReader(f, restkey='restKey')
rowCount = 0
errors = []
for row in reader:
try:
rowCount += 1
if 'chemicalClasses' in row:
classes = (c.strip() for c in row['chemicalClasses'].split(','))
chemicalClasses = []
for c in classes:
chemicalClass, created = ChemicalClass.objects.get_or_create(label=c)
chemicalClasses.append(chemicalClass)
if row.get('CAS') not in ('', None) and row.get('CSID') in ('', None):
CASResults = cs.simple_search(row['CAS'])
if len(CASResults) < 1:
errors.append(ValidationError('CAS Number returned no results from ChemSpider on row %(rowCount)d of uploaded csv.', params={'rowCount': rowCount}))
elif len(CASResults) == 1:
row['CSID'] = CASResults[0].csid # a little hacky, but it gets the job done
else:
errors.append(ValidationError('CAS number returns more than one ChemSpider ID on row %(rowCount)d of uploaded csv.', params={'rowCount': rowCount}))
elif row.get('CSID') in ('', None):
errors.append(ValidationError('No CSID provided on row %(rowCount)d of uploaded csv.', params={'rowCount': rowCount}))
kwargs = {}
kwargs['CSID'] = row.get('CSID')
kwargs['abbrev'] = row.get('abbrev')
kwargs['smiles'] = row.get('smiles')
kwargs['name'] = row.get('name')
kwargs['INCHI'] = row.get('INCHI')
compound = Compound(labGroup=labGroup, **kwargs)
for chemicalClass in chemicalClasses:
compound.lazyChemicalClasses.append(chemicalClass)
compoundsList.append(compound)
except ValidationError as e:
for message in e.messages:
errors.append(ValidationError(message + ' on row %(rowCount)d of uploaded csv', params={'rowCount': rowCount}))
if len(errors) > 0:
raise ValidationError(errors)
return compoundsList
def find_common_name(inchikey, formula):
# Try to find the common name for the compound, if not use the formula.
name = formula
if chemspikey:
cs = ChemSpider(chemspikey)
if (len(inchikey) > 0):
result = cs.search(inchikey)
if (len(result) == 1):
name = result[0].common_name
return name
def find_common_name(inchikey):
# Try to find the common name for the compound, if not, return None.
name = None
if chemspikey:
cs = ChemSpider(chemspikey)
if (len(inchikey) > 0):
result = cs.search(inchikey)
if (len(result) == 1):
name = result[0].common_name
return name
def structure_url(self):
from chemspipy import ChemSpider
try:
cs_key = settings.CHEMSPIDER_KEY
except AttributeError:
url = 'http://discovermagazine.com/~/media/Images/Zen%20Photo/N/nanoputian/3487.gif'
else:
cs = ChemSpider(cs_key)
IUPAC = self.name
search_results = cs.simple_search(IUPAC)
try:
url = search_results[0].image_url
except IndexError:
url = ""
return url
def structure_url(self): from chemspipy import ChemSpider try: cs_key = settings.CHEMSPIDER_KEY except AttributeError: url = 'http://discovermagazine.com/~/media/Images/Zen%20Photo/N/nanoputian/3487.gif' else: cs = ChemSpider(cs_key) IUPAC = self.name search_results = cs.simple_search(IUPAC) try: url = search_results[0].image_url except IndexError: url = "" return url
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 __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 index(req):
if req.method == "POST":
response = HttpResponse
# response["Access-Control-Allow-Origin"] = "*"
print "POST METHOD"
body = json.loads(req.body)
print body
print body["chemType"]
chemType = body["chemType"]
values = body["value"]
print chemType == "compound"
if chemType == "compound":
print "in this function"
CS = ChemSpider(security_token)
current_chem_symbol = CS.search(body)
print current_chem_symbol
print current_chem_symbol[0].common_name
returned_responses = [formulaToLatex(c.molecularFormula) for c in current_chem_symbol]
if len(returned_responses) > 4:
returned_responses = returned_responses[:4]
return JSONResponse({"latex": str(returned_resposnes)})
elif chemType == "element":
name = body["name"]
print "arrived at function"
for i in range(4):
print values[i]
print name
print textToLatex(name, values[0], values[1], values[2], values[3])
# return JSONResponse({"latex": textToLatex(name, values[0], values[1], values[2], values[3])})
return HttpResponse(
textToLatex(name, values[0], values[1], values[2], values[3]), content_type="text/plain"
)
else:
return
else:
template = loader.get_template("chemInterpreter/index.html")
return render(req, "chemInterpreter/index.html")
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
def get_image_url(self):
md = jsonpickle.decode(self.metadata)
if 'csid' in md:
# If this doc already has a csid, make the url
return 'http://www.chemspider.com/ImagesHandler.ashx?id=' + str(self.csid)
elif 'InChIKey' in md:
# If it doesnt but it does have an InChIKey get the csid and make the image url
from chemspipy import ChemSpider
cs = ChemSpider('b07b7eb2-0ba7-40db-abc3-2a77a7544a3d')
results = cs.search(md['InChIKey'])
if results:
# Return the image_url and also save the csid
csid = results[0].csid
md['csid'] = csid
self.metadata = jsonpickle.encode(md)
self.save()
return results[0].image_url
else:
return None
else:
# If it has neither, no image!
return None
def get_image_url(self):
md = jsonpickle.decode(self.metadata)
if 'csid' in md:
# If this doc already has a csid, make the url
return 'http://www.chemspider.com/ImagesHandler.ashx?id=' + str(self.csid)
elif 'InChIKey' in md or 'inchikey' in md:
# If it doesnt but it does have an InChIKey get the csid and make the image url
from chemspipy import ChemSpider
cs = ChemSpider('b07b7eb2-0ba7-40db-abc3-2a77a7544a3d')
ikey = md.get('InChIKey',md.get('inchikey'))
results = cs.search(ikey)
if results:
# Return the image_url and also save the csid
csid = results[0].csid
md['csid'] = csid
self.metadata = jsonpickle.encode(md)
self.save()
return results[0].image_url
else:
return None
else:
# If it has neither, no image!
return None
def smiles2cas(smiles_input):
myToken = 'a1d50aa3-6729-49df-a3e1-cd66240fab22'
cs = ChemSpider(security_token=myToken)
comp = cs.search(smiles_input)
for result in comp:
temp = result
res = temp.csid
res = str(res)
http = requests.session()
url = 'http://www.chemspider.com/MassSpecApi.asmx/GetExtendedCompoundInfoArray'
params = {'token': myToken}
http.post(url, data=params)
url_search = 'http://www.chemspider.com/Search.aspx?q=' + res
r = http.get(url_search)
soup = bs4.BeautifulSoup(r.text, "html.parser")
cas = [a.attrs.get('href') for a in soup.select('div.syn a[title="RN"]')]
for x in range(len(cas)):
cas[x] = re.findall(r"\"(.+?)\"", cas[x])
return (cas)
class SpiderRecovery:
def __init__(self, chemspider_api_key: str):
self._cs = ChemSpider(chemspider_api_key)
self._cs = None
self._has_stero = re.compile(
'(?:\([RSrsEZez+\-]\))|(?:[RSrsEZez][- \(])')
def recover_spider(self, name: str) -> Optional[str]:
"""Makes a best-effort attempt to recover SMILES strings from compound names unambiguously by searching ChemSpider.
Errs slightly on the side of failure.
If the compound name doesn't contain R, S, E, or Z (case-insensitive) in parantheses or followed by a hyphen or space,
assumes the compound has no defined sterocenters. In other words, it assumes minimal sterochemistry.
Returns the SMILES string if it was found unambiguously; otherwise returns None.
"""
results = self._cs.search(name)
if len(results) == 1:
return results[0].smiles
elif len(results) > 0: # try to recover if they're just enantiomers
connectivities = {result.inchikey[0:14] for result in results}
if len(connectivities) == 1:
if self._has_stero.match(name) is None:
no_sterocenters = {
result.smiles
for result in results if '@' not in result.smiles and
'/' not in result.smiles and '\\' not in result.smiles
}
if len(no_sterocenters) == 1:
return next(iter(no_sterocenters))
elif len(no_sterocenters) > 1:
warnings.warn(
"There are somehow {} compounds with the same connectivity and no defined sterocenters for {}"
.format(len(no_sterocenters), name))
return None # give up
def recover_spiders(
self,
names: Iterable[str],
sleep_seconds: float = 0.1) -> Iterator[Tuple[str, str]]:
"""Yields a SMILES string each time one is found. Returns a tuple of (name, smiles), which can be made into a dict."""
for name in names:
smiles = self.recover_spider(name)
time.sleep(sleep_seconds) # don't annoy the admins!
if smiles is not None:
yield name, smiles
class ChemspiderSearcher:
def __init__(self, api_key: str):
self.cs = ChemSpider(api_key)
def chemspider_names(
self,
names: Iterable[str],
partial_dict: Mapping[str, chemspipy.objects.Compound] = {},
sleep_secs_between: float = 0.1
) -> Mapping[str, chemspipy.objects.Compound]:
"""Build a dictionary mapping compound names to unique ChemSpider hits as chemspipy.objects.Compound objects, using partial_dict as a starting point.
Does not modify partial_dict. Warns for each compound that has multiple or no hits.
Immediately pickling the fetched results may be a good idea.
