def extractStep(self):
listIds = {}
smallMolID = "none"
for item in self.inputListID.get():
if hasattr(item, "_iteractsWithPDBId"):
tokens = item._iteractsWithPDBId.get().split(";")
for token in tokens:
pdbId = token.strip()
if not pdbId in listIds:
listIds[pdbId] = []
chemId = item.getDbId()
if hasattr(item, "_PDBChemId"):
chemId = item._PDBChemId.get()
smallMolID = "pdbchem"
listIds[pdbId].append(chemId)
outputDatabaseID = SetOfDatabaseID().create(path=self._getPath(),
suffix='PDBs')
for pdbId in listIds:
pdb = DatabaseID()
pdb.setDatabase("pdb")
pdb.setDbId(pdbId)
pdb._pdbId = pwobj.String(pdbId)
pdb._PDBLink = pwobj.String("https://www.rcsb.org/structure/%s" %
pdbId)
aux = " ; ".join(listIds[pdbId])
if smallMolID == "none":
pdb._interactsWithChemId = pwobj.String(aux)
elif smallMolID == "pdbchem":
pdb._interactsWithPDBChemId = pwobj.String(aux)
outputDatabaseID.append(pdb)
self._defineOutputs(outputPDBs=outputDatabaseID)
self._defineSourceRelation(self.inputListID, outputDatabaseID)
def searchStep(self):
outputDatabaseID = SetOfDatabaseID().create(path=self._getPath())
for item in self.inputListID.get():
newItem = DatabaseID()
newItem.copy(item)
newItem._uniprotId = pwobj.String("Not available")
newItem._uniprotLink = pwobj.String("Not available")
pdbId = item._pdbId.get()
print("Processing %s" % pdbId)
urlId = "https://www.rcsb.org/pdb/rest/das/pdb_uniprot_mapping/alignment?query=%s" % pdbId
if hasattr(item, "_chain"):
urlId += "." + item._chain.get().upper()
fnXml = self._getExtraPath("%s.xml" % pdbId)
if not os.path.exists(fnXml):
print("Fetching uniprot: %s" % urlId)
for i in range(3):
try:
urllib.request.urlretrieve(urlId, fnXml)
break
except: # The library raises an exception when the web is not found
pass
if os.path.exists(fnXml):
try:
tree = ET.parse(fnXml)
# print(ET.tostring(tree, pretty_print=True))
uniprotId = None
for child in tree.getroot().iter():
if child.tag.endswith("alignObject"):
if child.attrib['dbSource'] == "UniProt":
uniprotId = child.attrib['dbAccessionId']
break
if uniprotId:
newItem._uniprotId = pwobj.String(uniprotId)
newItem._uniprotLink = pwobj.String(
"https://www.uniprot.org/uniprot/%s" % uniprotId)
except:
print(" Cannot parse the Uniprot XML: %s" % fnXml)
outputDatabaseID.append(newItem)
self._defineOutputs(outputUniprot=outputDatabaseID)
self._defineSourceRelation(self.inputListID, outputDatabaseID)
def searchStep(self):
outputDatabaseID = SetOfDatabaseID().create(path=self._getPath())
fnList = []
for item in self.inputListID.get():
newItem = DatabaseID()
newItem.copy(item)
newItem._uniprotFile = pwobj.String("Not available")
newItem._unitprotSeqLength = pwobj.Integer(-1)
uniprotId = item._uniprotId.get()
print("Processing %s" % uniprotId)
urlId = "https://www.uniprot.org/uniprot/%s.fasta" % uniprotId
fnFasta = self._getExtraPath("%s.fasta" % uniprotId)
if not os.path.exists(fnFasta):
print("Fetching uniprot: %s" % urlId)
for i in range(3):
try:
urllib.request.urlretrieve(urlId, fnFasta)
if not fnFasta in fnList:
fnList.append(fnFasta)
break
except: # The library raises an exception when the web is not found
pass
if os.path.exists(fnFasta):
newItem._uniprotFile = pwobj.String(fnFasta)
newItem._unitprotSeqLength = pwobj.Integer(
sequenceLength(fnFasta))
outputDatabaseID.append(newItem)
fnAll = self._getPath("sequences.fasta")
with open(fnAll, 'w') as outfile:
for fname in fnList:
with open(fname) as infile:
for line in infile:
outfile.write(line)
outfile.write('\n\n')
seqFile = ProteinSequenceFile()
seqFile.setFileName(fnAll)
self._defineOutputs(outputUniprot=outputDatabaseID)
self._defineSourceRelation(self.inputListID, outputDatabaseID)
self._defineOutputs(outputSequence=seqFile)
self._defineSourceRelation(self.inputListID, seqFile)
def constructOutput(self, fnTxt):
fnDir, fnResults = os.path.split(fnTxt)
tokens = fnResults.split('-')
if len(tokens) > 1:
subset = tokens[1].split('.')[0]
else:
subset = ""
outputSet = SetOfDatabaseID.create(path=self._getPath(), suffix=subset)
for line in open(fnTxt, "r"):
line = line.strip()
if line == "":
continue
elif line.startswith("# Structural equivalences"):
break
elif line.startswith("#"):
continue
else:
tokens = line.split()
pdbId = DatabaseID()
tokens2 = tokens[1].split('-')
pdbId.setDatabase("pdb")
pdbId.setDbId(tokens[1])
pdbId._pdbId = pwobj.String(tokens2[0])
if len(tokens2) > 1:
pdbId._chain = pwobj.String(tokens2[1])
pdbId._PDBLink = pwobj.String(
"https://www.rcsb.org/structure/%s" % tokens2[0])
pdbId._DaliZscore = pwobj.Float(float(tokens[2]))
pdbId._DaliRMSD = pwobj.Float(float(tokens[3]))
pdbId._DaliSuperpositionLength = pwobj.Integer(int(tokens[4]))
