def applyTMAlignList(l_pr_ref, pr_out):
pathManage.generatePath(pr_out)
nb_pr_ref = len(l_pr_ref)
d_out = {}
i = 0
while i < nb_pr_ref:
j = i + 1
PDB1 = l_pr_ref[i].split("/")[-1][0:4]
# print PDB1
while j < nb_pr_ref:
PDB2 = l_pr_ref[j].split("/")[-1][0:4]
# folder TM align
pr_alignement = pr_out + PDB1 + "__" + PDB2 + "/"
#print pr_alignement
#print PDB1,i, PDB2,j
#print l_pr_ref[i]
# RUN
out_file = runOtherSoft.runTMalign(l_pr_ref[i], l_pr_ref[j],
pr_alignement)
# clean folders -> pb with several run -> clean too fast -> try / except
try:
CleanResultTMalign(pr_alignement)
except:
pass
# parse result
if not PDB1 in d_out.keys():
if not PDB2 in d_out.keys():
d_out[PDB1] = {}
d_out[PDB1][PDB2] = parseTMalign.parseOutputTMalign(
out_file[-1])
else:
d_out[PDB2][PDB1] = {}
d_out[PDB2][PDB1] = parseTMalign.parseOutputTMalign(
out_file[-1])
else:
d_out[PDB1][PDB2] = {}
d_out[PDB1][PDB2] = parseTMalign.parseOutputTMalign(
out_file[-1])
j = j + 1
i = i + 1
return d_out
def Superimpose(self, refFrame):
prSuperimposed = self.dMD["prRMSD"] + "superimpose/"
pathFolder.createFolder(prSuperimposed)
self.prSuperMatrix = prSuperimposed
# control to superimposition file exist
nbmatrix = float(len(listdir(prSuperimposed)))
nbth = (float(self.MDtime) / float(self.timeframe)) / float(
self.stepFrame) - 1
if nbmatrix == nbth:
print prSuperimposed
print "Full folder"
return
prtemp = self.dMD["prRMSD"] + "temp/"
pathFolder.createFolder(prtemp)
pTMalignScore = self.dMD["prRMSD"] + "TMalignScore"
if not path.exists(pTMalignScore):
TMalignScore = open(pTMalignScore, "w")
TMalignScore.write("Frame\tRMSD\tTMscore\n")
else:
TMalignScore = open(pTMalignScore, "a")
i = self.stepFrame
imax = float(self.MDtime) / float(self.timeframe)
while i < imax:
print i, imax
nframe1 = str("%05d" % (refFrame))
nframe2 = str("%05d" % (i))
pframe1 = self.dMD["prframe"] + "frame_" + nframe1 + ".pdb"
pframe2 = self.dMD["prframe"] + "frame_" + nframe2 + ".pdb"
# control if existing
pmatrix = prSuperimposed + str(nframe1) + "_" + str(nframe2)
if path.exists(pmatrix) and path.getsize(pmatrix) > 0:
i += self.stepFrame
continue
else:
pathFolder.createFolder(prtemp, clean=1) # clean folder temp
lsuperimposed = runExternalSoft.runTMalign(
pframe2, pframe1, prtemp)
dalign = parseTMalign.parseOutputTMalign(lsuperimposed[-1])
TMalignScore.write(
str(nframe2) + "\t" + str(dalign["RMSD"]) + "\t" +
str(dalign["TMscore1"]) + "\n")
move(lsuperimposed[-2], pmatrix)
i += self.stepFrame
TMalignScore.close()
def applyTMAlignList (l_pr_ref, pr_out):
pathManage.generatePath(pr_out)
nb_pr_ref = len (l_pr_ref)
d_out = {}
i = 0
while i < nb_pr_ref :
j = i + 1
PDB1 = l_pr_ref[i].split ("/")[-1][0:4]
# print PDB1
while j < nb_pr_ref :
PDB2 = l_pr_ref[j].split ("/")[-1][0:4]
# folder TM align
pr_alignement = pr_out + PDB1 + "__" + PDB2 + "/"
#print pr_alignement
#print PDB1,i, PDB2,j
#print l_pr_ref[i]
# RUN
out_file = runOtherSoft.runTMalign(l_pr_ref[i], l_pr_ref[j], pr_alignement)
# clean folders -> pb with several run -> clean too fast -> try / except
try : CleanResultTMalign (pr_alignement)
except : pass
# parse result
if not PDB1 in d_out.keys () :
if not PDB2 in d_out.keys () :
d_out[PDB1] = {}
d_out[PDB1][PDB2] = parseTMalign.parseOutputTMalign(out_file[-1])
else :
d_out[PDB2][PDB1] = {}
d_out[PDB2][PDB1] = parseTMalign.parseOutputTMalign(out_file[-1])
else :
d_out[PDB1][PDB2] = {}