Example usage:
for compounds in chemspider_names(['Trichostatin A', 'Oxamflatin', 'Vinblastine']):
print("{} → {}".format(result.csid, result.smiles))
Result:
UserWarning: Multiple (2) hits found for Oxamflatin
392575 → C[C@H](/C=C(\C)/C=C/C(=O)NO)C(=O)c1ccc(cc1)N(C)C
12773 → CC[C@@]1(C[C@H]2C[C@@](c3c(c4ccccc4[nH]3)CCN(C2)C1)(c5cc6c(cc5OC)N([C@@H]7[C@]68CCN9[C@H]8[C@@](C=CC9)([C@H]([C@@]7(C(=O)OC)O)OC(=O)C)CC)C)C(=O)OC)O
"""
def fetch(name: str) -> Optional[chemspipy.objects.Compound]:
results = []
for result in self.cs.search(name): # blocks
results.append(result)
if len(results) == 0:
warnings.warn("No results found for {}".format(name))
elif len(results) > 1:
warnings.warn('Multiple ({}) hits found for {}'.format(
len(results), name))
else:
return results[0]
new_dict = partial_dict.copy()
for name in set(names) - set(new_dict.keys()):
got = fetch(name)
time.sleep(sleep_secs_between)
if got is not None:
new_dict[name] = got
return new_dict
from chemspipy import ChemSpider
cs = ChemSpider('c48d4595-ead2-40e7-85c9-1e5d2a77754c')
def get_chem(query):
chem = None
results = cs.search(query)
if results:
name = results[0].common_name
smiles = results[0].smiles
chem = {'name': name, 'smiles': smiles}
return chem
def get_smiles(query):
chem = None
results = cs.search(query)
if results:
smiles = results[0].smiles
return smiles
else:
return None
import CoolProp
from chemspipy import ChemSpider
from chemspipy_key import key # private file with the key (DO NOT COMMIT!!)
import glob, json
cs = ChemSpider(key)
# Map from name to Chemspider ID
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 process(self,
input_text: str = "",
input_file: str = "",
output_file: str = "",
output_file_sdf: str = "",
sdf_append: bool = False,
input_type: str = "",
lang: str = "eng",
paged_text: bool = False,
format_output: bool = True,
opsin_types: list = None,
standardize_mols: bool = True,
convert_ions: bool = True,
write_header: bool = True,
iob_format: bool = False,
dry_run: bool = False,
csv_delimiter: str = ";",
normalize_text: bool = True,
remove_duplicates: bool = False,
annotate: bool = True,
annotation_sleep: int = 2,
chemspider_token: str = "",
continue_on_failure: bool = False) -> OrderedDict:
r"""
Process the input file with ChemSpot.
Parameters
----------
input_text : str
String to be processed by ChemSpot.
input_file : str
Path to file to be processed by ChemSpot.
output_file : str
File to write output in.
output_file_sdf : str
File to write SDF output in. SDF is from OPSIN converted entities.
sdf_append : bool
If True, append new molecules to existing SDF file or create new one if doesn't exist. SDF is from OPSIN converted entities.
input_type : str
| When empty, input (MIME) type will be determined from magic bytes.
| Or you can specify "pdf", "pdf_scan", "image" or "text" and magic bytes check will be skipped.
lang : str
| Language which will Tesseract use for OCR. Available languages: https://github.com/tesseract-ocr/tessdata
| Multiple languages can be specified with "+" character, i.e. "eng+bul+fra".
paged_text : bool
If True and `input_type` is "text" or `input_text` is provided, try to assign pages to chemical entities.
ASCII control character 12 (Form Feed, '\f') is expected between pages.
format_output : bool
| If True, the value of "content" key of returned dict will be list of OrderedDicts.
| If True and `output_file` is set, the CSV file will be written.
| If False, the value of "content" key of returned dict will be None.
opsin_types : list
| List of ChemSpot entity types. Entities of types in this list will be converted with OPSIN. If you don't want
to convert entities, pass empty list.
| OPSIN is designed to convert IUPAC names to linear notation (SMILES etc.) so default value of `opsin_types`
is ["SYSTEMATIC"] (these should be only IUPAC names).
| ChemSpot entity types: "SYSTEMATIC", "IDENTIFIER", "FORMULA", "TRIVIAL", "ABBREVIATION", "FAMILY", "MULTIPLE"
standardize_mols : bool
If True, use molvs (https://github.com/mcs07/MolVS) to standardize molecules converted by OPSIN.
convert_ions : bool
If True, try to convert ion entities (e.g. "Ni(II)") to SMILES. Entities matching ion regex won't be converted
with OPSIN.
write_header : bool
If True and if `output_file` is set and `output_format` is True, write a CSV write_header:
"smiles", "bond_length", "resolution", "confidence", "learn", "page", "coordinates"
iob_format : bool
If True, output will be in IOB format.
dry_run : bool
If True, only return list of commands to be called by subprocess.
csv_delimiter : str
Delimiter for output CSV file.
normalize_text : bool
If True, normalize text before performing NER. It is strongly recommended to do so, because without normalization
can ChemSpot produce unpredictable results which cannot be parsed.
remove_duplicates : bool
If True, remove duplicated chemical entities. Note that some entities-compounds can have different names, but
same notation (SMILES, InChI etc.). This will only remove entities with same names. Not applicable for IOB format.
annotate : bool
| If True, try to annotate entities in PubChem and ChemSpider. Compound IDs will be assigned by searching with
each identifier, separately for entity name, SMILES etc.
| If entity has InChI key yet, prefer it in searching.
| If "*" is present in SMILES, skip annotation.
| If textual entity has single result in DB when searched by name, fill in missing identifiers (SMILES etc.).
annotation_sleep: int
How many seconds to sleep between annotation of each entity. It's for preventing overloading of databases.
chemspider_token : str
Your personal token for accessing the ChemSpider API (needed for annotation). Make account there to obtain it.
continue_on_failure : bool
| If True, continue running even if ChemSpot returns non-zero exit code.
| If False and error occurs, print it and return.
Returns
-------
dict
Keys:
- stdout: str ... standard output from ChemSpot
- stderr: str ... standard error output from ChemSpot
- exit_code: int ... exit code from ChemSpot
- content
- list of OrderedDicts ... when `format_output` is True
- None ... when `format_output` is False
- normalized_text : str
"""
if opsin_types is None:
opsin_types = ["SYSTEMATIC"]
if input_text and input_file:
input_file = ""
self.logger.warning("Both 'input_text' and 'input_file' are set, but 'input_text' will be prefered.")
elif not input_text and not input_file:
raise ValueError("One of 'input_text' or 'input_file' must be set.")
if not input_type and not input_text:
possible_input_types = ["pdf", "image", "text"]
input_type = get_input_file_type(input_file)
if input_type not in possible_input_types:
raise ValueError("Input file type ({}) is not one of {}".format(input_type, possible_input_types))
elif input_type and not input_text:
possible_input_types = ["pdf", "pdf_scan", "image", "text"]
if input_type not in possible_input_types:
raise ValueError("Unknown 'input_type'. Possible 'input_type' values are {}".format(possible_input_types))
if input_type in ["pdf", "pdf_scan", "image"]:
input_text, _ = get_text(input_file, input_type, lang=lang, tessdata_prefix=os.environ["TESSDATA_PREFIX"])
input_file = ""
if annotate and not chemspider_token:
self.logger.warning("Cannot perform annotation in ChemSpider: 'chemspider_token' is empty.")
options = ChainMap({k: v for k, v in {"iob_format": iob_format}.items() if v},
self.options_internal)
output_file_temp = None
commands, _, _ = self.build_commands(options, self._OPTIONS_REAL, self.path_to_binary)
commands.insert(1, str(self.options_internal["max_memory"]))
commands.append("-t")
if normalize_text:
normalizer = Normalizer(strip=True, collapse=True, hyphens=True, quotes=True, slashes=True, tildes=True, ellipsis=True)
if input_file:
with open(input_file, mode="r") as f:
input_text = f.read()
input_text = normalizer(input_text)
if not input_text:
raise UserWarning("'input_text' is empty after normalization.")
input_text = self.normalize_text(text=input_text)
input_file_normalized = NamedTemporaryFile(mode="w", encoding="utf-8")
input_file_normalized.write(input_text)
input_file_normalized.flush()
input_file = input_file_normalized.name
else:
if input_text:
input_file_temp = NamedTemporaryFile(mode="w", encoding="utf-8")
input_file_temp.write(input_text)
input_file_temp.flush()
input_file = input_file_temp.name
commands.append(os.path.abspath(input_file))
commands.append("-o")
if format_output:
output_file_temp = NamedTemporaryFile(mode="w", encoding="utf-8")
commands.append(os.path.abspath(output_file_temp.name))
else:
commands.append(os.path.abspath(output_file))
if dry_run:
return " ".join(commands)
stdout, stderr, exit_code = common_subprocess(commands)
if "OutOfMemoryError" in stderr:
raise RuntimeError("ChemSpot memory error: {}".format(stderr))
to_return = {"stdout": stdout, "stderr": stderr, "exit_code": exit_code, "content": None,
"normalized_text": input_text if normalize_text else None}
if not continue_on_failure and exit_code > 0:
self.logger.warning("ChemSpot error:")
eprint("\n\t".join("\n{}".format(stderr).splitlines()))
return to_return
if normalize_text:
to_return["normalized_text"] = input_text
if not format_output:
return to_return
elif format_output:
with open(output_file_temp.name, mode="r", encoding="utf-8") as f:
output_chs = f.read()
entities = self.parse_chemspot_iob(text=output_chs) if iob_format else self.parse_chemspot(text=output_chs)
to_return["content"] = entities
if remove_duplicates and not iob_format:
seen = set()
seen_add = seen.add
to_return["content"] = [x for x in to_return["content"] if not (x["entity"] in seen or seen_add(x["entity"]))]
if input_type in ["pdf", "pdf_scan"] or paged_text:
page_ends = []
for i, page in enumerate(input_text.split("\f")):
if page.strip():
try:
page_ends.append(page_ends[-1] + len(page) - 1)
except IndexError:
page_ends.append(len(page) - 1)
if opsin_types:
if convert_ions:
to_convert = [x["entity"] for x in to_return["content"] if x["type"] in opsin_types and not self.re_ion.match(x["entity"])]
else:
to_convert = [x["entity"] for x in to_return["content"] if x["type"] in opsin_types]
if to_convert:
opsin = OPSIN(verbosity=self.verbosity)
opsin_converted = opsin.process(input=to_convert, output_formats=["smiles", "inchi", "inchikey"],
standardize_mols=standardize_mols, output_file_sdf=output_file_sdf,
sdf_append=sdf_append)
opsin_converted = iter(opsin_converted["content"])
else:
self.logger.info("Nothing to convert with OPSIN.")
