def getAncestry(self, parent, immediate=None):
"""Find out if a record is parent of another record using
sequence information and get operations"""
import sequence_alignment
from Sequence import SequenceOperations as SQ
full_record_sequence = self.aaseq
full_parent_sequence = parent.aaseq
if parent.has_key('Structure_alnseq_PDBfile') and parent.has_key(
'Structure_alnseq_EATrecord'):
parent_aln_PDBfile = parent.Structure_alnseq_PDBfile + '*'
parent_aln_record = parent.Structure_alnseq_EATrecord + '*'
else:
parent_aln_PDBfile = record_sequence, ignored_res = sequence_alignment.Protool2pir(
full_record_sequence)
parent_aln_record = parent_aln_PDBfile
if full_record_sequence and full_parent_sequence:
record_sequence, ignored_res = sequence_alignment.Protool2pir(
full_record_sequence)
parent_sequence, ignored_parent = sequence_alignment.Protool2pir(
full_parent_sequence)
# First try the simple option
operations = [None]
if len(record_sequence) == len(parent_sequence):
operations = SQ.findSequenceDifferences(
record_sequence,
parent_sequence,
full_parent_sequence,
PDBaln=parent_aln_PDBfile,
recordALN=parent_aln_record)
if len(operations) > 10 or len(record_sequence) != len(
parent_sequence):
NW_align = sequence_alignment.NW(record_sequence,
parent_sequence)
al_seq, al_parent, map_seq, map_parent = NW_align.Align(
verbose=True)
operations = SQ.findSequenceDifferences(
al_seq,
al_parent,
full_parent_sequence,
PDBaln=parent_aln_PDBfile,
recordALN=parent_aln_record)
if operations == [None]:
raise Exception()
if len(operations) >= 0 and not immediate:
return True, operations
elif len(operations) > 0 and immediate:
if len(operations) == 1:
return False, operations
else:
return False, operations
return False, False
def align_withseq(self, sequence):
"""Extract the sequence of the current molecule and align it with the sequence given.
Return an array giving the relation between the sequence position and the PDB residue ID"""
pdbpir = self.clean_seq(self.PirSeq(), remove_gaps=True)
sequence = self.clean_seq(sequence, remove_gaps=True)
import sequence_alignment
NW = sequence_alignment.NW(pdbpir, sequence)
al1, al2, map1, map2 = NW.Align(verbose=True)
map = []
residues = self.residues.keys()
residues.sort()
count = 0
#
for count in range(len(sequence)):
pdbrespos = map2[count]
if pdbrespos != '-':
map.append(residues[map2[count]])
else:
map.append(None)
#
# Check that we have 100% sequence identify for the aligned residues
#
print 'Aligned seqid', NW.aligned_seqid
return map, NW.aligned_seqid
def main():
import os, sys, math
sys.path.append('/home/people/tc/svn/Protool/')
import geometry
instance_geometry = geometry.geometry()
