def Builder(name_database, RX = 3.00, RFree = 0.25, one_PDB_by_lig = 0, debug = 1):
"""
Dataset Builder
in : - open file result of filter ligand PDB
out : - log file
- dataset file -> ligand with associated PDB
"""
if one_PDB_by_lig == 0 :
name_dataset = name_database + "/" + str (RX) + "_" + str (RFree) + "_multiPDB"
else :
name_dataset = name_database + "/" + str (RX) + "_" + str (RFree) + "_uniquePDB"
pr_database = pathManage.result(name_database)
pr_result = pathManage.result(name_dataset)
if debug : print "== Path result " + pr_result + "==\n"
# check dataSet exist !!!!!!
# short cut
l_file_dataset = pathManage.retriveDataSetFile (pr_result)
if len(l_file_dataset) != 0 :
return l_file_dataset
# load structure
d_lig_PDB = loadFile.LigandInPDB(pr_database + "resultLigandInPDB")
nb_lig = len(d_lig_PDB.keys())
print "NB ligand included database:", nb_lig
# print d_lig_PDB.keys().index("HSO") -> search index ligand
i = 0
while i < nb_lig:
name_lig = d_lig_PDB.keys()[i]
#######################################
# step 1 search chemical substructure #
#######################################
PDB_ref = d_lig_PDB[name_lig][0]
if debug : print PDB_ref, name_lig, i, nb_lig
# if not possible to load the ligand -> remove lig
try : l_atom_lig_ref = loadFile.ligandInPDBConnectMatrixLigand(PDB_ref, name_lig)
except :
if debug == 1 : print "Exit => load ligand-l59"
del d_lig_PDB[name_lig]
nb_lig = nb_lig - 1
continue
# search substructure interest
l_interest_sub = searchPDB.interestStructure(l_atom_lig_ref) # search interest structure
if debug : print "Interest substructures in " + str(name_lig) + "-" + str (PDB_ref) + " " + "-".join(l_interest_sub)
if l_interest_sub == []:
if debug == 1 : print "Exit => Not substructure-l68"
del d_lig_PDB[name_lig]
nb_lig = nb_lig - 1
continue
#######################################################
# Step 2 Control quality of PDB + ligand hooked + option one #
#######################################################
else :
# control dataset quality
if debug : print "List PDB checked -> ", d_lig_PDB[name_lig]
l_PDB = checkPDBfile.CheckComplexQuality(d_lig_PDB[name_lig], name_lig, RX, RFree, one_PDB_by_lig)
# remove the entrance key with the ligand
if l_PDB == []:
if debug == 1 : print "Exit => Not No PDB selected-l82"
del d_lig_PDB[name_lig]
nb_lig = nb_lig - 1
continue
else :
d_lig_PDB[name_lig] = l_PDB
i = i + 1
if debug == 1 : print "Number of ligand selected =>", nb_lig
# structure and file dataset and control RX + length bond
WriteDataset (d_lig_PDB, pr_result)
return Builder(name_database, RX , RFree , one_PDB_by_lig , debug = 1)
ligandWithPDB = loadFile.resultLigandPDB(repertory.resultSerineProtease() + "resultLigandInPDB")
listLigand = ligandWithPDB.keys()
print listLigand
nbLigand = len(listLigand)
print nbLigand
j = 0
while j < nbLigand :
ligand = listLigand[j]
print ligand , j
pdb = ligandWithPDB[ligand][0].split("_")[2]
pdb = lower(pdb)
atomLigand = loadFile.ligandInPDBConnectMatrixLigand(pdb, ligand)
writePDBfile.globalStruct(ligand + "_" + pdb + ".pdb", atomLigand)
nbAtom = len(atomLigand)
# print atomLigand
searchPDB.cycleGlobal(atomLigand)
# print atomLigand
listResult = []
for atom in atomLigand :
if atom["element"] == "N" or atom["element"] == "O" or atom["element"] == "C": # impact point
if atom["cycle"] == 1 :
cycleRetrieve = retrieveAtom.cycle(atom, atomLigand)
listResult.append(cycleRetrieve)
continue
for serialAtom in atom["connect"] :
atomRetrieve = retrieveAtom.serial(serialAtom, atomLigand)
def GlobalBondLength (name_database, RX_thresold = 1.5):
# directory
pr_result = pathManage.result(name_database + "/CXbound" + str (RX_thresold))
pr_database = pathManage.result(name_database)
# filout with distance
p_CN = pr_result + "distanceCN"
p_CO = pr_result + "distanceCO"
p_CC = pr_result + "distanceCC"
p_coplar = pr_result + "distanceCoplar"
filout_CN = open (p_CN, "w")
filout_CO = open (p_CO, "w")
filout_CC = open (p_CC, "w")
filout_coplar = open (p_coplar, "w")
# load PDB with logand
if not path.exists(pr_database + "resultLigandInPDB") :
print "ERROR => file with ligand and PDB does not exist"
return
else :
d_lig_PDB = loadFile.LigandInPDB(pr_database + "resultLigandInPDB")
nb_lig = len(d_lig_PDB.keys())
print d_lig_PDB.keys()
i = 0
while (i < nb_lig):
name_lig = d_lig_PDB.keys()[i]
l_PDB = d_lig_PDB[name_lig]
for PDB in l_PDB :
# controle RX
RX = parsing.Quality(PDB)[0]
# print RX
if RX <= RX_thresold :
l_atom_lig = loadFile.ligandInPDBConnectMatrixLigand(PDB, name_lig)
l_distCN = BondLengthCandX (l_atom_lig, "N")
l_distCO = BondLengthCandX (l_atom_lig, "O")
l_distCC = BondLengthCandX (l_atom_lig, "C")
l_coplarIII = CoplanarityIII(l_atom_lig)
if l_distCN != [] :
filout_CN.write ("\n".join (l_distCN) + "\n")
if l_distCO != [] :
filout_CO.write ("\n".join (l_distCO) + "\n")
if l_distCC != [] :
filout_CC.write ("\n".join (l_distCC) + "\n")
if l_coplarIII != [] :
filout_coplar.write ("\n".join (l_coplarIII) + "\n")
# take only one PDB by ligand not more
i = i + 1
continue
i = i + 1
filout_CO.close ()
filout_CN.close ()
filout_CC.close ()
filout_coplar.close ()
runScriptR.histDistance(p_CN, "CN")
runScriptR.histDistance(p_CO, "CO")
runScriptR.histDistance(p_CC, "CC")
runScriptR.histDistance(p_coplar, "coplar")