if annotate:
chemspider = ChemSpider(chemspider_token) if chemspider_token else None
for i, ent in enumerate(to_return["content"]):
if input_type in ["pdf", "pdf_scan"] or paged_text:
ent["page"] = str(bisect.bisect_left(page_ends, int(ent["start"])) + 1)
if convert_ions:
match_ion = self.re_ion.match(ent["entity"])
if match_ion:
match_ion = match_ion.groupdict()
match_charge = self.re_charge.search(match_ion["charge"])
if match_charge:
match_charge = match_charge.groupdict()
if match_charge["roman"]:
smiles = "[{}+{}]".format(match_ion["ion"], len(match_charge["roman"]))
elif match_charge["digit"]:
if "+" in match_ion["charge"]:
smiles = "[{}+{}]".format(match_ion["ion"], match_charge["digit"])
elif "-" in match_ion["charge"]:
smiles = "[{}-{}]".format(match_ion["ion"], match_charge["digit"])
elif match_charge["signs"]:
smiles = "[{}{}{}]".format(match_ion["ion"], match_charge["signs"][0],
len(match_charge["signs"]))
mol = MolFromSmiles(smiles)
if mol:
inchi = MolToInchi(mol)
if inchi:
ent.update(OrderedDict(
[("smiles", smiles), ("inchi", inchi), ("inchikey", InchiToInchiKey(inchi))]))
else:
ent.update(OrderedDict([("smiles", smiles), ("inchi", ""), ("inchikey", "")]))
else:
ent.update(OrderedDict([("smiles", ""), ("inchi", ""), ("inchikey", "")]))
else:
ent.update(OrderedDict([("smiles", ""), ("inchi", ""), ("inchikey", "")]))
if opsin_types and to_convert:
if ent["entity"] in to_convert:
ent_opsin = next(opsin_converted)
ent.update(OrderedDict([("smiles", ent_opsin["smiles"]), ("inchi", ent_opsin["inchi"]),
("inchikey", ent_opsin["inchikey"]), ("opsin_error", ent_opsin["error"])]))
elif convert_ions and self.re_ion.match(ent["entity"]):
ent.update(OrderedDict([("opsin_error", "")]))
elif (convert_ions and not self.re_ion.match(ent["entity"])) or (not convert_ions and ent["entity"] not in to_convert):
ent.update(OrderedDict([("smiles", ""), ("inchi", ""), ("inchikey", ""), ("opsin_error", "")]))
# TODO: this should be simplified...looks like garbage code
if annotate:
self.logger.info("Annotating entity {}/{}...".format(i + 1, len(to_return["content"])))
ent.update(OrderedDict([("pch_cids_by_inchikey", ""), ("chs_cids_by_inchikey", ""),
("pch_cids_by_name", ""), ("chs_cids_by_name", ""),
("pch_cids_by_smiles", ""), ("chs_cids_by_smiles", ""),
("pch_cids_by_inchi", ""), ("chs_cids_by_inchi", ""),
("pch_cids_by_formula", ""),
("pch_iupac_name", ""), ("chs_common_name", ""),
("pch_synonyms", "")]))
# do "double-annotation": some entities can be found in only one DB, updated and then searched in second DB
found_in_pch = False
found_in_chs = False
for _ in range(2):
results = []
# prefer InChI key
if "inchikey" in ent and ent["inchikey"]:
try:
results = get_compounds(ent["inchikey"], "inchikey")
if results:
if len(results) == 1:
result = results[0]
synonyms = result.synonyms
if synonyms:
ent["pch_synonyms"] = "\"{}\"".format("\",\"".join(synonyms))
ent["pch_iupac_name"] = result.iupac_name
if not found_in_chs:
ent["smiles"] = result.canonical_smiles or ent["smiles"]
ent["inchi"] = result.inchi or ent["inchi"]
ent["inchikey"] = result.inchikey or ent["inchikey"]
ent["pch_cids_by_inchikey"] = "\"{}\"".format(",".join([str(c.cid) for c in results]))
except (BadRequestError, NotFoundError, PubChemHTTPError, ResponseParseError, ServerError, TimeoutError, PubChemPyError):
pass
results = chemspider.search(ent["inchikey"]) if chemspider_token else []
if results:
if len(results) == 1:
result = results[0]
ent["chs_common_name"] = result.common_name
if not found_in_pch:
ent["smiles"] = result.smiles or ent["smiles"]
ent["inchi"] = result.stdinchi or ent["inchi"]
ent["inchikey"] = result.stdinchikey or ent["inchikey"]
ent["chs_cids_by_inchikey"] = "\"{}\"".format(",".join([str(c.csid) for c in results]))
else:
if (not found_in_pch and not found_in_chs) or (not found_in_pch and found_in_chs):
try:
results = get_compounds(ent["entity"] or ent["abbreviation"], "name")
if results:
if len(results) == 1:
found_in_pch = True
result = results[0]
synonyms = result.synonyms
if synonyms:
ent["pch_synonyms"] = "\"{}\"".format("\",\"".join(synonyms))
# only update identifiers if they weren't found in second DB
if not found_in_chs:
ent["smiles"] = result.canonical_smiles or ent["smiles"]
ent["inchi"] = result.inchi or ent["inchi"]
ent["inchikey"] = result.inchikey or ent["inchikey"]
ent["pch_iupac_name"] = result.iupac_name
ent["pch_cids_by_name"] = "\"{}\"".format(",".join([str(c.cid) for c in results]))
except (BadRequestError, NotFoundError, PubChemHTTPError, ResponseParseError, ServerError, TimeoutError, PubChemPyError):
pass
if (not found_in_pch and not found_in_chs) or (found_in_pch and not found_in_chs):
results = chemspider.search(ent["entity"] or ent["abbreviation"]) if chemspider_token else []
if results:
if len(results) == 1:
found_in_chs = True
result = results[0]
if not found_in_pch:
ent["smiles"] = result.smiles or ent["smiles"]
ent["inchi"] = result.stdinchi or ent["inchi"]
ent["inchikey"] = result.stdinchikey or ent["inchikey"]
ent["chs_common_name"] = result.common_name
ent["chs_cids_by_name"] = "\"{}\"".format(",".join([str(c.csid) for c in results]))
for search_field, col_pch, col_chs in [("smiles", "pch_cids_by_smiles", "chs_cids_by_smiles"),
("inchi", "pch_cids_by_inchi", "chs_cids_by_inchi"),
("formula", "pch_cids_by_formula", "")]:
results_pch = []
results_chs = []
if search_field == "smiles" and "smiles" in ent and ent["smiles"] and "*" not in ent["smiles"]:
if (not found_in_pch and not found_in_chs) or (not found_in_pch and found_in_chs):
try:
results_pch = get_compounds(ent["smiles"], "smiles")
except (BadRequestError, NotFoundError, PubChemHTTPError, ResponseParseError, ServerError, TimeoutError, PubChemPyError):
pass
if (not found_in_pch and not found_in_chs) or (found_in_pch and not found_in_chs):
results_chs = chemspider.search(ent["smiles"]) if chemspider_token else []
elif search_field == "inchi" and "inchi" in ent and ent["inchi"]:
if (not found_in_pch and not found_in_chs) or (not found_in_pch and found_in_chs):
try:
results_pch = get_compounds(ent["inchi"], "inchi")
except (BadRequestError, NotFoundError, PubChemHTTPError, ResponseParseError, ServerError, TimeoutError, PubChemPyError):
pass
if (not found_in_pch and not found_in_chs) or (found_in_pch and not found_in_chs):
results_chs = chemspider.search(ent["inchi"]) if chemspider_token else []
elif search_field == "formula":
if (not found_in_pch and not found_in_chs) or (not found_in_pch and found_in_chs):
try:
results_pch = get_compounds(ent["entity"], "formula")
except (BadRequestError, NotFoundError, PubChemHTTPError, ResponseParseError, ServerError, TimeoutError, PubChemPyError):
pass
# ChemSpider doesn't have search field for 'formula'
if results_pch:
ent[col_pch] = "\"{}\"".format(",".join([str(c.cid) for c in results_pch]))
if results_chs:
ent[col_chs] = "\"{}\"".format(",".join([str(c.csid) for c in results_chs]))
sleep(0.5)
sleep(annotation_sleep)
if not found_in_pch and not found_in_chs:
break
if output_file:
dict_to_csv(to_return["content"], output_file=output_file, csv_delimiter=csv_delimiter, write_header=write_header)
return to_return
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.common.by import By
from selenium.common.exceptions import TimeoutException
from dotenv import load_dotenv
load_dotenv()
DISCORD_TOKEN = os.getenv('DISCORD_TOKEN')
CHEMSPIDER_TOKEN = os.getenv('CHEMSPIDER_TOKEN')
WOLFRAM_TOKEN = os.getenv('WOLFRAM_TOKEN')
cs = ChemSpider(CHEMSPIDER_TOKEN)
wolfram = wolframalpha.Client(WOLFRAM_TOKEN)
client = commands.Bot(command_prefix='!')