## ## lactoferrin
## pdb1 = '1lfg'
## pdb2 = '1lfh'
## domain_range = range(1,88+1)+range(253,333+1)
## chain = 'A'
## ## trp repressor
## pdb1 = '1wrp'
## pdb2 = '2oz9' ## 2wrp
## pdb2 = '1zt9'
## domain_range = range(1,999+1)
## chain1 = 'R'
## chain2 = 'R'
## chain2 = 'A'
## exclude_chain = ''
## ## luciferase
## pdb1 = '1ba3'
## pdb2 = '1lci'
## domain_range = range(1,999+1)
## chain1 = 'A'
## chain2 = 'A'
## exclude_chain = ''
## ## G3P DH
## pdb1 = '1gd1'
## pdb2 = '2gd1'
## domain_range = range(1,999+1)
## chain1 = 'O'
## chain2 = 'O'
## exclude_chain = ''
## ## hexokinase
## pdb1 = '1hkg'
## pdb2 = '2yhx'
## domain_range = range(1,999+1)
## chain1 = 'A'
## chain2 = 'A'
## exclude_chain = ''
## ## adk
## pdb1 = '1ake'
## pdb2 = '4ake'
## domain_range = range(1,999+1)
## chain1 = 'A'
## chain2 = 'A'
## exclude_chain = 'B'
## ## t4l
## pdb1 = '2lzm'
## pdb2 = '150l'
#### domain_range = range(15,59+1)
#### domain_range = range(60,80+1)
## domain_range = range(81,162+1)
## chain1 = 'A'
## chain2 = 'D'
## exclude_chain = 'B'
l_input = [
## ##
## {'pdb1':'1ipd','pdb2':'1osj','chain1':'A','chain2':'A','range':range(1,98+1)+range(253,345+1)},
## {'pdb1':'1ipd','pdb2':'1osj','chain1':'A','chain2':'A','range':range(99,108+1)+range(109,252+1)},
###### shears
## ## aspartate amino transferase
## {'pdb1':'9aat','pdb2':'1ama','chain1':'A','chain2':'A','range':range(15,36+1)+range(349,410+1)},
## {'pdb1':'9aat','pdb2':'1ama','chain1':'A','chain2':'A','range':range(50,312+1)},
## alcohol dehydrogenase
{
'pdb1': '6adh',
'pdb2': '8adh1',
'chain1': 'A',
'chain2': 'A',
'range': range(1, 174 + 1) + range(322, 374 + 1)
},
{
'pdb1': '6adh',
'pdb2': '8adh2',
'chain1': 'A',
'chain2': 'A',
'range': range(193, 317 + 1)
},
## ## citrate synthase
## {'pdb1':'1cts','pdb2':'4cts','chain1':'A','chain2':'A','range':range(1,276+1)+range(386,999+1)},
###### hinges
## ## atpsulf
## {'pdb1':'1i2d','pdb2':'1m8p','chain1':'A','chain2':'A','range':range(1,389+1)},
## ## dnak (different spacegroups)
## {'pdb1':'1dkx','pdb2':'1dky','chain1':'A','chain2':'A','range':range(389,509+1)},
## ## dnak (different spacegroups)
## {'pdb1':'1ddt','pdb2':'1mdt','chain1':'A','chain2':'A','range':range(1,376+1)},
## ## ecpdpbp
## {'pdb1':'1dpp','pdb2':'1dpe','chain1':'A','chain2':'A','range':range(1,260+1)+range(479,999+1)},
## ## ef2
## {'pdb1':'1n0v','pdb2':'1n0u','chain1':'C','chain2':'A','range':range(1,478+1)}, ## large
## {'pdb1':'1n0v','pdb2':'1n0u','chain1':'C','chain2':'A','range':range(479,560+1)}, ## independent
## {'pdb1':'1n0v','pdb2':'1n0u','chain1':'C','chain2':'A','range':range(561,9999+1)}, ## small
## ## febp
## {'pdb1':'1d9v','pdb2':'1mrp','chain1':'A','chain2':'A','range':range(109,227+1)+range(292,309+1)},
## {'pdb1':'1d9v','pdb2':'1mrp','chain1':'A','chain2':'A','range':range(1,96+1)+range(228,262+1)},
## ## folylpolyglutamate synthetase
## {'pdb1':'1jbv','pdb2':'1jbw','chain1':'A','chain2':'A','range':range(1,295+1)},
## {'pdb1':'1jbv','pdb2':'1jbw','chain1':'A','chain2':'A','range':range(296,386+1)},
## ## glucose ABC transporter ATPase subunit (different spacegroups)
## {'pdb1':'1oxs','pdb2':'1oxu','chain1':'C','chain2':'C','range':range(1,209+1)},
## {'pdb1':'1oxs','pdb2':'1oxu','chain1':'C','chain2':'C','range':range(244,999+1)},
