def score(PDBfile):
"""
Calculates the m-score for a given PDB file
arguments:
PDBfile - the PDB file to score
hidden arguments:
aas.scr, pro.scr, gly.scr - the scoring tables
need to be present in working directory
"""
from pro_angle import find_residue
from Bio.PDB.PDBParser import PDBParser
from pro_length import length
(aas, gly, pro) = load_scores() ##define global tables
score = 0 #initialize
pars = PDBParser(PERMISSIVE = 1)
struct = pars.get_structure(PDBfile.rstrip('.pdb'), PDBfile)
model = struct.child_list[0]
chain = model.child_list[0]
pro_list = find_residue(chain, 'PRO')
gly_list = find_residue(chain, 'GLY')
aas_list = range(chain.child_list[1].id[1],
chain.child_list[len(chain)-1].id[1])
#need to remove pro/gly indices in first/last position
if pro_list.count(1) > 0:
pro_list.remove(1)
if pro_list.count(len(chain)-1) > 0:
pro_list.remove(len(chain)-1)
if gly_list.count(1) > 0:
gly_list.remove(1)
if gly_list.count(len(chain)-1) > 0:
gly_list.remove(len(chain)-1)
try:
for index in pro_list:
aas_list.remove(index) #remove pros from aas_list
for index in gly_list:
aas_list.remove(index) #remove glys from aas_list
except ValueError:
print 'incosistency in PDB file - will return score = 0'
return 0
else:
proscore = score_help(chain, pro_list, pro)
glyscore = score_help(chain, gly_list, gly)
aasscore = score_help(chain, aas_list, aas)
score = proscore+glyscore+aasscore
size=length(chain)
try:
score = (score/size)*1000 #normalize score
return score
except ZeroDivisionError:
print "calculated protein length 0 -> returning score 0"
score = 0
return score
def score(PDBfile):
"""
Calculates the m-score for a given PDB file
arguments:
PDBfile - the PDB file to score
hidden arguments:
aas.scr, pro.scr, gly.scr - the scoring tables
need to be present in working directory
"""
from pro_angle import find_residue
from Bio.PDB.PDBParser import PDBParser
from pro_length import length
(aas, gly, pro) = load_scores() ##define global tables
score = 0 #initialize
pars = PDBParser(PERMISSIVE=1)
struct = pars.get_structure(PDBfile.rstrip('.pdb'), PDBfile)
model = struct.child_list[0]
chain = model.child_list[0]
pro_list = find_residue(chain, 'PRO')
gly_list = find_residue(chain, 'GLY')
aas_list = range(chain.child_list[1].id[1],
chain.child_list[len(chain) - 1].id[1])
#need to remove pro/gly indices in first/last position
if pro_list.count(1) > 0:
pro_list.remove(1)
if pro_list.count(len(chain) - 1) > 0:
pro_list.remove(len(chain) - 1)
if gly_list.count(1) > 0:
gly_list.remove(1)
if gly_list.count(len(chain) - 1) > 0:
gly_list.remove(len(chain) - 1)
try:
for index in pro_list:
aas_list.remove(index) #remove pros from aas_list
for index in gly_list:
aas_list.remove(index) #remove glys from aas_list
except ValueError:
print 'incosistency in PDB file - will return score = 0'
return 0
else:
proscore = score_help(chain, pro_list, pro)
glyscore = score_help(chain, gly_list, gly)
aasscore = score_help(chain, aas_list, aas)
score = proscore + glyscore + aasscore
size = length(chain)
try:
score = (score / size) * 1000 #normalize score
return score
except ZeroDivisionError:
print "calculated protein length 0 -> returning score 0"
score = 0
return score
def score(PDBfile):
"""
Calculates the m-score for a given PDB file
arguments:
PDBfile - the PDB file to score
hidden arguments:
aas.scr, pro.scr, gly.scr - the scoring tables
need to be present in working directory
"""
from pro_angle import find_residue
from Bio.PDB.PDBParser import PDBParser
from pro_length import length
import os
import string
score = 0 #initialize
pars = PDBParser(PERMISSIVE = 1)
struct = pars.get_structure(PDBfile.rstrip('.pdb'), PDBfile)
model = struct.child_list[0]
chain = model.child_list[0]
score = float(0)
size=length(chain)
for res_index in range(1, size-2): #not first or last res
res = chain.child_list[res_index]
cur = res.resname
pre = chain.child_list[res_index-1].resname
pos = chain.child_list[res_index+1].resname
filename = pre + '_' + cur + '_' + pos + '.scr'
table_file = '/home/marciovm/proteins/bdtrimers/' + string.lower(cur) + '/' + filename
chain_index = chain.child_list[res_index].id[1]
table = load_scores(table_file)
if table != 0:
new = score_help(chain, chain_index, table)
else:
new = 0
score = score + new
try:
score = (score/size)*1000 #normalize score
return score
except ZeroDivisionError:
print "calculated protein length 0 -> returning score 0"
score = 0
return score
def score(PDBfile,
start_index=1,