op = webdriver.ChromeOptions()
op.binary_location = os.getenv('GOOGLE_CHROME_BIN')
op.add_argument('--headless')
op.add_argument('--no-sandbox')
op.add_argument('--disable-dev-sh-usage')
driver = webdriver.Chrome(executable_path=os.getenv('CHROMEDRIVER_PATH'),
chrome_options=op)
# for local testing purposes only; comment out when deployed to Heroku
#driver = webdriver.Firefox()
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
def __init__(self,argv):
''' load input arguments'''
self.CAS_file = argv[1]
''' my chemsphder token '''
self.cs=ChemSpider('d1778a9f-c41f-41f6-920e-fc6d9ff739ca')
def handle(self, *args, **kwargs):
"""Handle the command call."""
folder = kwargs['directory']
start_at_reactions = kwargs['reactions']
start_at_descriptors = kwargs['descriptors']
start_at_quantities = kwargs['quantities']
start_number = kwargs['start_number']
delete_all = kwargs['delete_all']
no_compound_prompts = kwargs['no_compound_prompts']
start_at_delete = not (
start_at_reactions or start_at_descriptors or start_at_quantities)
if start_at_delete:
self.stdout.write('Deleting reactions')
if delete_all:
PerformedReaction.objects.all().delete()
else:
with transaction.atomic():
with open(path.join(folder, 'performedReactions.tsv')) as reactions:
reader = csv.DictReader(reactions, delimiter='\t')
for i, r in enumerate(reader):
if start_at_delete and i < start_number:
continue
ref = convert_legacy_reference(r['reference'])
legacyID = r['id']
ps = PerformedReaction.objects.filter(
reference=ref)
if ps:
self.stdout.write(
'{}: Deleting reaction with reference {}'.format(i, ref))
ps.delete()
ps = PerformedReaction.objects.filter(
reference=ref.lower())
if ps:
self.stdout.write(
'{}: Deleting reaction with converted legacy reference {}'.format(i, ref))
ps.delete()
ps = PerformedReaction.objects.filter(
convertedLegacyRef=ref)
if ps:
self.stdout.write(
'{}: Deleting reaction with converted legacy reference {}'.format(i, ref))
ps.delete()
ps = PerformedReaction.objects.filter(
legacyID=legacyID)
if ps:
self.stdout.write(
'{}: Deleting reaction with legacy id {}'.format(i, legacyID))
ps.delete()
if start_at_reactions or start_at_delete:
warnings.simplefilter('error')
with open(path.join(folder, 'performedReactions.tsv')) as reactions:
self.stdout.write('Creating reactions')
reader = csv.DictReader(reactions, delimiter='\t')
for i, r in enumerate(reader):
if start_at_reactions and i < start_number:
continue
ref = convert_legacy_reference(r['reference'])
convertedLegacyRef = ref
ps = PerformedReaction.objects.filter(
convertedLegacyRef=convertedLegacyRef)
if ps.exists():
ref = '{}_{}'.format(ref, r['id'])
valid = False
notes = r[
'notes'] + ' Duplicate reference disambiguated with legacy id.'
for p in ps:
if p.convertedLegacyRef == p.reference:
p.valid = False
p.notes += u' Duplicate reference disambiguated with legacy id.'
p.reference = '{}_{}'.format(
p.convertedLegacyRef, p.legacyID)
p.save(calcDescriptors=False)
else:
valid = bool(int(r['valid']))
notes = r['notes']
p = PerformedReaction(
reference=ref,
legacyRef=r['reference'],
convertedLegacyRef=convertedLegacyRef,
labGroup=LabGroup.objects.get(
title=r['labGroup.title']),
legacyID=r['id'],
notes=notes,
user=User.objects.get(username=r['user.username']),
valid=valid,
legacyRecommendedFlag=(
r['legacyRecommendedFlag'] == 'Yes'),
insertedDateTime=r['insertedDateTime'],
public=int(r['public'])
)
self.stdout.write(
'{}: Creating reaction with reference {}'.format(i, ref))
p.full_clean()
p.save(calcDescriptors=False)
if start_at_delete or start_at_reactions or start_at_descriptors:
with open(path.join(folder, 'performedReactions.tsv')) as reactions:
self.stdout.write('Creating manual descriptors')
reader = csv.DictReader(reactions, delimiter='\t')
outValues = []
outBoolValues = []
purityValues = []
temperatureValues = []
timeValues = []
pHValues = []
preHeatStandingValues = []
teflonValues = []
slowCoolValues = []
leakValues = []
for i, r in enumerate(reader):
if start_at_descriptors and i < start_number:
continue
ref = convert_legacy_reference(r['reference'])
id = r['id']
self.stdout.write(
'{}: Reiterating for reaction with reference {}, legacyID {}'.format(i, ref, id))
p = PerformedReaction.objects.get(legacyID=id)
if r['duplicateOf.reference']:
convertedDupRef = convert_legacy_reference(
r['duplicateOf.reference'])
try:
p.duplicateOf = PerformedReaction.objects.get(
convertedLegacyRef=convertedDupRef)
p.save(calcDescriptors=False)
except PerformedReaction.DoesNotExist:
self.stderr.write('Reaction {} marked as duplicate of reaction {}, but the latter does not exist'.format(
ref, r['duplicateOf.reference']))
p.notes += 'Marked as duplicate of reaction with legacy reference {}, but it does not exist'.format(r[
'duplicateOf.reference'])
p.valid = False
p.save(calcDescriptors=False)
except PerformedReaction.MultipleObjectsReturned:
self.stderr.write('Reaction {} marked as duplicate of reaction {}, but more than one of the latter exists'.format(
ref, r['duplicateOf.reference']))
p.notes += 'Marked as duplicate of reaction with legacy reference {}, but more than one reaction with that reference exists'.format(r[
'duplicateOf.reference'])
p.valid = False
p.save(calcDescriptors=False)
outcomeValue = int(r['outcome']) if (
r['outcome'] in (str(x) for x in range(1, 5))) else None
try:
v = OrdRxnDescriptorValue.objects.get(
descriptor=outcomeDescriptor, reaction=p)
if v.value != outcomeValue:
v.value = outcomeValue
v.save()
except OrdRxnDescriptorValue.DoesNotExist:
outValue = outcomeDescriptor.createValue(
p, outcomeValue)
outValues.append(outValue)
value = True if (outcomeValue > 2) else False
try:
v = BoolRxnDescriptorValue.objects.get(
descriptor=outcomeBooleanDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except BoolRxnDescriptorValue.DoesNotExist:
outBoolValue = outcomeBooleanDescriptor.createValue(
p, value)
outBoolValues.append(outBoolValue)
value = int(r['purity']) if (
r['purity'] in ('1', '2')) else None
try:
v = OrdRxnDescriptorValue.objects.get(
descriptor=purityDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except OrdRxnDescriptorValue.DoesNotExist:
purityValue = purityDescriptor.createValue(p, value)
purityValues.append(purityValue)
value = (float(r['temp']) + 273.15) if (r['temp']
not in ('', '?')) else None
try:
v = NumRxnDescriptorValue.objects.get(
descriptor=temperatureDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
temperatureDescriptorValue = temperatureDescriptor.createValue(
p, value)
temperatureValues.append(temperatureDescriptorValue)
value = float(r['time']) * 60 if (r['time']
not in ['', '?']) else None
try:
v = NumRxnDescriptorValue.objects.get(
descriptor=timeDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
timeDescriptorValue = timeDescriptor.createValue(
p, value)
timeValues.append(timeDescriptorValue)
value = float(r['pH']) if (
r['pH'] not in ('', '?')) else None
try:
v = NumRxnDescriptorValue.objects.get(
descriptor=pHDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
pHDescriptorValue = pHDescriptor.createValue(p, value)
pHValues.append(pHDescriptorValue)
value = bool(r['pre_heat standing']) if (
r.get('pre_heat standing') not in ('', None)) else None
try:
v = NumRxnDescriptorValue.objects.get(
descriptor=preHeatStandingDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
preHeatStandingDescriptorValue = preHeatStandingDescriptor.createValue(
p, value)
preHeatStandingValues.append(
preHeatStandingDescriptorValue)
value = bool(int(r['teflon_pouch'])) if (
r.get('teflon_pouch') not in(None, '')) else None
try:
v = BoolRxnDescriptorValue.objects.get(
descriptor=teflonDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except BoolRxnDescriptorValue.DoesNotExist:
teflonDescriptorValue = teflonDescriptor.createValue(
p, value)
teflonValues.append(teflonDescriptorValue)
leak_string = r['leak']
if leak_string in (None, '', '?'):
value = None
elif leak_string.lower() == 'yes':
value = True
elif leak_string.lower() == 'no':
value = False
else:
raise RuntimeError(
"Unrecognized string '{}' in leak column".format(leak_string))
try:
v = BoolRxnDescriptorValue.objects.get(
descriptor=leakDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except BoolRxnDescriptorValue.DoesNotExist:
leakDescriptorValue = leakDescriptor.createValue(
p, value)
leakValues.append(leakDescriptorValue)
slow_cool_string = r['slow_cool']
if slow_cool_string in (None, '', '?'):
value = None
elif slow_cool_string.lower() == 'yes':
value = True
elif slow_cool_string.lower() == 'no':
value = False
else:
raise RuntimeError(
"Unrecognized string '{}' in slow_cool column".format(slow_cool_string))
try:
v = BoolRxnDescriptorValue.objects.get(
descriptor=slowCoolDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except BoolRxnDescriptorValue.DoesNotExist:
slowCoolDescriptorValue = slowCoolDescriptor.createValue(
p, value)
slowCoolValues.append(slowCoolDescriptorValue)
if len(outValues) > save_at_once:
self.stdout.write("Saving outValues...")
