def GetTopoAlignStat(topo1, topo2):
stat = []
posTM1 = myfunc.GetTMPosition(topo1)
if len(posTM1) > 0:
for (b, e) in posTM1:
segList1 = []
segList2 = []
cntTM = 0
cntGap = 0
cntSeq = 0
for j in xrange(b, e):
if topo1[j] == 'M':
segList2.append(topo2[j])
if topo2[j] == 'M':
cntTM += 1
elif topo2[j] == '-':
cntGap += 1
else:
cntSeq += 1
rd = {}
sizeSeg = len(segList2)
freqTM = myfunc.FloatDivision(cntTM, sizeSeg)
freqGap = myfunc.FloatDivision(cntGap, sizeSeg)
freqSeq = myfunc.FloatDivision(cntSeq, sizeSeg)
rd['seg2'] = ''.join(segList2)
rd['freqTM'] = freqTM
rd['freqGap'] = freqGap
rd['freqSeq'] = freqSeq
stat.append(rd)
return stat
def MaskTopologyBySignalPeptide(idList, topoList, signalpDict):
newTopoList = []
for i in xrange(len(idList)):
topo = topoList[i]
if idList[i] in signalpDict:
posTMList = myfunc.GetTMPosition(topo)
try:
posSigP = signalpDict[idList[i]]
(b,e) = (posTMList[0][0],posTMList[0][1])
cov = myfunc.coverage(0, posSigP, b, e)
if float(cov)/(e-b) > 0.5:
#mask
masked_state = topo[e]
newTopo = ( "".join([masked_state]*(e)) +
topo[e:])
newTopoList.append(newTopo)
if DEBUG:
print
print "posTM", (b,e), "SignalPeptide", posSigP
print topo
print newTopo
else:
newTopoList.append(topo)
except (KeyError, IndexError):
newTopoList.append(topo)
else:
newTopoList.append(topo)
return newTopoList
def main(g_params): #{{{
argv = sys.argv
numArgv = len(argv)
if numArgv < 2:
PrintHelp()
return 1
topofile = ""
outfile = ""
isGapLess = False
i = 1
isNonOptionArg = False
while i < numArgv:
if isNonOptionArg == True:
topofile = argv[i]
isNonOptionArg = False
i += 1
elif argv[i] == "--":
isNonOptionArg = True
i += 1
elif argv[i][0] == "-":
if argv[i] in ["-h", "--help"]:
PrintHelp()
return 1
elif argv[i] in ["-o", "--o"]:
outfile = argv[i + 1]
i += 2
elif argv[i] in ["-i", "--i"]:
topofile = argv[i + 1]
i += 2
elif argv[i] in ["-gapless", "--gapless"]:
isGapLess = True
i += 1
elif argv[i] in ["-q"]:
g_params['isQuiet'] = True
i += 1
else:
print >> sys.stderr, "Error! Wrong argument:", argv[i]
return 1
else:
topofile = argv[i]
i += 1
if topofile == "":
print >> sys.stderr, "topofile not set. exit"
return 1
try:
(idList, annoList, seqList) = myfunc.ReadFasta(topofile)
fpout = myfunc.myopen(outfile, sys.stdout, "w", False)
for i in xrange(len(idList)):
topo = seqList[i]
seqid = idList[i]
if isGapLess:
topo = topo.replace("-", "").replace(".", "")
posTMList = myfunc.GetTMPosition(topo)
print >> fpout, seqid, posTMList
myfunc.myclose(fpout)
except (IOError, IndexError):
pass
def IsIdenticalTopology_simple( topo1, topo2, min_TM_overlap = 5):#{{{
"""Check whether topo1 and topo2 are identical"""
# Created 2011-11-15, updated 2011-11-15
# Two topologies are considered identical (Krogh et al. 2001) if
# 1. numTM1 == numTM2
# 2. Each helix of the compared topology should overlap by at least N (e.g. 5)
# residues
# 3. Each helix is oriented in the same way
numTM1 = myfunc.CountTM(topo1)
numTM2 = myfunc.CountTM(topo2)
Nterm1 = GetNtermState(topo1)
Nterm2 = GetNtermState(topo2)
posTM1 = myfunc.GetTMPosition(topo1)
