def GroupRef (d_matrix, k_in, p_filout, thresold_group, l_lig):
# !!!!!!!!!!!!!!
# return p_filout
d_group = {}
d_group[1] = []
# l unique PDB
l_PDB = d_matrix.keys ()
for PDB1 in d_matrix.keys () :
for PDB2 in d_matrix[PDB1].keys ():
if not PDB2 in l_PDB :
l_PDB.append (PDB2)
for PDB_in in l_PDB :
f = 0
for group in d_group.keys () :
# case first requet
if f == 1 :
break
if group == 1 and d_group[group] == [] :
d_group[group].append (PDB_in)
break
for PDB_classed in d_group[group] :
val = 0.0
try : val = float (d_matrix[PDB_classed][PDB_in][k_in])
except : val = float (d_matrix[PDB_in][PDB_classed][k_in])
if val >= thresold_group :
d_group[group].append (PDB_in)
f = 1
break
# flag -> in dico group
if f == 0 :
d_group[group + 1] = [PDB_in]
filout = open (p_filout, "w")
filout.write ("PDB\tGroup\tFamily\n")
for group in d_group :
for pdb in d_group[group] :
for lig in l_lig :
# print lig, "====="
family = analysis.findFamily(pdb, pathManage.findFamilyFile (lig))
family = family [-1]
if family != "None" :
filout.write (str (pdb) + "\t" + str (group) + "\t" + str (family) + "\n")
break
filout.close ()
return p_filout
def GroupRef(d_matrix, k_in, p_filout, thresold_group, l_lig):
# !!!!!!!!!!!!!!
# return p_filout
d_group = {}
d_group[1] = []
# l unique PDB
l_PDB = d_matrix.keys()
for PDB1 in d_matrix.keys():
for PDB2 in d_matrix[PDB1].keys():
if not PDB2 in l_PDB:
l_PDB.append(PDB2)
for PDB_in in l_PDB:
f = 0
for group in d_group.keys():
# case first requet
if f == 1:
break
if group == 1 and d_group[group] == []:
d_group[group].append(PDB_in)
break
for PDB_classed in d_group[group]:
val = 0.0
try:
val = float(d_matrix[PDB_classed][PDB_in][k_in])
except:
val = float(d_matrix[PDB_in][PDB_classed][k_in])
if val >= thresold_group:
d_group[group].append(PDB_in)
f = 1
break
# flag -> in dico group
if f == 0:
d_group[group + 1] = [PDB_in]
filout = open(p_filout, "w")
filout.write("PDB\tGroup\tFamily\n")
for group in d_group:
for pdb in d_group[group]:
for lig in l_lig:
# print lig, "====="
family = analysis.findFamily(pdb,
pathManage.findFamilyFile(lig))
family = family[-1]
if family != "None":
filout.write(
str(pdb) + "\t" + str(group) + "\t" + str(family) +
"\n")
break
filout.close()
return p_filout
def manageResult (l_ligand, name_final, l_out = []):
pr_result = pathManage.result("final_" + name_final)
# remove the folder
# pr_pi = pathManage.result("final/phosphates")
# pr_ribose = pathManage.result("final/ribose")
for name_lig in l_ligand :
l_p_smile = pathManage.findListSmileFile(name_lig)
p_file_famile = pathManage.findFamilyFile (name_lig)
for p_smile in l_p_smile :
if search("ribose", p_smile) and search(".txt", p_smile) and search("smile", p_smile):
arrangeResult.globalArrangement(pr_result, p_smile, p_file_famile, name_lig, l_out)
elif search("smile", p_smile) and search(".txt", p_smile) :
arrangeResult.globalArrangement(pr_result, p_smile, p_file_famile, name_lig, l_out)
return 1
def manageResult(l_ligand, name_final, l_out=[]):
pr_result = pathManage.result("final_" + name_final)
# remove the folder
# pr_pi = pathManage.result("final/phosphates")
# pr_ribose = pathManage.result("final/ribose")
for name_lig in l_ligand:
l_p_smile = pathManage.findListSmileFile(name_lig)
p_file_famile = pathManage.findFamilyFile(name_lig)
for p_smile in l_p_smile:
if search("ribose", p_smile) and search(
".txt", p_smile) and search("smile", p_smile):
arrangeResult.globalArrangement(pr_result, p_smile,
p_file_famile, name_lig, l_out)
elif search("smile", p_smile) and search(".txt", p_smile):
arrangeResult.globalArrangement(pr_result, p_smile,
p_file_famile, name_lig, l_out)
return 1
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()