def writeSPDB(self):
aamap = AAmap()
fo = open(self.pdbfile + '.spdb', 'w')
for a in self.atoms:
fo.write('%f %f %f %d %s\n' %
(a.x, a.y, a.z, a.resSeq, aamap.getAAmap(a.resName)))
fo.close()
def getSeq(self):
aamap = AAmap()
seq=''
last_resSeq = -1
seqPos = 0
for i in xrange(0,len(self.atoms)):
a=self.atoms[i]
if last_resSeq != a.resSeq:
seq=seq+aamap.getAAmap(a.resName)
last_resSeq = a.resSeq
key = '%s%s' % (a.chainID, a.resSeq)
self.resDict[key] = (seqPos, seq[seqPos])
seqPos+=1
return seq
def filterClusters(self):
if len(self.pairwiseDict) == 0:
self.pairwise()
amap = AAmap()
for i in xrange(0, len(self.atoms)):
c = cluster(self.pdb, self.top, self.pfam, '', '', self.seqheader,
'', '', self.center, self.cutoff, self.scutoff,
self.flag, 1.0, self.desc)
c.addNeighbor(amap, self.atoms[i], i) # put itself in first
nbnum = 0
for j in xrange(0, len(self.atoms)):
key = "%d-%d" % (i, j)
if (self.pairwiseDict[key] <= self.cutoff) and (abs(i - j) >=
self.scutoff):
c.addNeighbor(amap, self.atoms[j], j)
nbnum = nbnum + 1
c.thetaPhi.append(
self.calculateThetaPhi(self.atoms[i], self.atoms[j]))
if nbnum < self.nbcutoff:
continue
c.pdbidx = c.pdbidx.lstrip() # will change meanDist
c.pdbResSeq = c.pdbResSeq.lstrip()
meanDist = self.clusterMeanDist(c)
if meanDist < 5.8:
print('%s,%0.2f,%s,%s,%s,%s') % (self.pdb, meanDist, ''.join(
sorted(c.str)), ''.join(sorted(
c.typeStr)), c.pdbResSeq, self.getSphericalStr(c))
self.clusters.append(c)
def writeFASTA(self):
fafile = self.pdb+'.fa'
aamap = AAmap()
seq=''
count = 0
last_resSeq = -1
for i in xrange(0,len(self.atoms)):
a=self.atoms[i]
if last_resSeq != a.resSeq:
seq=seq+aamap.getAAmap(a.resName)
last_resSeq = a.resSeq
count+=1
seq=seq+'\n'
header = '>%s/1-%d\n' % (self.pdb, count)
print header+seq
fp=open(fafile, 'w')
fp.write(header+seq)
fp.close()
def writeFASTA(self):
fafile = self.pdb + '.fa'
aamap = AAmap()
seq = ''
count = 0
last_resSeq = -1
for i in xrange(0, len(self.atoms)):
a = self.atoms[i]
if last_resSeq != a.resSeq:
seq = seq + aamap.getAAmap(a.resName)
last_resSeq = a.resSeq
count += 1
seq = seq + '\n'
header = '>%s/1-%d\n' % (self.pdb, count)
print header + seq
fp = open(fafile, 'w')
fp.write(header + seq)
fp.close()
def __init__(self, nafile):
self.pdb = nafile[0:4]
self.rsaDict = {}
self.resiDict = defaultdict(lambda: '')
self.alphabet = ['B', 'E']
aamap = AAmap()
lines = [line.strip() for line in open(nafile)]
for naline in lines:
head = naline[0:3]
if head == 'RES':
r = rsa(naline)
key = '%s%s%s' % (aamap.getAAmap(r.resn), r.chain, r.resi)
self.rsaDict[key] = r
varkey = '%s%s' % (aamap.getAAmap(r.resn), self.accessible(key))
varvalue = '%s%s%s ' % (self.resiDict[varkey], r.chain, r.resi)
self.resiDict[varkey] = varvalue
elif head == 'TOTAL':
key = 'TOTAL'
self.rasDict[key] = naline.split()
def resn2bfactor():
if len(sys.argv) < 3:
print 'resn2bfactor(): replace b factor values with residue type.'
print 'resn2bfactor(): used for pymol spectrum b'
return
scoreValue = {
'X':0,'-': 0,'.': 0,'A': 1,'C': 2,'D': 3,'E': 4,'F': 5,'G': 6,'H': 7,'I': 8,'K': 9,
'L': 10,'M': 11,'N': 12,'P': 13,'Q': 14,'R': 15,'S': 16,'T': 17,'V': 18,'W': 19,'Y': 20, 'B': 3
}
aamap = AAmap()
pdbfile = sys.argv[2]
p = protein(pdbfile)
outfile = '%s_rb.pdb' % pdbfile[:-4]
fout = open(outfile, 'w')
for a in p.atoms:
newBFactor = scoreValue[aamap.getAAmap(a.resName)]
print 'new b-factor: [%s : %s] -> %d' % (a.resName, aamap.getAAmap(a.resName), newBFactor)
a.tempFactor = newBFactor
fout.write(a.writeAtom())
fout.close()
print 'Output file: %s' % outfile
def getSeq(self):
aamap = AAmap()
seq = ''
#last_resSeq = -1 # 1a8v the first resi starts from -1 !!!!