## ## groel domain
## {'pdb1':'1aon','pdb2':'1oel','chain1':'A','chain2':'A','range':range(1,137+1)+range(410,999+1)},
## {'pdb1':'1aon','pdb2':'1oel','chain1':'A','chain2':'A','range':range(192,374+1)},
## {'pdb1':'1aon','pdb2':'1oel','chain1':'A','chain2':'A','range':range(138,190+1)+range(375,409+1)},
## ## lao bp
## {'pdb1':'2lao','pdb2':'1laf','chain1':'A','chain2':'E','range':range(1,90+1)+range(192,238+1)},
## {'pdb1':'2lao','pdb2':'1laf','chain1':'A','chain2':'E','range':range(91,191+1)},
## ## t4l
## {'pdb1':'1l96','pdb2':'1l97','chain1':'A','chain2':'A','range':range(13,59+1)},
## {'pdb1':'1l96','pdb2':'1l97','chain1':'A','chain2':'A','range':range(81,164+1)},
## ## maltodextrin bp
## {'pdb1':'1omp','pdb2':'3mbp','chain1':'A','chain2':'A','range':range(1,104+1)+range(268,313+1)},
## {'pdb1':'1omp','pdb2':'3mbp','chain1':'A','chain2':'A','range':range(113,258+1)+range(314,370+1)},
## ## mRNA capping enzyme
## {'pdb1':'1ckm','pdb2':'1ckm','chain1':'A','chain2':'B','range':range(1,237+1)+range(319,327+1)},
## {'pdb1':'1ckm','pdb2':'1ckm','chain1':'A','chain2':'B','range':range(241,303+1)},
## ## mura
## {'pdb1':'1ejd','pdb2':'1a2n','chain1':'A','chain2':'A','range':range(1,20+1)+range(230,417+1)},
## {'pdb1':'1ejd','pdb2':'1a2n','chain1':'A','chain2':'A','range':range(20,230+1)},
## ## oligopeptide bp
## {'pdb1':'1rkm','pdb2':'2rkm','chain1':'A','chain2':'A','range':range(1,263+1)+range(491,517+1)},
## {'pdb1':'1rkm','pdb2':'2rkm','chain1':'A','chain2':'A','range':range(277,477+1)},
## ## protein kinase A
## {'pdb1':'1jlu','pdb2':'1cmk','chain1':'E','chain2':'E','range':range(1,33+1)+range(125,310+1),
## {'pdb1':'1jlu','pdb2':'1cmk','chain1':'E','chain2':'E','range':range(34,124+1)},
## ## dna polymerase beta
## {'pdb1':'1bpd','pdb2':'2bpg','chain1':'A','chain2':'A','range':range(1,82+1)},
## {'pdb1':'1bpd','pdb2':'2bpg','chain1':'A','chain2':'A','range':range(106,132+1)},
## {'pdb1':'1bpd','pdb2':'2bpg','chain1':'A','chain2':'A','range':range(148,262+1)},
## {'pdb1':'1bpd','pdb2':'2bpg','chain1':'A','chain2':'A','range':range(262,335+1)},
## ## ribose bp
## {'pdb1':'1urp','pdb2':'2dri','chain1':'A','chain2':'A','range':range(1,98+1)+range(235,259+1)},
## {'pdb1':'1urp','pdb2':'2dri','chain1':'A','chain2':'A','range':range(104,234+1)+range(265,271+1)},
## ## thioredoxin reductase
## {'pdb1':'1tde','pdb2':'1f6m','chain1':'A','chain2':'E','range':range(1,112+1)+range(248,320+1)},
## {'pdb1':'1tde','pdb2':'1f6m','chain1':'A','chain2':'E','range':range(118,242+1)},
## ## dna bp
## {'pdb1':'1fgu','pdb2':'1jmc','chain1':'A','chain2':'A','range':range(183,283+1)},
## ## transferrin
## {'pdb1':'1bp5','pdb2':'1a8e','chain1':'A','chain2':'A','range':range(1,75+1)+range(249,316+1)},
## {'pdb1':'1bp5','pdb2':'1a8e','chain1':'A','chain2':'A','range':range(103,242+1)},
## ## uracil dna glycosylase
## {'pdb1':'1ssp','pdb2':'1akz','chain1':'E','chain2':'A','range':range(82,144+1)+range(191,240+1)},
## {'pdb1':'1ssp','pdb2':'1akz','chain1':'E','chain2':'A','range':range(166,182+1)+range(270,304+1)},
]