posTM2 = myfunc.GetTMPosition(topo2)
if numTM1 != numTM2:
return False
else:
if Nterm1 != Nterm2:
return False
else:
for i in range (numTM1):
(b1,e1) = posTM1[i]
(b2,e2) = posTM2[i]
(common_b, common_e) = (max(b1,b2), min(e1,e2))
overlap = common_e - common_b
if overlap <= 0:
return False
else:
cntCommonM = 0
for j in range(common_b, common_e):
if topo1[j] == 'M' and topo2[j] == 'M':
cntCommonM += 1
if cntCommonM >= min_TM_overlap:
break
# print ("cntCommonM=", cntCommonM, "min_TM_overlap=",
# min_TM_overlap)
if cntCommonM < min_TM_overlap:
return False
return True
def MatchTopology(targetTopo, topoList, min_TM_overlap = 5, seqid = ""):#{{{
## compare targetTopo to all topologies in the topoList
# return (matchList, numIDTtopo, numPredictor
numList = len(topoList)
matchList = []
# 0 for different topology
# 1 for identical topology
# -1 for empty topology
# debug
# print "SeqID: %s"%(seqid)
# print GetNtermState(targetTopo), myfunc.GetTMPosition(targetTopo)
# print
# for tt in topoList:
# if tt:
# print GetNtermState(tt), myfunc.GetTMPosition(tt)
# else:
# print "Null"
# print
NtermStateTarget = GetNtermState(targetTopo)
posTMtarget = myfunc.GetTMPosition(targetTopo)
numTMtarget = len(posTMtarget)
for i in range(numList):
if topoList[i] == "":
matchList.append(-1)
else:
NtermState = GetNtermState(topoList[i])
posTM = myfunc.GetTMPosition(topoList[i])
numTM = len(posTM)
if IsIdenticalTopology(NtermStateTarget, NtermState,
numTMtarget, numTM, posTMtarget, posTM, targetTopo,
topoList[i], min_TM_overlap):
matchList.append(1)
else:
matchList.append(0)
numIDTtopo = matchList.count(1)
numPredictor = matchList.count(1) + matchList.count(0)
return (matchList, numIDTtopo, numPredictor)
def CalKRBias(seq, topo, flank_win, max_dist):
kr_bias = None
KR_pos_list = [] # [[1,5], [1,3]]
posTM = myfunc.GetTMPosition(topo)
NtermState = lcmp.GetNtermState(topo)
numTM = len(posTM)
length = len(seq)
if numTM >= 1:
sum_KR_odd = 0
sum_KR_even = 0
for i in range(numTM + 1):
if i == 0 or i == numTM:
if i == 0:
b = max(0, posTM[i][0] - max_dist)
e = posTM[i][0] + flank_win
else:
b = posTM[i - 1][1] - flank_win
e = min(length, posTM[i - 1][1] + max_dist)
KRpos = GetKRPos(seq, b, e)
else:
if posTM[i][0] - posTM[i - 1][1] > 2 * max_dist:
b1 = posTM[i - 1][1] - flank_win
e1 = posTM[i - 1][1] + max_dist
b2 = posTM[i][0] - max_dist
e2 = posTM[i][0] + flank_win
KRpos = GetKRPos(seq, b1, e1)
KRpos += GetKRPos(seq, b2, e2)
else:
b = posTM[i - 1][1] - flank_win
e = posTM[i][0] + flank_win
# print (b,e)
# print len(seq)
# print "flank_win=",flank_win
# print "i=",i
# print "numTM=", len(posTM)
KRpos = GetKRPos(seq, b, e)
KR_pos_list.append(KRpos)
if i % 2 == 0:
sum_KR_odd += len(KRpos)
else:
sum_KR_even += len(KRpos)
kr_bias = sum_KR_odd - sum_KR_even
# print KR_pos_list
return (kr_bias, KR_pos_list, numTM)
def IsAllIdenticalTopology(topoList): #{{{
numSeq = len(topoList)
if numSeq <= 1:
return True
else:
posTMList = [myfunc.GetTMPosition(topo) for topo in topoList]
NtermStateList = [lcmp.GetNtermState(topo) for topo in topoList]
numTMList = [len(posTM) for posTM in posTMList]