OrdRxnDescriptorValue.objects.bulk_create(outValues)
outValues = []
self.stdout.write("...saved")
if len(outBoolValues) > save_at_once:
self.stdout.write("Saving outBoolValues...")
BoolRxnDescriptorValue.objects.bulk_create(
outBoolValues)
outBoolValues = []
self.stdout.write("...saved")
if len(purityValues) > save_at_once:
self.stdout.write("Saving purityValues...")
OrdRxnDescriptorValue.objects.bulk_create(purityValues)
purityValues = []
self.stdout.write("...saved")
if len(temperatureValues) > save_at_once:
self.stdout.write("Saving temperatureValues...")
NumRxnDescriptorValue.objects.bulk_create(
temperatureValues)
temperatureValues = []
self.stdout.write("...saved")
if len(timeValues) > save_at_once:
self.stdout.write("Saving timeValues...")
NumRxnDescriptorValue.objects.bulk_create(timeValues)
timeValues = []
self.stdout.write("...saved")
if len(pHValues) > save_at_once:
self.stdout.write("Saving pHValues...")
NumRxnDescriptorValue.objects.bulk_create(pHValues)
pHValues = []
self.stdout.write("...saved")
if len(preHeatStandingValues) > save_at_once:
self.stdout.write("Saving preHeatStandingValues...")
NumRxnDescriptorValue.objects.bulk_create(
preHeatStandingValues)
preHeatStandingValues = []
self.stdout.write("...saved")
if len(teflonValues) > save_at_once:
self.stdout.write("Saving teflonValues...")
BoolRxnDescriptorValue.objects.bulk_create(
teflonValues)
teflonValues = []
self.stdout.write("...saved")
if len(leakValues) > save_at_once:
self.stdout.write("Saving leakValues...")
BoolRxnDescriptorValue.objects.bulk_create(leakValues)
leakValues = []
self.stdout.write("...saved")
if len(slowCoolValues) > save_at_once:
self.stdout.write("Saving slowCoolValues...")
BoolRxnDescriptorValue.objects.bulk_create(
slowCoolValues)
slowCoolValues = []
self.stdout.write("...saved")
self.stdout.write("Saving all remaining values...")
OrdRxnDescriptorValue.objects.bulk_create(outValues)
BoolRxnDescriptorValue.objects.bulk_create(outBoolValues)
OrdRxnDescriptorValue.objects.bulk_create(purityValues)
NumRxnDescriptorValue.objects.bulk_create(temperatureValues)
NumRxnDescriptorValue.objects.bulk_create(timeValues)
NumRxnDescriptorValue.objects.bulk_create(pHValues)
NumRxnDescriptorValue.objects.bulk_create(
preHeatStandingValues)
BoolRxnDescriptorValue.objects.bulk_create(teflonValues)
BoolRxnDescriptorValue.objects.bulk_create(leakValues)
BoolRxnDescriptorValue.objects.bulk_create(slowCoolValues)
outValues = []
outBoolValues = []
purityValues = []
temperatureValues = []
timeValues = []
pHValues = []
preHeatStandingValues = []
teflonValues = []
leakValues = []
slowCoolValues = []
self.stdout.write("...saved")
with open(path.join(folder, 'compoundquantities.tsv')) as cqs, open(path.join(folder, 'compoundquantities_fixed.tsv'), 'a') as fixed_cqs:
self.stdout.write('Creating or updating compound quantities')
reader = csv.DictReader(cqs, delimiter='\t')
writer = csv.DictWriter(
fixed_cqs, reader.fieldnames + ['compound.old_abbrev'], delimiter='\t')
if not (start_at_quantities and start_number > 0):
writer.writeheader()
quantities = []
cs = ChemSpider(settings.CHEMSPIDER_TOKEN)
for i, r in enumerate(reader):
if start_at_quantities and (i < start_number):
continue
if not (r['compound.abbrev'] or r['compoundrole.name'] or r['amount']):
# this is just a blank entry
continue
legacyID = r['reaction.id']
reaction = PerformedReaction.objects.get(legacyID=legacyID)
compound_abbrev = r['compound.abbrev']
correct_abbrev = reagent_dict[
compound_abbrev] if compound_abbrev in reagent_dict else compound_abbrev
compound_found = False
while correct_abbrev and not compound_found:
try:
compound = Compound.objects.get(
abbrev=correct_abbrev, labGroup=reaction.labGroup)
compound_found = True
r['compound.old_abbrev'] = r['compound.abbrev']
r['compound.abbrev'] = correct_abbrev
except Compound.DoesNotExist:
if no_compound_prompts:
correct_abbrev = ''
else:
self.stderr.write(
'Could not find compound with abbreviation {}. Checking chemspider...'.format(correct_abbrev))
results = cs.simple_search(correct_abbrev)
if len(results) != 1:
CSID = raw_input(
'Could not find unique compound with abbreviation {}. Do you know the CSID? '.format(correct_abbrev))
if CSID.isdigit():
results = cs.simple_search(CSID)
if len(results) == 1:
compound = None
try:
compound = Compound.objects.get(
CSID=results[0].csid, labGroup=reaction.labGroup)
except Compound.DoesNotExist:
pass
user_response = None
while user_response is None:
if compound is None:
user_verification = raw_input('Found unique compound with CSID {} and name {} for abbreviation {}. This is NOT IN THE COMPOUND GUIDE. Is this correct? (y/n): '.format(
results[0].csid, results[0].common_name, correct_abbrev))
else:
user_verification = raw_input('Found unique compound with CSID {}, name {}, and abbreviation {} for abbreviation {} in the compound guide. Is this correct? (y/n): '.format(
compound.CSID, compound.name, compound.abbrev, correct_abbrev))
if user_verification and user_verification.lower()[0] == 'y':
user_response = True
elif user_verification and user_verification.lower()[0] == 'n':
user_response = False
if user_response:
if compound is not None:
correct_abbrev = compound.abbrev
reagent_dict[
compound_abbrev] = correct_abbrev
continue
else:
self.stderr.write('Creating compound with CSID {}, abbrevation {}, name {}'.format(
results[0].csid, correct_abbrev, results[0].common_name))
c = Compound(
CSID=results[0].csid, labGroup=reaction.labGroup, abbrev=correct_abbrev)
try:
c.csConsistencyCheck()
c.save(invalidateReactions=False)
continue
except ValidationError:
c.delete()
raise
self.stderr.write(
'Could not get unambiguous chemspider entry for abbreviation {}'.format(correct_abbrev))
self.stderr.write('Unknown Reactant {} with amount {} {} in reaction {}'.format(
r['compound.abbrev'], r['amount'], r['unit'], r['reaction.reference']))
correct_abbrev = raw_input(
'What is the correct abbreviation for this? ')
reagent_dict[compound_abbrev] = correct_abbrev
if compound_found:
self.stdout.write('{}: Creating quantity for compound {} and reaction {}'.format(
i, compound.abbrev, reaction.reference))
if r['compound.abbrev'] in ('water', 'H2O'):
r['density'] = 1
try:
mw = NumMolDescriptorValue.objects.get(
compound=compound, descriptor__heading='mw').value
except NumMolDescriptorValue.DoesNotExist:
compound.save(invalidateReactions=False)
mw = NumMolDescriptorValue.objects.get(
compound=compound, descriptor__heading='mw').value
if r['compound.old_abbrev'] is not None and r['compound.old_abbrev'] != r['compound.abbrev']:
reaction.notes += ' Compound abbreviation {} changed to {}'.format(
r['compound.old_abbrev'], r['compound.abbrev'])
reaction.save(calcDescriptors=False)
if r['compoundrole.name'] == 'pH':
reaction.notes += ' pH adjusting reagent used: {}, {}{}'.format(
compound, r['amount'], r['unit'])
reaction.save(calcDescriptors=False)
else:
compoundrole = None
while compoundrole is None and r['compoundrole.name'] in (None, '', '?'):
if r['compoundrole.name'] in (None, '', '?'):
classes = compound.chemicalClasses.all()
if classes.count() > 1:
self.stderr.write(
'{} has more than one chemical class: {}'.format(compound, classes))
role_label = raw_input('Which is the correct role for reagent {} in reaction {} with amount {} {}'.format(
compound, reaction, r['amount'], r['unit']))
elif classes.count() == 0:
self.stderr.write(
'{} has no chemical classes'.format(compound))
role_label = raw_input(
'What chemical class does {} belong to? '.format(compound))
cc = ChemicalClass.objects.get(
label=role_label)
compound.chemicalClasses.add(cc)
# Sanity check
assert(
compound.chemicalClasses.all().count() == 1)
else: # count == 1
role_label = classes[0].label
self.stderr.write('No reaction role listed for reagent {} with amount {} {} in reaction {}. '
'Using chemical class {}'.format(compound, reaction, r['amount'], r['unit'], role_label))
r['compoundrole.name'] = role_label
else:
role_label = r['compoundrole.name']
if not role_label:
reaction.notes += ' No role for reactant {} with amount {} {}'.format(
r['compound.abbrev'], r['amount'], r['unit'])
reaction.save(calcDescriptors=False)