last_resSeq = -9999 # 1a8v the first resi starts from -1 !!!!
seqPos = 0
resArray = []
resAtomsAll = []
resatoms = []
for i in xrange(0, len(self.atoms)):
a = self.atoms[i]
if last_resSeq != a.resSeq:
seq = seq + aamap.getAAmap(a.resName)
last_resSeq = a.resSeq
key = '%s%d' % (a.chainID, a.resSeq)
self.resDict[key] = (seqPos, seq[seqPos])
seqPos += 1
#resArray.append('%s %s %s' % (a.chainID,aamap.getAAmap(a.resName),str(a.resSeq)))
resArray.append(
(a.chainID, aamap.getAAmap(a.resName), a.resSeq))
if len(resatoms) > 0:
resAtomsAll.append(resatoms)
resatoms = []
resatoms.append(a)
# after loop add the last res into resatoms
# only resSeq change trigger adding above
if len(resatoms) > 0:
resAtomsAll.append(resatoms)
return seq, resArray, resAtomsAll
def main():
if len(sys.argv) < 3:
print 'python proc_dendrogram.py preffix cutoff'
exit
preffix = sys.argv[1]
cutoff = float(sys.argv[2])
# load tip pdb file
pr = protein(preffix)
aamap = AAmap()
n = len(pr.atoms)
resimap = {}
print 'writing %s.resimap ...' % (preffix)
fr = open(preffix+'.resimap', 'w')
px = []
count = 0
for a in pr.atoms:
px.append((a.x, a.y, a.z))
resimap[count] = ('%s%d' % (a.chainID, a.resSeq), aamap.getAAmap(a.resName))
fr.write('%d %s%d %s\n' % (count, a.chainID, a.resSeq, aamap.getAAmap(a.resName)))
count+=1
fr.close()
x = np.array(px)
# calculate pairwised distance
pdist = {}
print 'writing %s.pdist ...' % (preffix)
fo=open(preffix+'.pdist','w')
for i in xrange(0,len(x)):
for j in xrange(i+1,len(x)):
dist = np.linalg.norm(x[i]-x[j])
pdist['%d-%d' % (i,j)] = dist
fo.write('%d-%d : %f\n' % (i,j,dist))
fo.close()
# for hc extraction
hcdict = {}
hclist = []
existdict = {}
#linkage_matrix = linkage(x, "single")
linkage_matrix = linkage(x, "complete")
#ddata = augmented_dendrogram(linkage_matrix, color_threshold=1)
#plt.show()
print 'writing %s.hcluster ...' % (preffix)
fo1 = open(preffix+'.hcluster', 'w')
m = linkage_matrix
for i in xrange(0,len(m)):
#print '%d %d %d %f %d' % (n+i,m[i,0],m[i,1],m[i,2],m[i,3])
fo1.write('%d %d %d %f %d\n' % (n+i,m[i,0],m[i,1],m[i,2],m[i,3]))
hcline = '%d %d %d %f %d' % (n+i,m[i,0],m[i,1],m[i,2],m[i,3])
h = hc(hcline, n)
hcdict[h.clusterID] = h
hclist.append(h)
fo1.close()
# resolve leaves for each cluster
print 'resolving leaves ...'
for h in hclist:
h.getChildren(hcdict)
#h.dump()
print 'iterating clusters for largest proximity contact ...'
for i in xrange(0, n):
leafstr = '%d %d %d 0.0 1' % (i, i, i)
h = hc(leafstr, n)
h.leaves = [i]
hcdict[i] = h
#hcdict[i].dump()
# add single leaf in
for i in xrange(0, n):
existdict[i]= True
for h in hclist:
if h.dist <= cutoff:
if h.c1 in existdict and h.c2 in existdict: # both been checked before
#print '1AA'
if existdict[h.c1] == True and existdict[h.c2] == True:
ret = checkProximity2(hcdict[h.c1], hcdict[h.c2], pdist, cutoff)
existdict[h.clusterID] = ret
if ret == True: # combine both and delete sub cluster in the dict
existdict[h.c1] = False
existdict[h.c2] = False
elif existdict[h.c1] == False or existdict[h.c2] == False:
existdict[h.clusterID] = False
elif h.c1 in existdict and h.c2 not in existdict:
#print '1AB'
if existdict[h.c1] == False: # c1 is not a contact; get h
existdict[h.clusterID] = False
existdict[h.c2] = checkProximity(hcdict[h.c2], pdist, cutoff) # get c2
elif existdict[h.c1] == True: # c1 is a contact; get c2 then get h = c1 and c2
ret = checkProximity(hcdict[h.c2], pdist, cutoff) # get c2
existdict[h.c2] = ret
if ret == False:
existdict[h.clusterID] = False
elif ret == True: # h.c2 is a contact
ret1 = checkProximity2(hcdict[h.c1], hcdict[h.c2], pdist, cutoff)
existdict[h.clusterID] = ret1
if ret1 == True:
existdict[h.c1] = False
existdict[h.c2] = False
elif h.c1 not in existdict and h.c2 in existdict:
#print '1BA'