d_out[PDB1][PDB2] = parseTMalign.parseOutputTMalign(out_file[-1])
j = j + 1
i = i + 1
return d_out
def analysisBS (name_lig, ID_seq = '0.0', debug = 1):
pr_result = pathManage.result(name_lig)
pr_out = pathManage.result(name_lig + "/sameBS")
# log files
p_log_file = pr_out + "log.txt"
filout_log = open (p_log_file, "w")
# dictionnar with files
d_file_BS = {}
d_file_BS["global"] = open (pr_out + name_lig + "_", "w")
d_file_BS["global"].write ("name_bs\tRMSD_prot\tRMSD_BS_ca\tRMSD_BS_all\tD_max\tl_at_BS\tidentic\n")
d_file_BS["summary"] = open (pr_out + "summary.txt", "w")
pr_dataset = pathManage.dataset(name_lig)
l_folder_ref = listdir(pr_result)
nb_BS = 0
nb_BS_filtered = 0
nb_same_BS = 0
for PDB_ref in l_folder_ref :
if debug : print PDB_ref
if len (PDB_ref) != 4 :
continue
p_pdb_ref = pathManage.findPDBRef(pr_dataset + PDB_ref + "/")
l_p_query = pathManage.findPDBQueryTransloc (pathManage.result(name_lig) + PDB_ref + "/")
if debug : print l_p_query
for p_query in l_p_query :
# read TM Align
if debug : print p_query.split ("/")[-1][7:-4]
p_TMalign = pathManage.alignmentOutput(name_lig) + p_pdb_ref.split ("/")[-1][0:-4] + "__" + p_query.split ("/")[-1][7:-4] + "/RMSD"
try : score_align = parseTMalign.parseOutputTMalign(p_TMalign)
except :
filout_log.write ("ERROR TM align " + p_TMalign + "\n")
continue
nb_BS = nb_BS + 1
if score_align["IDseq"] >= ID_seq :
nb_BS_filtered = nb_BS_filtered + 1
l_p_substruct_ref = pathManage.findSubstructRef (pr_dataset + PDB_ref + "/", name_lig)
# sub BS
for p_substruct_ref in l_p_substruct_ref :
struct_substitued = p_substruct_ref.split ("_")[-2]
# write header
if not struct_substitued in d_file_BS.keys () :
d_file_BS[struct_substitued] = open (pr_out + name_lig + "_" + struct_substitued + "_", "w")
d_file_BS[struct_substitued].write ("name_bs\tRMSD_prot\tRMSD_BS_ca\tRMSD_BS_all\tD_max\tl_at_BS\tidentic\n")
RMSD_bs = analysis.computeRMSDBS (p_pdb_ref, p_query, p_substruct_ref, pr_out)
if RMSD_bs != [] :
d_file_BS[struct_substitued].write (p_substruct_ref.split("/")[-1][0:-4] + "_*_" + p_query.split ("/")[-1][0:-4] + "\t" + str(score_align["RMSD"]) + "\t" + str(RMSD_bs[1]) + "\t" + str(RMSD_bs[0]) + "\t" + str(RMSD_bs[2]) + "\t" + str(RMSD_bs[-2]) + "\t" + str(RMSD_bs[-1]) + "\n")
p_ligand_ref = pathManage.findligandRef(pr_dataset + PDB_ref + "/", name_lig)
RMSD_bs_lig = analysis.computeRMSDBS (p_pdb_ref, p_query, p_ligand_ref, pr_out)
if RMSD_bs_lig != [] :
d_file_BS["global"].write (p_ligand_ref.split("/")[-1][0:-4] + "_*_" + p_query.split ("/")[-1][0:-4] + "\t" + str(score_align["RMSD"]) + "\t" + str(RMSD_bs_lig[1]) + "\t" + str(RMSD_bs_lig[0]) + "\t" + str(RMSD_bs_lig[2]) + "\t" + str(RMSD_bs_lig[-2]) + "\t" + str(RMSD_bs_lig[-1]) + "\n")
if RMSD_bs_lig [-1] == 1 :
nb_same_BS = nb_same_BS + 1
# write summary
d_file_BS["summary"].write ("BS global: " + str (nb_BS) + "\n")
d_file_BS["summary"].write ("BS - IDseq " + str (ID_seq) + "%: " + str (nb_BS_filtered) + "\n")
d_file_BS["summary"].write ("BS - same atom number: " + str (nb_same_BS) + "\n")
filout_log.close ()
# close files and run histograms
for k_dico in d_file_BS.keys () :
p_file = d_file_BS[k_dico].name
d_file_BS[k_dico].close ()
if name_lig == "ATP" :
runOtherSoft.RhistogramRMSD(p_file, max_RMSD = 5.0)
elif name_lig == "ADP" :
runOtherSoft.RhistogramRMSD(p_file, max_RMSD = 4.0)
elif name_lig == "AMP" :
runOtherSoft.RhistogramRMSD(p_file, max_RMSD = 4.0)