for i in range(len(l_input)):
pdb1 = l_input[i]['pdb1']
pdb2 = l_input[i]['pdb2']
chain1 = l_input[i]['chain1']
chain2 = l_input[i]['chain2']
domain_range = l_input[i]['range']
os.system('cp /oxygenase_local/data/pdb/%s/pdb%s.ent %s.pdb' % (
pdb1[1:3],
pdb1,
pdb1,
))
os.system('cp /oxygenase_local/data/pdb/%s/pdb%s.ent %s.pdb' % (
pdb2[1:3],
pdb2[:4],
pdb2[:4],
))
ss_range1, l_missing1, seqres1, l_modres = parse_header(
pdb1,
chain1,
)
ss_range2, l_missing2, seqres2, l_modres = parse_header(
pdb2[:4],
chain2,
)
ss_range = list(set(ss_range1) & set(ss_range2))
l_missing = list(set(l_missing1) | set(l_missing2))
if len(seqres1) != len(seqres2):
d_replace = {
'TPO': 'THR',
'PTR': 'TYR',
## 'SER':'CYS', ## 1tde v 1f6m
}
for i in range(len(seqres1)):
if seqres1[i] in d_replace.keys():
seqres1[i] = d_replace[seqres1[i]]
for i in range(len(seqres2)):
if seqres2[i] in d_replace.keys():
seqres2[i] = d_replace[seqres2[i]]
if not (''.join(seqres1) in ''.join(seqres2)
or ''.join(seqres2) in ''.join(seqres1)):
import sys
sys.path.append('/home/people/tc/svn/EAT_DB/')
import sequence_alignment
d_res = {
'ALA': 'A',
'CYS': 'C',
'ASP': 'D',
'GLU': 'E',
'PHE': 'F',
'GLY': 'G',
'HIS': 'H',
'ILE': 'I',
'LYS': 'K',
'LEU': 'L',
'MET': 'M',
'ASN': 'N',
'PRO': 'P',
'GLN': 'Q',
'ARG': 'R',
'SER': 'S',
'THR': 'T',
'VAL': 'V',
'TRP': 'W',
'TYR': 'Y',
}
seq1 = ''
for res in seqres1:
seq1 += d_res[res]
seq2 = ''
for res in seqres2:
seq2 += d_res[res]
instance = sequence_alignment.NW(seq1, seq2)
s1, s2 = instance.Align(verbose=False)[:2]
l1 = len(s1) - len(s1.lstrip('-'))
l2 = len(s2) - len(s2.lstrip('-'))
r1 = len(s1) - len(s1.rstrip('-'))
r2 = len(s2) - len(s2.rstrip('-'))
print seqres1
print seqres2
print len(seqres1)
print len(seqres2)
print pdb1, pdb2
print l1, l2, r1, r2
print seqres2[l1:len(seqres2) - r1]
print seqres1[l2:len(seqres1) - r2]
l_coordinates1 = parse_coordinates(
pdb1,
chain1,
domain_range,
ss_range,
l_missing,
)
l_coordinates2 = parse_coordinates(
pdb2[:4],
chain2,
domain_range,
ss_range,
l_missing,
)
## l_coordinates1 = l_coordinates1[l2:len(l_coordinates1)-r2]
## l_coordinates2 = l_coordinates2[l1:len(l_coordinates2)-r1]
if len(l_coordinates1) != len(l_coordinates2):
print len(l_coordinates1)
print len(l_coordinates2)
stop
rmsd = instance_geometry.superpose(l_coordinates1, l_coordinates2)
print pdb1, pdb2, round(rmsd, 1), len(l_coordinates1) / 3.
tv1 = instance_geometry.fitcenter
rm = instance_geometry.rotation
tv2 = instance_geometry.refcenter
apply_transformation_matrix(
pdb1,
chain1,
l_modres,
[0, 0, 0],
[[1, 0, 0], [0, 1, 0], [0, 0, 1]],
[0, 0, 0],
)
apply_transformation_matrix(
pdb2,
chain2,
l_modres,
tv1,
rm,
tv2,
)
l_coordinates1 = parse_coordinates(
pdb1 + '_rotated',
chain1,
range(1, 9999),
ss_range,
l_missing,
)
l_coordinates2 = parse_coordinates(
pdb2 + '_rotated',
chain2,
range(1, 9999),
ss_range,
l_missing,
)
SUM = 0.
