def extractReference(p_list_ligand, p_dir_dataset, p_dir_result, substruct):
# struct reference
d_dataset = {}
# retrieve list of ligand in PDB
d_ligand = tool.parseLigandPDBList(p_list_ligand)
# download PDB and fasta associated
l_p_PDB = []
l_p_fasta = []
for PDB_ID in d_ligand[substruct]:
PDB_ID = PDB_ID.upper()
p_pdb = downloadFile.importPDB(PDB_ID,
p_dir_dataset,
dir_by_PDB=1,
debug=1,
dbPDB="/home/borrel/PDB/")
p_fasta = downloadFile.importFasta(
PDB_ID,
p_dir_dataset,
dir_by_PDB=1,
debug=1,
fastaGlobal="/home/borrel/Yue_project/pdb_seqres.txt")
if p_pdb != 0 and p_fasta != 0:
l_p_pdb_chain = separeByChain(p_pdb)
l_p_fasta_chain = separeChainFasta(p_fasta)
d_dataset[PDB_ID] = {}
d_dataset[PDB_ID]["p_pdb"] = p_pdb
d_dataset[PDB_ID]["p_fasta"] = p_fasta
d_dataset[PDB_ID]["p_pdb_chain"] = l_p_pdb_chain
d_dataset[PDB_ID]["p_fasta_chain"] = l_p_fasta_chain
d_dataset[PDB_ID]["conserve"] = 1
# plot resolution
p_file_RX = p_dir_result + "resolution_ref.txt"
file_RX = open(p_file_RX, "w")
for PDB_ID in d_dataset.keys():
RX = parsePDB.resolution(d_dataset[PDB_ID]["p_pdb"])
try:
d_dataset[PDB_ID]["RX"] = float(RX)
except:
d_dataset[PDB_ID]["RX"] = 100.0
file_RX.write(PDB_ID + "\t" + str(RX) + "\n")
file_RX.close()
runOtherSoft.Rhistogram(p_file_RX, "RX_ref_no_filter")
return d_dataset
def extractReference (p_list_ligand, p_dir_dataset, p_dir_result, substruct):
# struct reference
d_dataset = {}
# retrieve list of ligand in PDB
d_ligand = tool.parseLigandPDBList (p_list_ligand)
# download PDB and fasta associated
l_p_PDB = []
l_p_fasta = []
for PDB_ID in d_ligand[substruct] :
PDB_ID = PDB_ID.upper()
p_pdb = downloadFile.importPDB(PDB_ID, p_dir_dataset, dir_by_PDB = 1, debug = 1, dbPDB = "/home/borrel/PDB/" )
p_fasta = downloadFile.importFasta(PDB_ID, p_dir_dataset, dir_by_PDB = 1, debug = 1, fastaGlobal = "/home/borrel/Yue_project/pdb_seqres.txt")
if p_pdb != 0 and p_fasta != 0 :
l_p_pdb_chain = separeByChain (p_pdb)
l_p_fasta_chain = separeChainFasta(p_fasta)
d_dataset[PDB_ID] = {}
d_dataset[PDB_ID] ["p_pdb"] = p_pdb
d_dataset[PDB_ID] ["p_fasta"] = p_fasta
d_dataset[PDB_ID] ["p_pdb_chain"] = l_p_pdb_chain
d_dataset[PDB_ID] ["p_fasta_chain"] = l_p_fasta_chain
d_dataset[PDB_ID] ["conserve"] = 1
# plot resolution
p_file_RX = p_dir_result + "resolution_ref.txt"
file_RX = open (p_file_RX, "w")
for PDB_ID in d_dataset.keys () :
RX = parsePDB.resolution(d_dataset[PDB_ID]["p_pdb"])
try : d_dataset[PDB_ID] ["RX"] = float(RX)
except : d_dataset[PDB_ID] ["RX"] = 100.0
file_RX.write (PDB_ID + "\t" + str (RX) + "\n")
file_RX.close ()
runOtherSoft.Rhistogram (p_file_RX, "RX_ref_no_filter")
return d_dataset
def qualityExtraction (l_ligand, name_folder, p_list_ligand, thresold_sheap) :
pr_result = pathManage.result("final_" + name_folder)
filout = open(pr_result + "quality_extraction.txt", "w")
# number PDB by ligand, without filter
filout.write ("Number PDB by ligand:\n")
d_dataset = tool.parseLigandPDBList(p_list_ligand)
for ligand in l_ligand :
filout.write (str (ligand) + ": " + str (len (d_dataset[ligand])) + "\n")
# number references
filout.write ("\n*************\n\nNumber references by ligands:\n")
for ligand in l_ligand :
pr_result_ligand = pathManage.result(ligand)
nb_ref = -2
l_file = listdir(pr_result_ligand)
for f in l_file :
if path.isdir (pr_result_ligand + "/" + f) :
nb_ref = nb_ref + 1
filout.write (ligand + ": " + str (nb_ref) + "\n")
# number of query by ref in means and max and min (after blast)
filout.write ("\n*************\n\nNumber means queries by references:\n")