else :
runOtherSoft.RhistogramRMSD(p_file, max_RMSD = 3.5)
return 1
def analysisBS(name_lig, ID_seq='0.0', debug=1):
pr_result = pathManage.result(name_lig)
pr_out = pathManage.result(name_lig + "/sameBS")
# log files
p_log_file = pr_out + "log.txt"
filout_log = open(p_log_file, "w")
# dictionnar with files
d_file_BS = {}
d_file_BS["global"] = open(pr_out + name_lig + "_", "w")
d_file_BS["global"].write(
"name_bs\tRMSD_prot\tRMSD_BS_ca\tRMSD_BS_all\tD_max\tl_at_BS\tidentic\n"
)
d_file_BS["summary"] = open(pr_out + "summary.txt", "w")
pr_dataset = pathManage.dataset(name_lig)
l_folder_ref = listdir(pr_result)
nb_BS = 0
nb_BS_filtered = 0
nb_same_BS = 0
for PDB_ref in l_folder_ref:
if debug: print PDB_ref
if len(PDB_ref) != 4:
continue
p_pdb_ref = pathManage.findPDBRef(pr_dataset + PDB_ref + "/")
l_p_query = pathManage.findPDBQueryTransloc(
pathManage.result(name_lig) + PDB_ref + "/")
if debug: print l_p_query
for p_query in l_p_query:
# read TM Align
if debug: print p_query.split("/")[-1][7:-4]
p_TMalign = pathManage.alignmentOutput(name_lig) + p_pdb_ref.split(
"/")[-1][0:-4] + "__" + p_query.split("/")[-1][7:-4] + "/RMSD"
try:
score_align = parseTMalign.parseOutputTMalign(p_TMalign)
except:
filout_log.write("ERROR TM align " + p_TMalign + "\n")
continue
nb_BS = nb_BS + 1
if score_align["IDseq"] >= ID_seq:
nb_BS_filtered = nb_BS_filtered + 1
l_p_substruct_ref = pathManage.findSubstructRef(
pr_dataset + PDB_ref + "/", name_lig)
# sub BS
for p_substruct_ref in l_p_substruct_ref:
struct_substitued = p_substruct_ref.split("_")[-2]
# write header
if not struct_substitued in d_file_BS.keys():
d_file_BS[struct_substitued] = open(
pr_out + name_lig + "_" + struct_substitued + "_",
"w")
d_file_BS[struct_substitued].write(
"name_bs\tRMSD_prot\tRMSD_BS_ca\tRMSD_BS_all\tD_max\tl_at_BS\tidentic\n"
)
RMSD_bs = analysis.computeRMSDBS(p_pdb_ref, p_query,
p_substruct_ref, pr_out)
if RMSD_bs != []:
d_file_BS[struct_substitued].write(
p_substruct_ref.split("/")[-1][0:-4] + "_*_" +
p_query.split("/")[-1][0:-4] + "\t" +
str(score_align["RMSD"]) + "\t" + str(RMSD_bs[1]) +
"\t" + str(RMSD_bs[0]) + "\t" + str(RMSD_bs[2]) +
"\t" + str(RMSD_bs[-2]) + "\t" + str(RMSD_bs[-1]) +
"\n")
p_ligand_ref = pathManage.findligandRef(
pr_dataset + PDB_ref + "/", name_lig)
RMSD_bs_lig = analysis.computeRMSDBS(p_pdb_ref, p_query,
p_ligand_ref, pr_out)
if RMSD_bs_lig != []:
d_file_BS["global"].write(
p_ligand_ref.split("/")[-1][0:-4] + "_*_" +
p_query.split("/")[-1][0:-4] + "\t" +
str(score_align["RMSD"]) + "\t" + str(RMSD_bs_lig[1]) +
"\t" + str(RMSD_bs_lig[0]) + "\t" +
str(RMSD_bs_lig[2]) + "\t" + str(RMSD_bs_lig[-2]) +
"\t" + str(RMSD_bs_lig[-1]) + "\n")
if RMSD_bs_lig[-1] == 1:
nb_same_BS = nb_same_BS + 1
# write summary
d_file_BS["summary"].write("BS global: " + str(nb_BS) + "\n")
d_file_BS["summary"].write("BS - IDseq " + str(ID_seq) + "%: " +
str(nb_BS_filtered) + "\n")
d_file_BS["summary"].write("BS - same atom number: " + str(nb_same_BS) +
"\n")
filout_log.close()
# close files and run histograms
for k_dico in d_file_BS.keys():
p_file = d_file_BS[k_dico].name
d_file_BS[k_dico].close()
if name_lig == "ATP":
runOtherSoft.RhistogramRMSD(p_file, max_RMSD=5.0)
elif name_lig == "ADP":
runOtherSoft.RhistogramRMSD(p_file, max_RMSD=4.0)
elif name_lig == "AMP":
runOtherSoft.RhistogramRMSD(p_file, max_RMSD=4.0)
else:
runOtherSoft.RhistogramRMSD(p_file, max_RMSD=3.5)
return 1