n = len(l_coordinates1)
for i in range(n):
SUM += sum((l_coordinates1[i] - l_coordinates2[i])**2)
RMSD = math.sqrt(SUM / n)
print RMSD
return
def main(
pdb1,
pdb2,
chains1_align,
chains2_align,
):
chains1_apply = chains1_align
chains2_apply = chains2_align
import os, sys, math
sys.path.append('/home/people/tc/svn/Protool/')
import geometry
instance_geometry = geometry.geometry()
domain_range = range(0, 9999)
os.system('cp /data/pdb-v3.2/%s/pdb%s.ent %s.pdb' % (
pdb1[1:3],
pdb1,
pdb1,
))
os.system('cp /data/pdb-v3.2/%s/pdb%s.ent %s.pdb' % (
pdb2[1:3],
pdb2,
pdb2,
))
ss_range1, l_missing1, seqres1, l_modres = parse_header(
pdb1,
chains1_align,
)
ss_range2, l_missing2, seqres2, l_modres = parse_header(
pdb2,
chains2_align,
)
ss_range = list(set(ss_range1) & set(ss_range2))
l_missing = list(set(l_missing1) | set(l_missing2))
if len(seqres1) != len(seqres2):
d_replace = {
'TPO': 'THR',
'PTR': 'TYR',
## 'SER':'CYS', ## 1tde v 1f6m
}
for i in range(len(seqres1)):
if seqres1[i] in d_replace.keys():
seqres1[i] = d_replace[seqres1[i]]
for i in range(len(seqres2)):
if seqres2[i] in d_replace.keys():
seqres2[i] = d_replace[seqres2[i]]
if not (''.join(seqres1) in ''.join(seqres2)
or ''.join(seqres2) in ''.join(seqres1)):
import sys
sys.path.append('/home/people/tc/svn/EAT_DB/')
import sequence_alignment
d_res = {
'ALA': 'A',
'CYS': 'C',
'ASP': 'D',
'GLU': 'E',
'PHE': 'F',
'GLY': 'G',
'HIS': 'H',
'ILE': 'I',
'LYS': 'K',
'LEU': 'L',
'MET': 'M',
'ASN': 'N',
'PRO': 'P',
'GLN': 'Q',
'ARG': 'R',
'SER': 'S',
'THR': 'T',
'VAL': 'V',
'TRP': 'W',
'TYR': 'Y',
}
seq1 = ''
for res in seqres1:
seq1 += d_res[res]
seq2 = ''
for res in seqres2:
seq2 += d_res[res]
instance = sequence_alignment.NW(seq1, seq2)
s1, s2 = instance.Align(verbose=False)[:2]
l1 = len(s1) - len(s1.lstrip('-'))
l2 = len(s2) - len(s2.lstrip('-'))
r1 = len(s1) - len(s1.rstrip('-'))
r2 = len(s2) - len(s2.rstrip('-'))
print seqres1
print seqres2
print len(seqres1)
print len(seqres2)
print pdb1, pdb2
print l1, l2, r1, r2
print seqres2[l1:len(seqres2) - r1]
print seqres1[l2:len(seqres1) - r2]
else:
s1 = ''.join(seqres1)
s2 = ''.join(seqres2)
if s1 in s2:
seqres2 = seqres2[s2.index(s1) / 3:]
else:
seqres1 = seqres1[s1.index(s2) / 3:]
if len(seqres1) != len(seqres2):
print len(seqres1), len(seqres2)
stop
l_coordinates1 = parse_coordinates(
pdb1,
chains1_align,
domain_range,
ss_range,
l_missing,
)
l_coordinates2 = parse_coordinates(
pdb2,
chains2_align,
domain_range,
ss_range,
l_missing,
)
## l_coordinates1 = l_coordinates1[l2:len(l_coordinates1)-r2]
## l_coordinates2 = l_coordinates2[l1:len(l_coordinates2)-r1]
if len(l_coordinates1) == 0 or len(l_coordinates2) == 0:
stop
if len(l_coordinates1) != len(l_coordinates2):
print len(l_coordinates1)
print len(l_coordinates2)
stop
rmsd = instance_geometry.superpose(l_coordinates1, l_coordinates2)
print pdb1, pdb2
print 'rmsd', round(rmsd, 1)
print 'residues', len(seqres1), len(seqres2)
print 'coordinates', len(l_coordinates1)
tv1 = instance_geometry.fitcenter
rm = instance_geometry.rotation
tv2 = instance_geometry.refcenter
lines1 = apply_transformation_matrix(
pdb1,
chains1_apply,
l_modres,
[0, 0, 0],
[[1, 0, 0], [0, 1, 0], [0, 0, 1]],
[0, 0, 0],
)
lines2 = apply_transformation_matrix(
pdb2,
chains2_apply,
l_modres,
tv1,
rm,
tv2,
)
fd = open('rotated_%s%s.pdb' % (
pdb1,
pdb2,
), 'w')
fd.writelines(lines1 + lines2)
fd.close()
l_coordinates1 = parse_coordinates(
'rotated_' + pdb1,
chains1_apply,
range(-9999, 9999),
ss_range,
l_missing,
)
l_coordinates2 = parse_coordinates(
'rotated_' + pdb2,
chains2_apply,
range(-9999, 9999),
ss_range,
l_missing,
)
SUM = 0.