p_family_all = pathManage.result() + "reference_family_all.txt"
filout_family_all = open (p_family_all, "w")
d_family_all = {}
for ligand in l_ligand :
d_nb_query = {}
d_family = {}
p_filout_family = pathManage.result() + "reference_family_" + ligand + ".txt"
p_filout_family_count = pathManage.result () + "count_family_" + ligand + ".txt"
filout_family = open (p_filout_family, "w")
filout_family_count = open (p_filout_family_count, "w")
pr_result_ligand = pathManage.result(ligand)
nb_ref = 0
l_file = listdir(pr_result_ligand)
for f in l_file :
if path.isdir (pr_result_ligand + "/" + f) and len (f) == 4:
# count by family
family_ref = analysis.findFamily(f, pathManage.findFamilyFile (ligand))
filout_family.write ("\t".join (family_ref) + "\n")
if not family_ref[-1] in d_family.keys () :
d_family[family_ref[-1]] = 0
d_family[family_ref[-1]] = d_family[family_ref[-1]] + 1
# file all
if not family_ref[-1] in d_family_all.keys () :
d_family_all[family_ref[-1]] = 0
d_family_all[family_ref[-1]] = d_family_all[family_ref[-1]] + 1
# count number of references
nb_ref = nb_ref + 1
d_nb_query[f] = 0
l_file_queries = listdir(pr_result_ligand + "/" + f + "/")
for file_query in l_file_queries :
if search ("CX",file_query) :
d_nb_query[f] = d_nb_query[f] + 1
filout.write (ligand + ": " + str(np.sum(d_nb_query.values ())) + "\n")
filout.write (ligand + ": " + str(np.mean(d_nb_query.values ())) + "+/-" + str(np.std (d_nb_query.values ())) + "\n")
filout.write ("MAX " + str (ligand) + ": " + str (max (d_nb_query.values ())) + " " + str (d_nb_query.keys ()[d_nb_query.values ().index (max (d_nb_query.values ()))]) +"\n")
# family
filout_family_count.write ("\t".join(d_family.keys ()) + "\n")
l_values = [str(x) for x in d_family.values ()]
filout_family_count.write ("\t".join(l_values) + "\n")
filout_family.close ()
filout_family_count.close ()
runOtherSoft.piePlot(p_filout_family_count)
# all family
filout_family_all.write ("\t".join(d_family_all.keys ()) + "\n")
l_values = [str(x) for x in d_family_all.values ()]
filout_family_all.write ("\t".join(l_values) + "\n")
filout_family_all.close ()
runOtherSoft.piePlot(p_family_all)
# number subref by ligand
filout.write ("\n*************\n\nNumber of subref considered:\n")
for ligand in l_ligand :
d_nb_sub = {}
d_nb_sub_sheap = {}
pr_result_ligand = pathManage.result(ligand)
l_ref = listdir(pr_result_ligand)
for ref in l_ref :
if path.isdir (pr_result_ligand + "/" + ref) and len (ref) == 4:
l_file_queries = listdir(pr_result_ligand + "/" + ref + "/")
for file_query in l_file_queries :
if search ("substituent",file_query) and search (".pdb",file_query):
atom_substituate = file_query.split ("_")[-2]
try : value_sheap = float(file_query.split ("_")[-1][:-4])
except : continue
if not atom_substituate in d_nb_sub.keys () :
d_nb_sub[atom_substituate] = 0
d_nb_sub[atom_substituate] = d_nb_sub[atom_substituate] + 1
if value_sheap > thresold_sheap :
if not atom_substituate in d_nb_sub_sheap :
d_nb_sub_sheap[atom_substituate] = 0
d_nb_sub_sheap[atom_substituate] = d_nb_sub_sheap[atom_substituate] + 1
filout.write ("\n" + ligand + "\n")
for atom_substituate in d_nb_sub.keys () :
filout.write (atom_substituate + ": " + str (d_nb_sub[atom_substituate]) + "\n")
try : filout.write (atom_substituate + " ShaEP: " + str (d_nb_sub_sheap[atom_substituate]) + "\n")
except : filout.write (atom_substituate + " ShaEP: 0\n")
filout.close()