for i in xrange(numSeq - 1):
for j in xrange(i + 1, numSeq):
if not lcmp.IsIdenticalTopology(
NtermStateList[i], NtermStateList[j], numTMList[i],
numTMList[j], posTMList[i], posTMList[j], topoList[i],
topoList[j], min_TM_overlap):
return False
return True
def RunHHSearchPairwise(tableinfoFile, #{{{
hhprofilepathList, hhprofilepathMapDictList,
hhsearchpathList, hhsearchpathMapDictList,
topoDict, outpath, dupfile):
fpoutDup = None
if dupfile != "":
fpoutDup = myfunc.myopen(dupfile, sys.stdout, "w", False)
hdl = myfunc.ReadLineByBlock(tableinfoFile)
if hdl.failure:
return 1
cnt = 0
lines = hdl.readlines()
while lines != None:
for line in lines:
if not line or line[0] == "#":
continue
strs = line.split()
try:
seqid1 = strs[0]
seqid2 = strs[1]
except (IndexError, ValueError):
print >> sys.stderr, "Bad record line \"%s\""%(line)
continue
try:
topo1 = topoDict[seqid1]
except KeyError:
topo1 = ""
try:
topo2 = topoDict[seqid2]
except KeyError:
topo2 = ""
seqlen1 = len(topo1)
seqlen2 = len(topo2)
pairlist = [(seqid1, seqlen1), (seqid2, seqlen2)]
pairlist = sorted(pairlist, key=lambda x:x[1], reverse=False) # short - long
hhrfile = "%s%s%s_%s.hhr"%(outpath, os.sep, seqid1, seqid2)
if g_params['isUsePreBuildHHSearchResult']:
keystr = "%s_%s"%(seqid1, seqid2)
tmp_hhrfile = GetProfileFileName(hhsearchpathList,
hhsearchpathMapDictList, keystr, ".hhr")
if os.path.exists(tmp_hhrfile):
hhrfile = tmp_hhrfile
else:
print >> sys.stderr, "hhrfile %s does not exist in"\
" the prebuilt path"%(hhrfile)
# update seqid1 and seqid2 (shorter - longer)
seqid1 = pairlist[0][0] # shorter sequence
seqid2 = pairlist[1][0] # longer sequence
try:
topo1 = topoDict[seqid1]
except KeyError:
topo1 = ""
try:
topo2 = topoDict[seqid2]
except KeyError:
topo2 = ""
seqlen1 = len(topo1)
seqlen2 = len(topo2)
numTM1 = len(myfunc.GetTMPosition(topo1))
numTM2 = len(myfunc.GetTMPosition(topo2))
if not os.path.exists(hhrfile) or g_params['isForceOverWrite']:
a3mfile = GetProfileFileName(hhprofilepathList, #query
hhprofilepathMapDictList, pairlist[0][0], ".a3m")
hhmfile = GetProfileFileName(hhprofilepathList, #template
hhprofilepathMapDictList, pairlist[1][0], ".hhm")
if a3mfile == "" or not os.path.exists(a3mfile):
print >> sys.stderr, "a3mfile not found for %s. Ignore." %(pairlist[0][0])
elif hhmfile == "" or not os.path.exists(hhmfile):
print >> sys.stderr, "hhmfile not found for %s. Ignore." %(pairlist[1][0])
else:
tmp_hhrfile = "%s.tmp"%(hhrfile)
cmd = "hhsearch -i %s -d %s -o %s -v 0 -nocons -nopred -nodssp" % (
a3mfile, hhmfile, tmp_hhrfile)
os.system(cmd)
if os.path.exists(tmp_hhrfile):
os.system("/bin/mv -f %s %s"%(tmp_hhrfile, hhrfile))
print hhrfile, "output"
if fpoutDup and os.path.exists(hhrfile):
ss_isdup = ""
hitinfo = {}
# if IsDuplicatedByHHSearch(hhrfile, seqid1, seqid2, cnt):
# ss_isdup = 'y'
# else:
# ss_isdup = 'n'
hitinfo = CheckDuplication(hhrfile, seqid1, seqid2, topoDict, cnt)
if hitinfo != {}:
fpoutDup.write("%s-%s %s %4d %4d %4d %4d" %(
seqid1, seqid2, hitinfo['isDup'],
len(topo1), len(topo2), numTM1, numTM2))