else:
try:
compoundrole = CompoundRole.objects.get(
label=role_label)
except CompoundRole.DoesNotExist:
user_response = None
if role_label in role_dict:
new_role_label = role_dict[role_label]
else:
while user_response is None:
user_verification = raw_input(
'Compound role {} does not exist. Would you like to add it? ')
if user_verification and user_verification.lower()[0] == 'y':
user_response = True
elif user_verification and user_verification.lower()[0] == 'n':
user_response = False
if user_response:
compoundrole = CompoundRole.objects.create(
label=role_label)
else:
new_role_label = raw_input(
'What should this label be? ')
role_dict[
role_label] = new_role_label
r['compoundrole.name'] = new_role_label
self.stdout.write('\tadding {} with role {} to {}'.format(
compound.abbrev, role_label, reaction.reference))
if r['amount'] in ('', '?'):
amount = None
reaction.notes += ' No amount for reactant {} with role {}'.format(
r['compound.abbrev'], r['compoundrole.name'])
reaction.save(calcDescriptors=False)
elif r['unit'] == 'g':
amount = float(r['amount']) / mw
elif r['unit'] == 'd' or r['unit'] == 'mL':
valid_density = False
while not valid_density:
if compound.abbrev in density_dict:
r['density'] = density_dict[
compound.abbrev]
try:
density = float(r['density'])
valid_density = True
except (TypeError, ValueError):
self.stderr.write("Density '{}' cannot be converted to float. (Compound {} with amount {} {} in reaction {})".format(
r['density'], compound, r['amount'], r['unit'], reaction))
r['density'] = raw_input(
'What is the density? ')
density_dict[compound.abbrev] = r[
'density']
if r['unit'] == 'd':
amount = float(
r['amount']) * 0.0375 * density / mw
elif r['unit'] == 'mL':
amount = float(
r['amount']) * density / mw
else:
raise RuntimeError('invalid unit entered')
# convert to millimoles
if amount is not None:
amount = (amount * 1000)
cqq = CompoundQuantity.objects.filter(
compound=compound, reaction=reaction)
if cqq.exists():
cqq.delete()
quantity = CompoundQuantity(
compound=compound, reaction=reaction, role=compoundrole, amount=amount)
quantities.append(quantity)
if len(quantities) > save_at_once:
self.stdout.write('Saving...')
CompoundQuantity.objects.bulk_create(
quantities)
quantities = []
else:
self.stderr.write('Unknown Reactant {} with amount {} {} in reaction {}'.format(
r['compound.abbrev'], r['amount'], r['unit'], r['reaction.reference']))
reaction.notes += ' Unknown Reactant {} with amount {} {}'.format(
r['compound.abbrev'], r['amount'], r['unit'])
reaction.valid = False
reaction.save(calcDescriptors=False)
writer.writerow(r)
self.stdout.write('Saving...')
CompoundQuantity.objects.bulk_create(quantities)
quantities = []
description="Script to obtain SMILES for a solutes in a list")
argparser.add_argument('-db', '--db', help="the molecule database")
argparser.add_argument('-solvent',
'--solvent',
help="the solvent",
default="water")
argparser.add_argument('-solutes', '--solutes', help="the list of solutes")
args = argparser.parse_args()
db = dblib.SolvDb(filename=args.db, type="abs", filehandle="^0")
solutes = [s.strip() for s in open(args.solutes, 'r').readlines()]
if os.getenv("SPIDERKEY") is None:
print "SPIDERKEY environmental variable not set! Exit."
quit()
cs = ChemSpider(os.getenv("SPIDERKEY"))
# Loop over all the database entries in the solute lists
n = 0
for entry in db.itersolutelist(args.solvent, solutes):
if os.path.exists(entry.FileHandle + ".smi"): continue
hits = cs.search(entry.SoluteName)
if len(hits) > 0:
smi = hits[0].smiles
with open(entry.FileHandle + ".smi", "w") as f:
f.write("%s\n" % smi)
else:
print entry.SoluteName, entry.FileHandle
n += 1
print "Looped over %d solutes" % n
def handle(self, *args, **kwargs):
folder = kwargs['directory']
#if not path.isfile(path.join(folder, 'performedReactionsNoDupsLower.tsv')):
#self.stdout.write('Writing file with all references that were uppercase (now lower) and duplicate references disambiguated (arbitrarily)')
#with open(path.join(folder, 'performedReactions.tsv')) as in_file, open(path.join(folder, 'performedReactionsNoDupsLower.tsv'), 'w') as out_file:
#references = set()
#reader = csv.DictReader(in_file, delimiter='\t')
#writer = csv.DictWriter(out_file, delimiter='\t', fieldnames=reader.fieldnames)
#writer.writeheader()
#case_count = 0
#valid_case_count = 0
#dup_count = 0
#for r in reader:
#ref = r['reference'].lower()
#if ref != r['reference']:
#self.stderr.write('Reference {} was not in lowercase. Converted.'.format(r['reference']))
#case_count += 1
#if r['valid'] == '1':
#valid_case_count += 1
#if ref in references:
#r['notes'] += ' Duplicated reference'
#r['valid'] = 0
#dup_count += 1
#i = 1
#new_ref = ref
#while new_ref in references:
#new_ref = '{}_dup{}'.format(ref, i)
#i += 1
#self.stderr.write('Reference {} duplicated {} times. Renamed and invalidated'.format(ref, i))
#ref = new_ref
#references.add(ref)
#r['reference'] = ref
#writer.writerow(r)
#self.stderr.write('{} references converted to lowercase. {} were valid'.format(case_count, valid_case_count))
#self.stderr.write('{} references with _dupX appended to remove duplicate reference'.format(dup_count))
#with open(path.join(folder, 'performedReactionsNoDupsLower.tsv')) as reactions:
#reader = csv.DictReader(reactions, delimiter='\t')
#for r in reader:
#if not PerformedReaction.objects.filter(reference=r['reference'].lower()).exists():
#p = PerformedReaction(
#reference = r['reference'],
#labGroup = LabGroup.objects.get(title=r['labGroup.title']),
#notes = r['notes'],
#user = User.objects.get(username=r['user.username']),
#valid = int(r['valid']),
#legacyRecommendedFlag=r['legacyRecommendedFlag']=='Yes',
#insertedDateTime=r['insertedDateTime'],
#public=int(r['public'])
#)
#self.stdout.write('Creating reaction with reference {}'.format(p.reference))
#p.validate_unique()
#p.save(calcDescriptors=False)
with open(path.join(folder, 'performedReactionsNoDupsLower.tsv')) as reactions:
reader = csv.DictReader(reactions, delimiter='\t')
outValues = []
outBoolValues = []
purityValues = []
temperatureValues = []
timeValues = []
pHValues = []
preHeatStandingValues = []
teflonValues = []
for r in reader:
self.stdout.write('Reiterating for reaction with reference {}'.format(r['reference'].lower()))
ps = PerformedReaction.objects.filter(reference=r['reference'].lower())
if ps.count() > 1:
ps = ps.filter(valid=True)
if ps.exists():
if ps.count() > 1:
raise RuntimeError('{} has more than one reaction'.format(r['reference'].lower()))
p = ps[0]
try:
p.duplicateOf = PerformedReaction.objects.get(reference=r['duplicateOf.reference'].lower())
p.save()
except PerformedReaction.DoesNotExist:
pass
#outValue = OrdRxnDescriptorValue.objects.get_or_create(descriptor=outcomeDescriptor, reaction=p)[0]
outcomeValue = int(r['outcome']) if (r['outcome'] in (str(x) for x in range (1, 5))) else None
try:
v = OrdRxnDescriptorValue.objects.get(descriptor=outcomeDescriptor, reaction=p)
if v.value != outcomeValue:
v.value = outcomeValue
v.save()
except OrdRxnDescriptorValue.DoesNotExist:
outValue = outcomeDescriptor.createValue(p, outcomeValue)
#outValue.save()
outValues.append(outValue)
#outBoolValue = BoolRxnDescriptorValue.objects.get_or_create(descriptor=outcomeBooleanDescriptor, reaction=p)[0]
value = True if (outcomeValue > 2) else False
try:
v = BoolRxnDescriptorValue.objects.get(descriptor=outcomeBooleanDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except BoolRxnDescriptorValue.DoesNotExist:
#outBoolValue.save()
outBoolValue = outcomeBooleanDescriptor.createValue(p, value)
outBoolValues.append(outBoolValue)
#purityValue = OrdRxnDescriptorValue.objects.get_or_create(descriptor=purityDescriptor, reaction=p)[0]
value = int(r['purity']) if (r['purity'] in ('1', '2')) else None
try:
v = OrdRxnDescriptorValue.objects.get(descriptor=purityDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except OrdRxnDescriptorValue.DoesNotExist:
#purityValue.save()
purityValue = purityDescriptor.createValue(p, value)
purityValues.append(purityValue)
#temperatureDescriptorValue = NumRxnDescriptorValue.objects.get_or_create(descriptor=temperatureDescriptor, reaction=p)[0]
