def readData(propCod):
"""Reads files with data extracted from DBS.
:param propCod: (str) Property code.
:return:
tables: (dict) Tables of property and source (LARS code).
{property code {LARS code: [values]}}
"""
tables = {}
for prop in propCod:
tables[prop] = {}
for larsCod in propCod[prop]:
fileName = getFileName(prop, larsCod)
if not os.path.exists(fileName):
continue
try:
tab = pd.read_csv(fileName)
except FileNotFoundError:
raise myExceptions.NoFile(fileName)
tables[prop][larsCod] = tab
return tables
def readFieFile(fileName):
"""Reads family isomer enumeration (fie) file.
:param fileName: (str) File name.
:return:
df: (pandas DataFrame) Table with ENU code, molecular formula and SMILEs string.
"""
if not os.path.exists(fileName):
raise myExceptions.NoFile(fileName)
# df = pd.read_csv(fileName, sep='\s+', comment='#', names=['nam', 'frm', 'smiles'])
df = pd.read_csv(fileName, sep='\s+', names=['nam', 'frm', 'smiles'])
return df
def run():
"""Writes identifiers to txt file."""
nArgs = len(sys.argv)
if nArgs != 3 and nArgs != 4:
raise myExceptions.ArgError('3 or 4', nArgs)
name = sys.argv[2]
name = name.split('/')[-1]
enuMolFile = 'out/{}.json'.format(name)
if not os.path.exists(enuMolFile):
raise myExceptions.NoFile(enuMolFile)
with open(enuMolFile) as jsonFile:
enuData = json.load(jsonFile, object_hook=moleculeDecoder)
if len(sys.argv) == 4:
fieFile = sys.argv[3]
isomers = IO.readFieFile(fieFile)
isomers = utils.canonicalizeSmiles(isomers)
enuData = selectMolecules(isomers, enuData)
outFileName = 'out/00_{}.lst'.format(name)
out = open(outFileName, 'w')
for smiles in enuData:
mol = enuData[smiles]
# frm = mol.form
frm = mol.form_pcp
cas = mol.cas
smiles = mol.smiles
if not cas:
cas = '%'
name = mol.name_pcp
if not name:
name = '%'
inchi = mol.inchi_pcp
name = name.replace(' ', '_')
out.write('{:10} {:12} {:40} {:30} {}\n'.format(
frm, cas, name, inchi, smiles))
out.close()
def getDbsEntries(fileName):
"""Reads dbs entries from file.
:param fileName: (str) File name.
:return:
dbsEntries: (dict) Dictionary of DBS entries.
"""
if not os.path.exists(fileName):
raise myExceptions.NoFile(fileName)
with open(fileName) as jsonFile:
dbsEntries = json.load(jsonFile, object_hook=dbsEntryDecoder)
return dbsEntries
def __init__(self, fileName):
"""Constructs all the necessary attributes for the dbsConfiguration object.
A configuration file consists of sections, lead by a "[section]" header,
and followed by "name: value" entries, with continuations and such in
the style of RFC 822.
See 'configparser' documentation for more information.
:param fileName: (str) File name from which configuration is read.
"""
if not os.path.exists(fileName):
raise myExceptions.NoFile(fileName)
config = configparser.ConfigParser()
config.read(fileName)
self.config = config
def run(dbsConfig):
"""Get identifiers and all available data for a given molecule in DBS
and save data to data/ directory.
:param dbsConfig: (dbsConfiguration object) DBS configuration object.
"""
nArgs = len(sys.argv)
if nArgs != 3:
raise myExceptions.ArgError(nArgs, 3)
# I decided to keep both files (molListFile and enuMolFile)
# because I can easily comment out entries in molListFile,
# because of "#" in smiles strings I cannot use this symbol to start a comment.
# 00_file.lst
molListFile = sys.argv[2]
# molList = np.genfromtxt(molListFile, dtype=None, encoding='utf-8')
if not os.path.exists(molListFile):
raise myExceptions.NoFile(molListFile)
molList = pd.read_csv(molListFile, sep='\s+', header=None, names=['frm', 'cas', 'nam', 'inchi', 'smiles'])
dbsFileName = dbsConfig.getDbsFileName()
dbsEntries = dbs.getDbsEntries(dbsFileName)
print('Getting Identifiers')
path = dbsConfig.getPath()
molList = getCids(dbsEntries, molList, path)
# getCidsofSynonyms(dbsEntries, molList)
print('Getting Data')
propCod = dbsConfig.getPropCod()
tables = getData(dbsEntries, molList, path, propCod)
print('Writing Data')
writeData(tables)