if 'hit' in hitinfo:
for j in xrange(len(hitinfo['hit'])):
hit = hitinfo['hit'][j]
ss_hit = "%d-%d(nTM=%d) %d-%d(nTM=%d)"%(
hit['posQuery'][0], hit['posQuery'][1], hit['numTM1'],
hit['posTemplate'][0], hit['posTemplate'][1], hit['numTM2'])
fpoutDup.write(" | %35s"%(ss_hit))
fpoutDup.write("\n")
cnt += 1
lines = hdl.readlines()
hdl.close()
myfunc.myclose(fpoutDup)
return 0
def CheckDuplication(hhrfile, seqid1, seqid2, topoDict, cnt):#{{{
hitinfo = {}
try:
# Read in hhsearch hits
fpin = open(hhrfile,"r")
lines = fpin.readlines()
fpin.close()
except IOError:
print >> sys.stderr, "Failed to read hhrfile %s"%hhrfile
return {}
try:
topo1 = topoDict[seqid1]
except KeyError:
topo1 = ""
try:
topo2 = topoDict[seqid2]
except KeyError:
topo2 = ""
lengthQuery = 0
lengthTemplate = 0
hitList = []
numLine = len(lines)
i = 0
while i < numLine:
line = lines[i]
if line.find("Match_columns") == 0:
try:
lengthQuery = int(line.split()[1])
hitinfo['seqLen1'] = lengthQuery
except (IndexError, ValueError):
print >> sys.stderr, "Error in hhrfile %s. Ignore"%(hhrfile)
return {}
i += 1
elif line.find(" No Hit") == 0:
j = 1
while i+j < numLine and lines[i+j] != "":
hit = ExtractHit(lines[i+j])
if hit != {} and hit['evalue'] <= EVALUE_THRESHOLD:
posQuery = hit['posQuery']
posTemplate = hit['posTemplate']
if topo1 != "":
hit['numTM1'] = len(myfunc.GetTMPosition(topo1[posQuery[0]:posQuery[1]]))
else:
hit['numTM1'] = 0
if topo2 != "":
hit['numTM2'] = len(myfunc.GetTMPosition(topo2[posTemplate[0]:posTemplate[1]]))
else:
hit['numTM2'] = 0
hitList.append(hit)
else:
break
j += 1
break
else:
i += 1
hitList = sorted(hitList, key=lambda x:x['evalue'], reverse=False)
hitinfo['hit'] = hitList
numHit = len(hitList)
if numHit < 2: # there should be at least two hits
isDup = False
else:
isDup = IsDuplicated(hitList, len(topo1), len(topo2))
if isDup:
ss_isdup = 'y'
hitinfo['isDup'] = 'y'
else:
ss_isdup = 'n'
hitinfo['isDup'] = 'n'
sys.stdout.write("%d: %s-%s %s numHit=%d\n" %(cnt, seqid1,
seqid2, ss_isdup, numHit))
return hitinfo
def main(g_params):#{{{
argv = sys.argv
numArgv = len(argv)
if numArgv < 2:
PrintHelp()
return 1
outfile = ""
seqdbfile = ""
infile = ""
i = 1
isNonOptionArg=False
while i < numArgv:
if isNonOptionArg == True:
infile = argv[i]
isNonOptionArg = False
i += 1
elif argv[i] == "--":
isNonOptionArg = True
i += 1
elif argv[i][0] == "-":
if argv[i] in ["-h", "--help"]:
PrintHelp()
return 1
elif argv[i] in ["-outfile", "--outfile"]:
outfile = argv[i+1]
i += 2
elif argv[i] in ["-seqdb", "--seqdb"]:
seqdbfile = argv[i+1]
i += 2
elif argv[i] in ["-q"]:
g_params['isQuiet'] = True
i += 1
else:
print >> sys.stderr, "Error! Wrong argument:", argv[i]
return 1
else:
infile = argv[i]
i += 1
if infile == "":
print >> sys.stderr, "topofile not set"
return 1
elif not os.path.exists(infile):
print >> sys.stderr, "topofile %s does not exist"%(infile)
return 1
# if seqdbfile == "":
# print >> sys.stderr, "seqdbfile file not set"
# return 1
# elif not os.path.exists(seqdbfile):
# print >> sys.stderr, "seqdbfile file %s does not exist"%(seqdbfile)
# return 1
# seqDict = GetSeqDict(seqdbfile)
# if seqDict == {}:
# print >> sys.stderr, "Failed to read seqdbfile %s"%(seqdbfile)
# return 1
fpout = myfunc.myopen(outfile, sys.stdout, "w", False)