pdbId._DaliSeqLength = pwobj.Integer(int(tokens[5]))
pdbId._DaliSeqIdentity = pwobj.Float(float(tokens[6]))
pdbId._DaliDescription = pwobj.String(" ".join(tokens[7:]))
outputSet.append(pdbId)
outputDict = {'outputDatabaseIds%s' % subset: outputSet}
self.protocol._defineOutputs(**outputDict)
self.protocol._defineSourceRelation(self.protocol.inputStructure,
outputSet)
def operateStep(self):
outputDict = {}
if self.operation.get() == 1:
# Union
for database in self.multipleInputListID:
for databaseEntry in database.get():
add = True
if self.removeDuplicates.get():
add = not databaseEntry.getDbId() in outputDict
if add:
dbEntry = DatabaseID()
dbEntry.copy(databaseEntry, copyId=False)
outputDict[databaseEntry.getDbId()] = dbEntry
elif self.operation.get() == 0 or self.operation.get(
) == 2 or self.operation.get() == 3:
# Unique, Intersection, Difference
outputList2 = []
if self.operation.get() == 2 or self.operation.get() == 3:
for databaseEntry in self.inputListID2.get():
outputList2.append(databaseEntry.getDbId())
for databaseEntry in self.inputListID.get():
add = False
if self.operation.get() == 0: # Unique
add = not databaseEntry.getDbId() in outputDict
elif self.operation.get() == 2: # Intersection
add = databaseEntry.getDbId() in outputList2
if self.removeDuplicates.get():
add = add and not databaseEntry.getDbId() in outputDict
elif self.operation.get() == 3: # Difference
add = not databaseEntry.getDbId() in outputList2
if self.removeDuplicates.get():
add = add and not databaseEntry.getDbId() in outputDict
if add:
dbEntry = DatabaseID()
dbEntry.copy(databaseEntry)
outputDict[databaseEntry.getDbId()] = dbEntry
elif self.operation.get() == 4:
# Change ID
newLabel = True
for name, _ in self.inputListID.get().getFirstItem().getAttributes(
):
if self.newDb.get() == name:
newLabel = False
break
for databaseEntry in self.inputListID.get():
dbEntry = DatabaseID()
dbEntry.copy(databaseEntry)
if hasattr(dbEntry, self.newDbId.get()):
if newLabel:
dbEntry.setDatabase(self.newDb.get())
else:
dbEntry.setDatabase(
dbEntry.getAttributeValue(self.newDb.get()))
dbEntry.setDbId(
dbEntry.getAttributeValue(self.newDbId.get()))
add = True
if self.removeDuplicates.get():
add = add and not dbEntry.getDbId() in outputDict
if add:
outputDict[dbEntry.getDbId()] = dbEntry
elif self.operation.get() == 5:
# Keep columns
keepList = [x.strip() for x in self.keepColumns.get().split()]
keepList.append("database")
keepList.append("dbId")
ignoreList = []
for name, _ in self.inputListID.get().getFirstItem().getAttributes(
):
if not name in keepList:
ignoreList.append(name)
for databaseEntry in self.inputListID.get():
dbEntry = DatabaseID()
dbEntry.copy(databaseEntry, ignoreAttrs=ignoreList)
add = True
if self.removeDuplicates.get():
add = add and not dbEntry.getDbId() in outputDict
if add:
outputDict[dbEntry.getDbId()] = dbEntry
elif self.operation.get() == 6:
# Filter columns
referenceValue = self.filterValue.get()
value = self.inputListID.get().getFirstItem().getAttributeValue(
self.filterColumn.get())
if isinstance(value, float):
referenceValue = float(referenceValue)
elif isinstance(value, int):
referenceValue = int(referenceValue)
for databaseEntry in self.inputListID.get():
dbEntry = DatabaseID()
dbEntry.copy(databaseEntry)
add = False
value = dbEntry.getAttributeValue(self.filterColumn.get())
if isinstance(value, Float):
value = float(value)
elif isinstance(value, Integer):
value = int(value)
filterOp = self.filterOp.get()
if filterOp == 0: # ==
add = value == referenceValue
elif filterOp == 1: # >
add = value > referenceValue
elif filterOp == 2: # >=
add = value > referenceValue
elif filterOp == 3: # <
add = value < referenceValue
elif filterOp == 4: # <=
add = value <= referenceValue
elif filterOp == 5: # !=
add = value != referenceValue
elif filterOp == 6: #startswith
add = value.startswith(referenceValue)
elif filterOp == 7: # endswith
add = value.endswith(referenceValue)
elif filterOp == 8: # contains
add = referenceValue in value
elif filterOp == 9: # does not startswith
add = not (value.startswith(referenceValue))
elif filterOp == 10: # does not endswith
add = not (value.endswith(referenceValue))
elif filterOp == 11: # does not contains
add = not (referenceValue in value)
if self.removeDuplicates.get():
add = add and not dbEntry.getDbId() in outputDict
if add:
outputDict[dbEntry.getDbId()] = dbEntry
outputDatabaseID = SetOfDatabaseID().create(path=self._getPath())
for dbId in outputDict:
outputDatabaseID.append(outputDict[dbId])
self._defineOutputs(output=outputDatabaseID)
self._defineSourceRelation(self.inputListID, outputDatabaseID)