value = (float(r['temp']) + 273.15) if (r['temp'] not in ('', '?')) else None
try:
v = NumRxnDescriptorValue.objects.get(descriptor=temperatureDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
#temperatureDescriptorValue.save()
temperatureDescriptorValue = temperatureDescriptor.createValue(p, value)
temperatureValues.append(temperatureDescriptorValue)
#timeDescriptorValue = NumRxnDescriptorValue.objects.get_or_create(descriptor=timeDescriptor, reaction=p)[0]
value = float(r['time'])*60 if (r['time'] not in ['', '?']) else None
try:
v = NumRxnDescriptorValue.objects.get(descriptor=timeDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
#timeDescriptorValue.save()
timeDescriptorValue = timeDescriptor.createValue(p, value)
timeValues.append(timeDescriptorValue)
#pHDescriptorValue = NumRxnDescriptorValue.objects.get_or_create(descriptor=pHDescriptor, reaction=p)[0]
value = float(r['pH']) if (r['pH'] not in ('', '?')) else None
try:
v = NumRxnDescriptorValue.objects.get(descriptor=pHDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
#pHDescriptorValue.save()
pHDescriptorValue = pHDescriptor.createValue(p, value)
pHValues.append(pHDescriptorValue)
#preHeatStandingDescriptorValue = NumRxnDescriptorValue.objects.get_or_create(descriptor=preHeatStandingDescriptor, reaction=p)[0]
value = bool(r['pre_heat standing']) if (r.get('pre_heat standing') not in ('', None)) else None
try:
v = NumRxnDescriptorValue.objects.get(descriptor=preHeatStandingDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except NumRxnDescriptorValue.DoesNotExist:
#preHeatStandingDescriptorValue.save()
preHeatStandingDescriptorValue = preHeatStandingDescriptor.createValue(p, value)
preHeatStandingValues.append(preHeatStandingDescriptorValue)
#teflonDescriptorValue = BoolRxnDescriptorValue.objects.get_or_create(descriptor=teflonDescriptor, reaction=p)[0]
value = bool(int(r['teflon_pouch'])) if (r.get('teflon_pouch') not in(None, '')) else None
try:
v = BoolRxnDescriptorValue.objects.get(descriptor=teflonDescriptor, reaction=p)
if v.value != value:
v.value = value
v.save()
except BoolRxnDescriptorValue.DoesNotExist:
#teflonDescriptorValue.save()
teflonDescriptorValue = teflonDescriptor.createValue(p, value)
teflonValues.append(teflonDescriptorValue)
if len(outValues) > 500:
self.stdout.write("Saving...")
OrdRxnDescriptorValue.objects.bulk_create(outValues)
BoolRxnDescriptorValue.objects.bulk_create(outBoolValues)
OrdRxnDescriptorValue.objects.bulk_create(purityValues)
NumRxnDescriptorValue.objects.bulk_create(temperatureValues)
NumRxnDescriptorValue.objects.bulk_create(timeValues)
NumRxnDescriptorValue.objects.bulk_create(pHValues)
NumRxnDescriptorValue.objects.bulk_create(preHeatStandingValues)
BoolRxnDescriptorValue.objects.bulk_create(teflonValues)
outValues = []
outBoolValues = []
purityValues = []
temperatureValues = []
timeValues = []
pHValues = []
preHeatStandingValues = []
teflonValues = []
self.stdout.write("...saved")
with open(path.join(folder, 'compound_labs.tsv')) as compounds:
reader = csv.DictReader(compounds, delimiter='\t')
cs = ChemSpider(settings.CHEMSPIDER_TOKEN)
for r in reader:
l = LabGroup.objects.get(title=r['labGroup.title'])
if not Compound.objects.filter(abbrev=r['abbrev']).exists():
self.stdout.write('Importing compound with abbreviation {} and name {}'.format(r['abbrev'], r['name']))
if r.get('custom') != '1':
try:
if r.get('CSID') not in ('', None):
c = Compound(CSID=r['CSID'], labGroup=l, abbrev=r['abbrev'])
c.csConsistencyCheck()
c.save()
else:
if r.get('CAS_ID') not in (None, ''):
CASResults = cs.simple_search(r['CAS_ID'])
else:
CASResults = []
if len(CASResults) != 1:
nameResults = cs.simple_search(r.get('name'))
if len(nameResults) != 1:
raise RuntimeError('Could not get unambiguous chemspider entry for CAS ID {} with name {}. Got {} responses'.format(r['CAS_ID'], r['name'], len(CASResults)))
else:
c = Compound(CSID=nameResults[0].csid, labGroup=l, abbrev=r['abbrev'])
c.csConsistencyCheck()
c.save()
else:
c = Compound(CSID=CASResults[0].csid, labGroup=l, abbrev=r['abbrev'])
c.csConsistencyCheck()
c.save()
except ValidationError as e:
c.delete()
raise e
else:
if r.get('INCHI') is None:
r['INCHI'] = ''
if r.get('smiles') is None:
r['smiles'] = ''
c = Compound.objects.get_or_create(labGroup=l, custom=True, name=r['name'], abbrev=r['abbrev'], formula=r['formula'], smiles=r['smiles'], INCHI=r['INCHI'])[0]
self.stdout.write(c.name.encode('utf-8'))
c.save()
with open(path.join(folder, 'compound_chemicalClasses.tsv')) as chemicalClasses:
reader = csv.DictReader(chemicalClasses, delimiter='\t')
for r in reader:
self.stdout.write('working with class {}'.format(r['chemicalClass.label']))
cs = Compound.objects.filter(abbrev=r['compound.abbrev'])
if cs.count() > 0:
c1 = ChemicalClass.objects.get_or_create(label=r['chemicalClass.label'])[0]
for c2 in cs:
if not c1 in c2.chemicalClasses.all():
c2.chemicalClasses.add(c1)
c2.save()
with open(path.join(folder, 'compoundquantities.tsv')) as cqs:
reader = csv.DictReader(cqs, delimiter='\t')
quantities = []
for r in reader:
try:
reaction = PerformedReaction.objects.get(reference=r['reaction.reference'].lower())
compound = Compound.objects.get(abbrev=r['compound.abbrev'], labGroup=reaction.labGroup)
if r['compound.abbrev'] in ('water', 'H2O'):
r['density'] = 1
mw = NumMolDescriptorValue.objects.get(compound=compound, descriptor__heading='mw').value
if r['compoundrole.name'] != 'pH':
self.stdout.write('adding {} to {}'.format(compound.abbrev, reaction.reference))
compoundrole = CompoundRole.objects.get_or_create(label=r['compoundrole.name'])[0]
if r['amount'] in ('', '?'):
amount = None
elif r['unit'] == 'g':
amount = float(r['amount'])/mw
elif r['unit'] == 'd':
amount = float(r['amount'])*0.0375*float(r['density'])/mw
elif r['unit'] == 'mL':
amount = float(r['amount'])*float(r['density'])/mw
else:
raise RuntimeError('invalid unit entered')
if amount is not None:
amount = (amount * 1000)
cqq = CompoundQuantity.objects.filter(role=compoundrole, compound=compound, reaction=reaction)
if cqq.count() > 1:
cqq.delete()
elif cqq.count() == 0:
quantity = CompoundQuantity(role=compoundrole, compound=compound, reaction=reaction)
quantity.amount = amount
#quantities.append(quantity)
quantity.save(recalculate=False)
else:
reaction.notes += ' pH adjusting reagent used: {}, {}{}'.format(r['compound.abbrev'], r['amount'], r['unit'])
reaction.save(calcDescriptors=False)
except Compound.DoesNotExist as e:
self.stderr.write('Unknown Reactant {} with amount {} {} in reaction {}'.format(r['compound.abbrev'], r['amount'], r['unit'], r['reaction.reference']))
reaction.notes += ' Unknown Reactant {} with amount {} {}'.format(r['compound.abbrev'], r['amount'], r['unit'])
reaction.valid = False
reaction.save(calcDescriptors=False)
except PerformedReaction.DoesNotExist as e:
raise e
class make_descs:
def __init__(self,argv):
''' load input arguments'''
self.CAS_file = argv[1]
''' my chemsphder token '''
self.cs=ChemSpider('d1778a9f-c41f-41f6-920e-fc6d9ff739ca')
def querySMILEs(self):
'''
Read CAS file as input
To calculate SMILEs through querying chemspider
'''
index_num=-1
self.CAS = []
self.SMILEs=[]
self.missing = []
with open (self.CAS_file,'rb') as csvfile:
csv_read=csv.reader(csvfile,delimiter=',')
for row in csv_read:
index_num += 1
row = row[0]
row = row.split('//')
row = row[0]
chem_this=self.cs.search(row)
try:
print "Working on chemical ", index_num, row
# raw_input(chem_this[0].smiles)
self.CAS.append(row)
self.SMILEs.append(chem_this[0].smiles)
except IndexError:
print "Can't find index: ", index_num, row
self.SMILEs.append(row)
self.missing.append(index_num)
continue
# delete missing rows
for index in sorted(self.missing,reverse=True):
del self.CAS[index]
del self.SMILEs[index]
# Clean up SMILEs
self._cleanSMILEs()
# Write SMILEs
self._writeSEMILs()
return self.SMILEs
def _writeSEMILs(self):
resultsfile=open('./cas/SMILEs.csv','wb')
wr=csv.writer(resultsfile,dialect='excel')
for eachSMILEs in self.SMILEs:
wr.writerow([eachSMILEs])
def _cleanSMILEs(self):
'''
Clean up SMILEs
Make sure there is not '+', "-", and "."