fpin = open (infile, "rb");
if not fpin:
print >> sys.stderr, "Failed to open input file %s"%(infile)
return 1
unprocessedBuffer="";
isEOFreached = False;
processedTopoIDSet = set([]);
while 1:
buff = fpin.read(BLOCK_SIZE);
if len(buff) < BLOCK_SIZE:
isEOFreached=True;
buff = unprocessedBuffer + buff;
recordList = [];
unprocessedBuffer = myfunc.ReadFastaFromBuffer(buff,recordList, isEOFreached);
if len(recordList) > 0:
idListTopo = [r[0] for r in recordList];
topoList = [r[2] for r in recordList];
for i in xrange(len(idListTopo)):
seqid = idListTopo[i]
topo = topoList[i]
posTM = myfunc.GetTMPosition(topo)
if len(posTM) > 0:
cnt = 0
for (b,e) in posTM:
seg = topo[b:e]
fpout.write("%s\t%4d\t%s\n"%(seqid, cnt+1, seg))
cnt += 1
if isEOFreached == True:
break;
fpin.close();
myfunc.myclose(fpout)
def CountIdenticalTopology(pred_topodict, real_topodict, agreement, TM_type,
fpout_wrong, seqDict, pred_method_item): #{{{
"""
return (cntIDT, cntINV)
"""
numPredTopo = len(pred_topodict)
cntIDT = 0
cntINV = 0
cntDIFF = 0
for seqid in pred_topodict:
predtopo = pred_topodict[seqid]
try:
realtopo = real_topodict[seqid]
except KeyError:
print >> sys.stderr, "%s a nonTM protein predicted as TM protein" % (
seqid)
realtopo = "i" * len(predtopo)
pass
pred_NtermState = lcmp.GetNtermState(predtopo)
real_NtermState = lcmp.GetNtermState(realtopo.replace('.', '-'))
pred_posTM = myfunc.GetTMPosition(predtopo)
real_posTM = myfunc.GetTMPosition(realtopo)
pred_numTM = len(pred_posTM)
real_numTM = len(real_posTM)
# if g_params['isDEBUG'] and seqid == "3cx5I":
# print "pred_NtermState = <%s>"%pred_NtermState
# print "real_NtermState = <%s>"% real_NtermState
# print "pred_posTM = ", pred_posTM
# print "real_posTM = ", real_posTM
if lcmp.IsIdenticalTopology(pred_NtermState, real_NtermState,
pred_numTM, real_numTM, pred_posTM,
real_posTM, predtopo, realtopo,
g_params['min_TM_overlap']):
cntIDT += 1
else:
if fpout_wrong != None:
# output the wrongly predict topology
fpout_wrong.write(">%s Number %d mtd_%s\n" %
(seqid, cntDIFF + 1, pred_method_item))
try:
seq = seqDict[seqid]
fpout_wrong.write("%-10s %s\n" % ("AASeq", seq))
except KeyError:
seq = ""
fpout_wrong.write("%-10s %s\n" % ("RealTop", realtopo))
fpout_wrong.write("%-10s %s\n" % ("PredTop", predtopo))
fpout_wrong.write("\n")
if lcmp.IsInvertedTopology(pred_NtermState, real_NtermState,
pred_numTM, real_numTM, pred_posTM,
real_posTM, predtopo, realtopo,
g_params['min_TM_overlap']):
cntINV += 1
if g_params['isDEBUG']:
print >> sys.stderr, "%-7s(real %3s) nTM=%2d %s" % (
seqid, agreement, real_numTM, realtopo)
print >> sys.stderr, "%-7s(pred %3s) nTM=%2d %s" % (
seqid, agreement, pred_numTM, predtopo)
print >> sys.stderr
cntDIFF += 1
return (cntIDT, cntINV)
def WriteInfo(groupList, seqlenDict, seqannoDict, pfamidDefDict,#{{{
clanidDefDict, topoDict, groupedPfamScanDict, htmlname, fpout):
cnt = 0