n = len(l_coordinates1)
for i in range(n):
SUM += sum((l_coordinates1[i] - l_coordinates2[i])**2)
RMSD = math.sqrt(SUM / n)
print 'RMSD all atoms', RMSD
return RMSD, l_coordinates1, l_coordinates2
def checkPDBSequence(self, name):
"""Check the PDB sequence against a newly added structure, optional.
Adds the amino acid seq of the PDB file, overwriting the old one"""
# Extract the sequence
import sequence_alignment
pdb_1, ignored_res1 = sequence_alignment.Protool2pir(self.X.sequence)
print 'IGNORED', ignored_res1
if ignored_res1 != {}:
igroups = ignored_res1.keys()
igroups.sort()
import tkMessageBox
tkMessageBox.showwarning(
'Unknown entities in PDB file',
'I ignored the following residue types/molecules in the PDB file:\n%s'
% (str(igroups)))
# Get the entry sequence
accept_alignment_automatically = None
record_AA = DB.get_AA_sequence(name)
if record_AA:
record_AA1, ignored_res = sequence_alignment.Protool2pir(record_AA)
# If we do not have an amino acid sequence for the record, then
# we simply use the one from the PDB file and accept the alignment
# straight away
accept_alignment_automatically = 1
import copy
record_AA1 = copy.deepcopy(pdb_1)
# Also deposit the amino acid sequence in the protein record
DB.data[name]['aaseq'] = copy.deepcopy(self.X.sequence)
# Align the two sequences
NW_align = sequence_alignment.NW(pdb_1, record_AA1)
al_pdb, al_record, map_pdb, map_record = NW_align.Align()
self.al_pdb = al_pdb
self.al_record = al_record
# Find regions of overlap
ids = 0
for count in range(len(al_pdb)):
res_pdb = al_pdb[count]
res_rec = al_record[count]
if res_pdb == res_rec:
ids = ids + 1
print 'Sequence identity %5.3f' % (100.0 * float(ids) /
float(len(al_pdb)))
AlignmentMap = {}
AlignmentMap['OrigAa'] = al_record
AlignmentMap['AlignedAa'] = al_pdb
#Make alignment window
AlignWindow = Toplevel()
self.AlingWindow = AlignWindow
AlignWindow.geometry('+100+200')
AlignWindow.title('Please check alignment')
AlignWindow.button = Button(AlignWindow, {
"text": "Alignment OK",
"fg": "black",
"command": storePDB
})
AlignWindow.button.grid(row=3, column=0)
AlignWindow.button = Button(
AlignWindow, {
"text": "Alignment not OK",
"fg": "black",
"command": AlignWindow.destroy
})
AlignWindow.button.grid(row=3, column=1)
AlignWindow.Slider = Scrollbar(AlignWindow, orient=HORIZONTAL)
AlignWindow.Slider.grid(row=1, column=0, sticky='news', columnspan=2)
listbox = Listbox(AlignWindow, {
"height": 2,
"width": 80,
"font": "courier 14"
})
listbox.insert('end', "PEAT_DB record: " + al_record)
listbox.insert('end', "PDB file : " + al_pdb)
listbox.grid(row=0, column=0, columnspan=2)
listbox.config(xscrollcommand=AlignWindow.Slider.set)
AlignWindow.Slider.config(command=listbox.xview)
return AlignmentMap