end_index=9999,
smoothed='y',
mutated_list=' '):
"""
Calculates the m-score for a given PDB file
arguments:
PDBfile - the PDB file to score
start_index - child_list stat index of residues to look at
end_index - child_list end index
hidden arguments:
aas.scr, pro.scr, gly.scr - the scoring tables
need to be present in working directory
"""
from pro_angle import find_residue
from Bio.PDB.PDBParser import PDBParser
from pro_length import length
import os
import string
score = 0 #initialize
pars = PDBParser(PERMISSIVE=1)
struct = pars.get_structure(PDBfile.rstrip('.pdb'), PDBfile)
model = struct.child_list[0]
chain = model.child_list[0]
score = float(0)
size = length(chain)
if end_index > (size - 2):
end_index = size - 2
#non-mutated score
if mutated_list == ' ':
for res_index in range(start_index, end_index): #not first or last res
res = chain.child_list[res_index]
cur = res.resname
pre = chain.child_list[res_index - 1].resname
pos = chain.child_list[res_index + 1].resname
if smoothed == 'y':
filename = pre + '_' + cur + '_' + pos + '.smooth.scr'
else:
filename = pre + '_' + cur + '_' + pos + '.scr'
table_file = '/home/marciovm/proteins/bdtrimers/' + \
string.lower(cur) + '/' + filename
chain_index = chain.child_list[res_index].id[1]
# print 'loading table file: ' + table_file ### debugging
table = load_scores(table_file)
if table != 0:
new = score_help(chain, chain_index, table)
else:
new = 0
score = score + new
else:
mutated_form = mutated_list.split()
for res_index in range(start_index, end_index):
res = chain.child_list[res_index]
mutated_index = res_index - start_index
cur = mutated_form[res_index - start_index]
if mutated_index == 0:
pre = chain.child_list[res_index - 1].resname
else:
pre = mutated_form[mutated_index - 1]
if mutated_index == end_index - start_index - 1:
pos = chain.child_list[res_index + 1].resname
else:
pos = mutated_form[mutated_index + 1]
if smoothed == 'y':
filename = pre + '_' + cur + '_' + pos + '.smooth.scr'
else:
filename = pre + '_' + cur + '_' + pos + '.scr'
table_file = '/home/marciovm/proteins/bdtrimers/' + \
string.lower(cur) + '/' + filename
chain_index = chain.child_list[res_index].id[1]
# print 'loading table file: ' + table_file ##debugging
table = load_scores(table_file)
if table != 0:
new = score_help(chain, chain_index, table)
else:
new = 0
score = score + new
try:
score = (score / size) * 1000 #normalize score
return score
except ZeroDivisionError:
print "calculated protein length 0 -> returning score 0"
score = 0
return score
def compile_trimers():
import os
import commands
import sys
import string
from Bio.PDB.PDBParser import PDBParser
from pro_length import length
os.chdir('/home/marciovm/proteins/bigdist') ##choose directory
list = commands.getoutput('ls *.ent').split() #list of pdb files w/ chain names
start = sys.argv[1]
end = sys.argv[2]
aas = ['GLY', 'ALA', 'VAL', 'LEU', 'ILE', 'SER', 'THR', 'ASN', 'GLN',
'PHE', 'TYR', 'TRP', 'CYS', 'MET', 'PRO', 'ASP', 'GLU', 'LYS',
'ARG', 'HIS']
for pdb_index in range(string.atoi(start), string.atoi(end)):
pdb = list[pdb_index]
pars = PDBParser(PERMISSIVE = 1)
struct = pars.get_structure(pdb, pdb)
model = struct.child_list[0]
if pdb[4] == '.':
chain_name = ' '
else:
chain_name = string.upper(pdb[4])
chain = model.child_dict[chain_name]
print 'processing ' + pdb + ' index: ' + str(pdb_index)
##find first real residue
index = 0
res = 'first'
while aas.count(res) == 0 and index < 10000:
try:
res = chain.child_list[index].resname
except KeyError:
pass #nothing here
index += 1
res = chain.child_list[index].resname ##should be 2nd real residue
##find last real residue
final_res_index = length(chain) + index - 3 ##2nd to last real residue
##look at everything in between
for aa_index in range(index, final_res_index):
aacur = chain.child_list[aa_index].resname
aapre = chain.child_list[aa_index - 1].resname
aapost= chain.child_list[aa_index + 1].resname
seq = aapre + '_' + aacur + "_" + aapost + '.dat'
filename = ('/home/marciovm/proteins/bdtrimers/' + string.lower(aacur) + '/' + seq)
try:
file_out = open(filename, 'a')
except IOError:
print "IOError line 63, nostandard aa name in PDB"
else:
tuple = calc_all_dihedrals(chain, aa_index)
if tuple:
if len(tuple) > 2:
file_out.write(str(tuple[0]).zfill(6) + ' '
+str(tuple[1]).zfill(6)+ ' '
+str(tuple[2]).zfill(6)+ ' '
+str(tuple[3]).zfill(6)+'\n')
else:
file_out.write(str(tuple[0]).zfill(6)+ ' '
+str(tuple[1]).zfill(6)+'\n')
file_out.close()