for tup in groupList:
try:
ss = tup[0]
seqidlist = tup[2]
famidlist = ss.split("\t")
fpout.write("Group %d: %d seqs, %d domains "%(cnt+1, tup[1],
len(famidlist)))
for famid in famidlist:
if famid[0] == 'P':
famdef = pfamidDefDict[famid]
else:
famdef = clanidDefDict[famid]
fpout.write(" %s (%s)"%(famid, famdef))
fpout.write("\n")
fpout.write("#%-3s %10s %4s %3s %15s %5s\n"%("No",
"SeqID", "Len", "nTM", "DomainCoverage", "nTM_within"))
cntseq = 0
for seqid in seqidlist:
try:
seqlen = seqlenDict[seqid]
except KeyError:
seqlen = -1
pass
fpout.write("%-4d %10s %4d"%(cntseq+1, seqid, seqlen))
try:
topo = topoDict[seqid]
posTM = myfunc.GetTMPosition(topo)
except KeyError:
print >> sys.stderr, "topo not found for %s"%seqid
fpout.write("\n")
continue
fpout.write(" %3d"%(len(posTM)))
pfamscan_hit = groupedPfamScanDict[seqid]
for famid in famidlist:
try:
b1 = pfamscan_hit[famid]['alnBeg']
e1 = pfamscan_hit[famid]['alnEnd']
(posTM_covered, indexList_covered) = GetCoveredTM((b1,e1), posTM)
fpout.write("%15s %5s %4s"%("(%d,%d)"%(b1,e1),
"%d TM"%(len(posTM_covered)),
"%d-%d"%(indexList_covered[0]+1,
indexList_covered[len(indexList_covered)-1]+1)
))
except (KeyError):
print >> sys.stderr, "%s not in pfamscan_hit"%(famid)
pass
fpout.write("\n")
cntseq += 1
except (KeyError, IndexError):
print >> sys.stderr, "Error for %s"%(tup[0])
pass
cnt += 1
fpout.write("\n")
def AnaLocalTopoAln(idList, topoList, localseqpairDict, fpout, fpout1):
# fpout: write result for those one end with no TM region
# fpout1: write result for those one end with one TM region but less then 5
# residues of TM region
numseq = len(idList)
numpair = numseq / 2
for i in xrange(numpair):
id1 = idList[2 * i]
id2 = idList[2 * i + 1]
topo1 = topoList[2 * i]
topo2 = topoList[2 * i + 1]
lengthAln = len(topo1)
try:
unaligned_str = localseqpairDict[(id1, id2)][2]
except (KeyError, IndexError):
print >> sys.stderr, "no local alignment found for %s %s" % (id1,
id2)
continue
alignedPosList = myfunc.GetSegPos(unaligned_str, "1")
if len(alignedPosList) != 1:
print >> sys.stderr, "aligned region not equal 1 for %s %s" % (id1,
id2)
continue
else:
alignedPos = alignedPosList[0]
if alignedPos[0] == 0 and alignedPos[1] == lengthAln:
print "%s %s: Full aligned" % (id1, id2)
else:
alignedPos = alignedPosList[0]
topo_Nterm1 = topo1[:alignedPos[0]]
topo_Cterm1 = topo1[alignedPos[1]:]
topo_Nterm2 = topo2[:alignedPos[0]]
topo_Cterm2 = topo2[alignedPos[1]:]
posTM_Nterm1 = myfunc.GetTMPosition(topo_Nterm1)
posTM_Cterm1 = myfunc.GetTMPosition(topo_Cterm1)
posTM_Nterm2 = myfunc.GetTMPosition(topo_Nterm2)
posTM_Cterm2 = myfunc.GetTMPosition(topo_Cterm2)
s_Nterm = GetUnglianedTermStatus(posTM_Nterm1, posTM_Nterm2)
s_Cterm = GetUnglianedTermStatus(posTM_Cterm1, posTM_Cterm2)
# s1_Nterm, s2_Nterm is used to record the status of those
# unaligned terminals with one has a splitted TM helices and
# another has >= 1 TM helix
s1_Nterm = GetUnglianedTermStatus1(posTM_Nterm1, posTM_Nterm2)
s1_Cterm = GetUnglianedTermStatus1(posTM_Cterm1, posTM_Cterm2)
if s_Nterm < 2 and s_Cterm < 2 and (s_Nterm + s_Cterm) > 0:
if len(posTM_Nterm1) > 0:
num_res_unaligned_Nterm = len(topo_Nterm2.replace("-", ""))
numTM_unaligned_Nterm = len(posTM_Nterm1)
num_res_to_TM_Nterm = len(topo_Nterm1) - posTM_Nterm1[
len(posTM_Nterm1) - 1][1]
elif len(posTM_Nterm2) > 0:
num_res_unaligned_Nterm = len(topo_Nterm1.replace("-", ""))
numTM_unaligned_Nterm = len(posTM_Nterm2)
num_res_to_TM_Nterm = len(topo_Nterm2) - posTM_Nterm2[
len(posTM_Nterm2) - 1][1]
else:
num_res_unaligned_Nterm = 0
numTM_unaligned_Nterm = 0
num_res_to_TM_Nterm = 0
if len(posTM_Cterm1) > 0:
num_res_unaligned_Cterm = len(topo_Cterm2.replace("-", ""))
numTM_unaligned_Cterm = len(posTM_Cterm1)
num_res_to_TM_Cterm = posTM_Cterm1[0][0]
elif len(posTM_Cterm2) > 0:
num_res_unaligned_Cterm = len(topo_Cterm1.replace("-", ""))
numTM_unaligned_Cterm = len(posTM_Cterm2)
num_res_to_TM_Cterm = posTM_Cterm2[0][0]
else:
num_res_unaligned_Cterm = 0
numTM_unaligned_Cterm = 0
num_res_to_TM_Cterm = 0
ss = "%s %s %4d %4d %4d %4d %4d %4d"
print >> fpout, ss % (
id1, id2, num_res_unaligned_Nterm, num_res_to_TM_Nterm,
numTM_unaligned_Nterm, num_res_unaligned_Cterm,
num_res_to_TM_Cterm, numTM_unaligned_Cterm)
if ((s1_Nterm == 1.5 or s1_Cterm == 1.5) and s1_Nterm < 2
and s1_Cterm < 2):
num_res_unaligned_Nterm = -1
numRes_PartHelix_Nterm = -1
numTM_unaligned_Nterm = -1
num_res_to_TM_Nterm = -1
num_res_unaligned_Cterm = -1
numRes_PartHelix_Cterm = -1
numTM_unaligned_Cterm = -1
num_res_to_TM_Cterm = -1
if s1_Nterm == 1.5:
if len(
posTM_Nterm1
) == 1 and posTM_Nterm1[0][1] - posTM_Nterm1[0][0] < 5:
num_res_unaligned_Nterm = len(
topo_Nterm1.replace("-", ""))
numRes_PartHelix_Nterm = posTM_Nterm1[0][
1] - posTM_Nterm1[0][0]
numTM_unaligned_Nterm = len(posTM_Nterm2)
num_res_to_TM_Nterm = len(topo_Nterm2) - posTM_Nterm2[
len(posTM_Nterm2) - 1][1]
elif len(
posTM_Nterm2
) == 1 and posTM_Nterm2[0][1] - posTM_Nterm2[0][0] < 5:
num_res_unaligned_Nterm = len(
topo_Nterm2.replace("-", ""))
numRes_PartHelix_Nterm = posTM_Nterm2[0][
1] - posTM_Nterm2[0][0]
numTM_unaligned_Nterm = len(posTM_Nterm1)
num_res_to_TM_Nterm = len(topo_Nterm1) - posTM_Nterm1[
len(posTM_Nterm1) - 1][1]
if s1_Cterm == 1.5:
if len(
posTM_Cterm1
) == 1 and posTM_Cterm1[0][1] - posTM_Cterm1[0][0] < 5:
num_res_unaligned_Cterm = len(
topo_Cterm1.replace("-", ""))
numRes_PartHelix_Cterm = posTM_Cterm1[0][
1] - posTM_Cterm1[0][0]
numTM_unaligned_Cterm = len(posTM_Cterm2)
num_res_to_TM_Cterm = len(topo_Cterm2) - posTM_Cterm2[
len(posTM_Cterm2) - 1][1]
elif len(
posTM_Cterm2
) == 1 and posTM_Cterm2[0][1] - posTM_Cterm2[0][0] < 5:
num_res_unaligned_Cterm = len(
topo_Cterm2.replace("-", ""))
numRes_PartHelix_Cterm = posTM_Cterm2[0][
1] - posTM_Cterm2[0][0]
numTM_unaligned_Cterm = len(posTM_Cterm1)
num_res_to_TM_Cterm = len(topo_Cterm1) - posTM_Cterm1[
len(posTM_Cterm1) - 1][1]
ss = "%s %s %4d %4d %4d %4d %4d %4d %4d %4d"
print >> fpout1, ss % (
id1, id2, num_res_unaligned_Nterm, numRes_PartHelix_Nterm,
num_res_to_TM_Nterm, numTM_unaligned_Nterm,
num_res_unaligned_Cterm, numRes_PartHelix_Cterm,
num_res_to_TM_Cterm, numTM_unaligned_Cterm)