'''
assert self.SMILEs is not None
for indx, eachSMILEs in enumerate(self.SMILEs):
eachSMILEs = eachSMILEs.replace("+","")
eachSMILEs = eachSMILEs.replace("-","")
eachSMILEs = eachSMILEs.replace(".","")
self.SMILEs[indx] = eachSMILEs
print "SMILEs have been cleaned up!"
def calculateDescs(self):
'''
Use subprocess to call Dragon command line shell
And calculate predefined descriptors.
These descriptors are suited for the LCIA module at this time
'''
dragonShellCall = "./Dragon/dragon6shell.exe"
dragonScriptCall = "./Dragon/test_script.drs"
proc = sb.Popen([dragonShellCall,"-s" ,dragonScriptCall])
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)
import CoolProp
from chemspipy import ChemSpider
from chemspipy_key import key # private file with the key (DO NOT COMMIT!!)
import glob, json
cs = ChemSpider(key)
# Map from name to Chemspider ID
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']
# Tool uses the ChemSpiPy library to assist in accessing the ChemSpider Database
# Syntax to run command: python ChemSpider.py -(f/n) term
# -f name -> get the formula for the common name formula
# -n formula -> get the common name for the formula
import sys
# allows us to use command line arguments
if len(sys.argv) < 3:
print("Incorrect input.\n\t==> python ChemSpider.py [-f/-n] <argument>")
sys.exit()
from chemspipy import ChemSpider
cs = ChemSpider('3e05e0a6-9f49-4dff-ba0e-a9d6ca3d04ea')
# imports the ChemSpider api, and passes our access token to it
for result in cs.search(sys.argv[2])[:5]:
# Give the first five results for -f.
if sys.argv[1] == "-f":
print(result.common_name)
print(result.molecular_formula)
# print(result.common_name)
if sys.argv[1] == "-n":
print(result.common_name)
break
if not args.from_db:
from chemspipy import ChemSpider
if not args.export_db_only:
import pandas as pd
if args.from_db or args.export_db_only or args.export_db_csv:
import shelve
## ==================== set up chemspider ====================
if not args.from_db:
possiblefile = os.path.expanduser(args.token)
if os.path.exists(possiblefile): # is file
with open(possiblefile) as f:
csp = ChemSpider(f.read().strip())
else:
csp = ChemSpider(args.token) # else is token
else:
csp = None
spq = spiderquery(csp, args.prefix + '_p')
## ==================== list of compounds ====================
if args.inputfile:
with open(args.inputfile) as csvfile:
f = csv.reader(csvfile)
compounds = []
j = 0
for i, row in enumerate(f):
import os
from chemspipy import ChemSpider
from sgenlib import smiles
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Program to find molecules in ChemSpdie",)
parser.add_argument('filename',help="the filename of a list")
args = parser.parse_args()
if os.getenv("SPIDERKEY") is None :
print "SPIDERKEY environmental variable not set! Exit."
quit()
cs = ChemSpider(os.getenv("SPIDERKEY"))
molecules = []
with open(args.filename,"r") as f :
molecules = [line.strip() for line in f.readlines()]
for mol in molecules:
hits = cs.search(mol)
if len(hits) == 0 :
print mol+"\t!!"
else :
"""try :
print "//".join([h.common_name for h in hits])
except :
print mol+"\t!!!"
"""
if __name__ == '__main__':
argparser = argparse.ArgumentParser(description="Script to obtain SMILES for a solutes in a list")
argparser.add_argument('-db', '--db', help="the molecule database")
argparser.add_argument('-solvent', '--solvent', help="the solvent", default="water")
argparser.add_argument('-solutes','--solutes',help="the list of solutes")
args = argparser.parse_args()
db = dblib.SolvDb(filename=args.db,type="abs",filehandle="^0")
solutes = [s.strip() for s in open(args.solutes,'r').readlines()]
if os.getenv("SPIDERKEY") is None :
print "SPIDERKEY environmental variable not set! Exit."
quit()
cs = ChemSpider(os.getenv("SPIDERKEY"))
# Loop over all the database entries in the solute lists
n = 0
for entry in db.itersolutelist(args.solvent,solutes):
if os.path.exists(entry.FileHandle+".smi") : continue
hits = cs.search(entry.SoluteName)
if len(hits) > 0 :
smi = hits[0].smiles
with open(entry.FileHandle+".smi","w") as f :
f.write("%s\n"%smi)
else :
print entry.SoluteName, entry.FileHandle
n += 1
print "Looped over %d solutes"%n
from chemspipy import ChemSpider
from sgenlib import smiles
if __name__ == "__main__":
parser = argparse.ArgumentParser(
description="Program to build molecules", )
parser.add_argument('filename', help="the filename of a list")
args = parser.parse_args()
if os.getenv("SPIDERKEY") is None:
print "SPIDERKEY environmental variable not set! Exit."
quit()
cs = ChemSpider(os.getenv("SPIDERKEY"))
lines = []
with open(args.filename, "r") as f:
lines = [line.strip() for line in f.readlines()]
molecules = sorted(list(set(lines)),
cmp=lambda x, y: cmp(lines.index(x), lines.index(y)))
for mol in molecules:
hits = cs.search(mol)
if len(hits) == 0:
print mol + "\t!!"
else:
molsmiles = hits[0].smiles
mol2 = mol.strip()
mol2.replace(" ", "_")
from nose.tools import eq_, ok_, raises
import requests
import six
from chemspipy import ChemSpider, MOL2D, MOL3D, BOTH
from chemspipy.errors import ChemSpiPyAuthError, ChemSpiPyServerError
logging.basicConfig(level=logging.WARN)
logging.getLogger('chemspipy').setLevel(logging.DEBUG)
# Security token is retrieved from environment variables
CHEMSPIDER_SECURITY_TOKEN = os.environ['CHEMSPIDER_SECURITY_TOKEN']
# Chemspider instances with and without a security token
cs = ChemSpider(security_token=CHEMSPIDER_SECURITY_TOKEN)
cs2 = ChemSpider()
def test_no_security_token():
"""Test ChemSpider can be initialized with no parameters."""
eq_(cs2.security_token, None)
def test_security_token():
"""Test security token is set correctly when initializing ChemSpider"""
eq_(cs.security_token, CHEMSPIDER_SECURITY_TOKEN)
def test_chemspider_repr():
"""Test ChemSpider object repr."""
return random.choice(security_token)
else:
print("You need Security_token.txt providing security token. Please contact me as soon as.")
# 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))
from django.utils.text import slugify
from chemspipy import ChemSpider
from professor_oak.models import ScoreMixin
log = logging.getLogger(__name__)
# Load the chemspider API for accessing the RSC structure database
try:
cs_key = settings.CHEMSPIDER_KEY
except AttributeError:
log.warn('CHEMSPIDER_KEY not found in localsettings.py')
chemspider_api = None
else:
chemspider_api = ChemSpider(cs_key)
class Hazard(models.Model):
"""A hazard type as defined by the global harmonized system.
Attributes
----------
- pictogram : Image file that represents this image. If not
provided, we will look in `static_files/ghs_pictograms/` for one
that matches the `name` attribute
"""
PHYSICAL = 'p'
HEALTH = 'h'
PHYSICAL_AND_HEALTH = 'ph'
ENVIRONMENTAL = 'e'
from chemspipy import ChemSpider
cs = ChemSpider("CHEMSPIDER_API_KEY")
def remove_prefix(disctext):
"""Removes the command prefix from the passed string"""
cprefix = "!chem "
if disctext.content.startswith(cprefix):
return disctext.content[len(cprefix):]
def getCompound(id):
"""Gets a compound object from a compound name or id"""
slist = cs.search(id) # Returns list of compounds
compound = slist[0] # Fetches top item of list as compound object
return compound
import os
import pickle
import numpy as np
import sys
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "ms2ldaviz.settings_simon")
import django
import jsonpickle
django.setup()
from chemspipy import ChemSpider
from basicviz.models import *
if __name__ == '__main__':
cs = ChemSpider('b07b7eb2-0ba7-40db-abc3-2a77a7544a3d')
exp_name = sys.argv[1]
e = Experiment.objects.get(name = exp_name)
print e
docs = Document.objects.filter(experiment = e)
for doc in docs:
md = jsonpickle.decode(doc.metadata)
ik = md.get('InChIKey',md.get('inchikey',None))
print ik
# search in chemspi
results = cs.search(ik)
if len(results) > 0:
m = results[0].mol_2d
if len(m) > 0:
doc.mol_string = m
def find_common_name(inchikey):
# Try to find the common name for the compound, if not, return None.
name = None
if chemspikey:
cs = ChemSpider(chemspikey)
if (len(inchikey) > 0):
result = cs.search(inchikey)
if (len(result) == 1):
name = result[0].common_name
return name