def score(PDBfile, start_index = 1, end_index = 9999, smoothed = 'y', mutated_list = ' '):
"""
Calculates the m-score for a given PDB file
arguments:
PDBfile - the PDB file to score
start_index - child_list stat index of residues to look at
end_index - child_list end index
hidden arguments:
aas.scr, pro.scr, gly.scr - the scoring tables
need to be present in working directory
"""
from pro_angle import find_residue
from Bio.PDB.PDBParser import PDBParser
from pro_length import length
import os
import string
score = 0 #initialize
pars = PDBParser(PERMISSIVE = 1)
struct = pars.get_structure(PDBfile.rstrip('.pdb'), PDBfile)
model = struct.child_list[0]
chain = model.child_list[0]
score = float(0)
size=length(chain)
if end_index > (size - 2) :
end_index = size - 2
#non-mutated score
if mutated_list == ' ':
for res_index in range(start_index, end_index): #not first or last res
res = chain.child_list[res_index]
cur = res.resname
pre = chain.child_list[res_index-1].resname
pos = chain.child_list[res_index+1].resname
if smoothed == 'y':
filename = pre + '_' + cur + '_' + pos + '.smooth.scr'
else:
filename = pre + '_' + cur + '_' + pos + '.scr'
table_file = '/home/marciovm/proteins/bdtrimers/' + \
string.lower(cur) + '/' + filename
chain_index = chain.child_list[res_index].id[1]
# print 'loading table file: ' + table_file ### debugging
table = load_scores(table_file)
if table != 0:
new = score_help(chain, chain_index, table)
else:
new = 0
score = score + new
else:
mutated_form = mutated_list.split()
for res_index in range(start_index, end_index):
res = chain.child_list[res_index]
mutated_index = res_index - start_index
cur = mutated_form[res_index-start_index]
if mutated_index == 0:
pre = chain.child_list[res_index-1].resname
else:
pre = mutated_form[mutated_index-1]
if mutated_index == end_index - start_index - 1:
pos = chain.child_list[res_index+1].resname
else:
pos = mutated_form[mutated_index+1]
if smoothed == 'y':
filename = pre + '_' + cur + '_' + pos + '.smooth.scr'
else:
filename = pre + '_' + cur + '_' + pos + '.scr'
table_file = '/home/marciovm/proteins/bdtrimers/' + \
string.lower(cur) + '/' + filename
chain_index = chain.child_list[res_index].id[1]
# print 'loading table file: ' + table_file ##debugging
table = load_scores(table_file)
if table != 0:
new = score_help(chain, chain_index, table)
else:
new = 0
score = score + new
try:
score = (score/size)*1000 #normalize score
return score
except ZeroDivisionError:
print "calculated protein length 0 -> returning score 0"
score = 0
return score
def compile_trimers():
import os
import commands
import sys
import string
from Bio.PDB.PDBParser import PDBParser
from pro_length import length
os.chdir('/home/marciovm/proteins/bigdist') ##choose directory
list = commands.getoutput(
'ls *.ent').split() #list of pdb files w/ chain names
start = sys.argv[1]
end = sys.argv[2]
aas = [
'GLY', 'ALA', 'VAL', 'LEU', 'ILE', 'SER', 'THR', 'ASN', 'GLN', 'PHE',
'TYR', 'TRP', 'CYS', 'MET', 'PRO', 'ASP', 'GLU', 'LYS', 'ARG', 'HIS'
]
for pdb_index in range(string.atoi(start), string.atoi(end)):
pdb = list[pdb_index]
pars = PDBParser(PERMISSIVE=1)
struct = pars.get_structure(pdb, pdb)
model = struct.child_list[0]
if pdb[4] == '.':
chain_name = ' '
else:
chain_name = string.upper(pdb[4])
chain = model.child_dict[chain_name]
print 'processing ' + pdb + ' index: ' + str(pdb_index)
##find first real residue
index = 0
res = 'first'
while aas.count(res) == 0 and index < 10000:
try:
res = chain.child_list[index].resname
except KeyError:
pass #nothing here
index += 1
res = chain.child_list[index].resname ##should be 2nd real residue
##find last real residue
final_res_index = length(chain) + index - 3 ##2nd to last real residue
##look at everything in between
for aa_index in range(index, final_res_index):
aacur = chain.child_list[aa_index].resname
aapre = chain.child_list[aa_index - 1].resname
aapost = chain.child_list[aa_index + 1].resname
seq = aapre + '_' + aacur + "_" + aapost + '.dat'
filename = ('/home/marciovm/proteins/bdtrimers/' +
string.lower(aacur) + '/' + seq)
try:
file_out = open(filename, 'a')
except IOError:
print "IOError line 63, nostandard aa name in PDB"
else:
tuple = calc_all_dihedrals(chain, aa_index)
if tuple:
if len(tuple) > 2:
file_out.write(
str(tuple[0]).zfill(6) + ' ' +
str(tuple[1]).zfill(6) + ' ' +
str(tuple[2]).zfill(6) + ' ' +
str(tuple[3]).zfill(6) + '\n')
else:
file_out.write(
str(tuple[0]).zfill(6) + ' ' +
str(tuple[1]).zfill(6) + '\n')
file_out.close()