if existdict[h.c2] == False: # c2 is not a contact; get h
existdict[h.clusterID] = False
existdict[h.c1] = checkProximity(hcdict[h.c1], pdist, cutoff) # get c1
elif existdict[h.c2] == True: # c2 is a contact; get c1 then get h = c1 and c2
ret = checkProximity(hcdict[h.c1], pdist, cutoff) # get c1
existdict[h.c1] = ret
if ret == False:
existdict[h.clusterID] = False
elif ret == True: # h.c1 is a contact
ret1 = checkProximity2(hcdict[h.c1], hcdict[h.c2], pdist, cutoff)
existdict[h.clusterID] = ret1
if ret1 == True:
existdict[h.c1] = False
existdict[h.c2] = False
elif h.c1 not in existdict and h.c2 not in existdict:
#print '1BB'
r1 = checkProximity(hcdict[h.c1], pdist, cutoff)
existdict[h.c1] = r1
r2 = checkProximity(hcdict[h.c2], pdist, cutoff)
existdict[h.c2] = r2
if r1 == False or r2 == False:
existdict[h.clusterID] = False
elif r1 == True and r2 == True:
ret = checkProximity2(hcdict[h.c1], hcdict[h.c2], pdist, cutoff)
if ret == True:
existdict[h.c1] = False
existdict[h.c2] = False
elif h.dist > cutoff:
#print '0XX'
existdict[h.clusterID] = False
if h.c1 not in existdict:
existdict[h.c1] = checkProximity(hcdict[h.c1], pdist, cutoff)
if h.c2 not in existdict:
existdict[h.c2] = checkProximity(hcdict[h.c2], pdist, cutoff)
# print out the result
print 'writing result into %s.hcg' % preffix
fout = open(preffix+'.hcg', 'w')
count=0
for hid in existdict:
#if hid >= N and existdict[hid] == True:
if existdict[hid] == True:
#fout.write('%d: %r, %s' % (hid, existdict[hid], hcdict[hid].writeString()))
fout.write('%s,%s\n' % (preffix, hcdict[hid].writeLeaves(resimap)))
count+=len(hcdict[hid].leaves)
print '%d leaves in total\n' % count
def writeSPDB(self):
aamap = AAmap()
fo = open(self.pdbfile+'.spdb', 'w')
for a in self.atoms:
fo.write('%f %f %f %d %s\n' % (a.x, a.y, a.z, a.resSeq, aamap.getAAmap(a.resName)))
fo.close()
def __init__(self,
pdbname,
chain='all',
top='',
pfam='',
center='CA',
cutoff=5,
scutoff=1,
flag=0,
desc='',
nbcutoff=4):
self.atoms = []
#dictionary for pairwise distance
self.pairwiseDict = {}
self.clusters = []
#pdb, top, pfam, str, pdbidx, seqheader, alignstr, alignidx, center, cutoff, scutoff, flag, desc
#self.pdb = pdbname[len(pdbname)-8:len(pdbname)-4]
self.pdbfile = pdbname
self.pdb = pdbname[:-4]
self.chain = chain
self.top = top
self.pfam = pfam
self.center = center
self.cutoff = cutoff
self.scutoff = scutoff
self.seqheader = self.pdb
self.flag = flag
self.desc = desc
self.nbcutoff = nbcutoff
self.ca = []
fin = open(pdbname, 'r')
lines = fin.readlines()
fin.close()
lastname = ''
lastres = ''
aamap = AAmap()
for i in xrange(0, len(lines)):
line = lines[i]
# load only one model
if 'END' in line[0:6]:
break
if line[17:20].strip() not in aamap.AAA2A:
continue
if self.chain != 'all':
if (self.chain != line[21]):
continue
if line[0:6] == 'ATOM ':
at = atom(lines[i])
if (at.name == lastname) and (at.resSeq == lastres):
#print '[%s]::alter loc:\n%s' % (self.pdbfile, lines[i])
#if (line[16]==' ' or line[16]=='A'): # to avoid alternative location
continue
else:
self.atoms.append(at)
if at.name.strip() == 'CA':
self.ca.append(at)
lastname = at.name
lastres = at.resSeq
# map for Chain+Resi : (index in sequence, ResName)
# 'B529': (132, 'V')
self.resDict = {} # assigned in self.getSeq() function
# resAtoms, a list of lists, each (element) list contains atoms of residues
# resArray, gives a list of keys eg. (A,Q,70), (A,I,71), (A,V,72)
self.seq, self.resArray, self.resAtoms = self.getSeq()
# some residue does not have CA!! 1e6i.aln.pdb the last residue
#aamap = AAmap()
#self.seq = ''.join([aamap.getAAmap(a.resName) for a in self.ca])
# map for sequence index: Chain+Resi(ResName)
# 132 : 'B529(V)'
self.seqDict = {-1: '.'}
for r in self.resDict:
self.seqDict[self.resDict[r][0]] = '%s(%s)' % (r,
self.resDict[r][1])
