def __init__(self,path, pdb_id='pdb', cif_id='cif'):
# copied from input_output.py _read_structure
self.file_name = os.path.basename(path).split('.')[0]
self.file_sufix = os.path.basename(path).split('.')[1]
self.dir_path = os.path.dirname(path)
self.params = CP.read_charmm_FF()
self.chains = []
self.models = {}
if self.file_sufix == 'pdb':
self.header = struct.parse_pdb_header(path)
self.structure = struct.PDBParser(QUIET=True).get_structure(pdb_id, path)
self.has_sequence = False
elif self.file_sufix == 'cif':
self.header = struct.MMCIF2Dict()
self.structure = struct.MMCIFParser().get_structure(cif_id, path)
self.has_sequence = True
else:
print("ERROR: Unreognized file type " + self.file_sufix + " in " + self.file_name)
sys.exit(1)
def main():
usage = "usage: %prog [options] arg"
d = "This program reads a CSV file that has been generated by Super_Structure.\
One residue or terminal will be deleted at the time.\n \
WARNING: Deleting residues will leave a 'hole' in the structure. Amino Acids will not be renumbered.\n \
This program can only delete residues or terminals that are in the parameter file.\
The program will create a new and modified CSV file with the name of the input file plus the entity number."
opt_parser = optparse.OptionParser(usage,description=d)
opt_parser.add_option("--rem", type="str",help="Enter Instruction for removing amino acid or terminal in hard \n \
quotes '\"'. Place: Amino Acid Number, Entity ID, Chain ID, \
Residue or Terminal to be deleted separated by comas.\n \
Example \"1,1,A,ACE\", \"1,1,A,CTER\" or \"20,2,A,LYS\". \n \
Chain ID or terminal name are case sensitive and \
do not need to go in quotes.\n\
For residues all atoms will be deleted. For terminals, only the \
atoms that correspond to the terminal will be deleted.")
opt_parser.add_option("--inp", type="str",help="Path to CSV file for removing residue.")
opt_parser.add_option("--out", type="str",help="Name of output CSV file after removal of amino acid or terminal.")
opt_parser.add_option("--par", type="str",help="Path to charmm parameters folder.")
options, args = opt_parser.parse_args()
if not os.path.exists(options.inp):
print "Error: File path for Super Structure CSV file does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
params = CP.read_charmm_FF(options.par)
insulin = SS.Super_Structure(params, options.inp,'add_linker')
parse_list = options.rem.split(',')
#insulin = SS.Super_Structure(params, '/home/noel/Projects/Protein_design/EntropyMaxima/examples/Linker_minimization/2zta.csv','add_linker')
#parse_list = "1,1,A,CTER".split(',')
message = ''
if len(parse_list) == 4:
amino_acid_number = int(parse_list[0])
entity_number = int(parse_list[1])
chain = str(parse_list[2]).upper()
term_res = str(parse_list[3]).upper()
# So far this only works with natural aminoacids and ACE and CTER
if term_res in ['ILE','GLN','GLY','GLU','CYS','ASP','SER','HSD','HSE','PRO','HSP','ASN','VAL','THR','TRP','CTER',\
'LYS','PHE','ALA','MET','ACE','LEU','ARG','TYR']:
message += 'Deleting a '+term_res+' from '
message += 'amino acid number '+str(amino_acid_number)+' in entity id '+str(entity_number)+' and chain '+chain+'.\n'
insulin.delete_aa(amino_acid_number,entity_number,chain,term_res)
# If amino acid that is at the protein terminal is deleted, the terminal must be deleted too.
min_aa = min(insulin.Full_Structure.aaid[(insulin.Full_Structure.ent_id == entity_number) & (insulin.Full_Structure.chain == chain)])
max_aa = max(insulin.Full_Structure.aaid[(insulin.Full_Structure.ent_id == entity_number) & (insulin.Full_Structure.chain == chain)])
if amino_acid_number == min_aa:
del_terminal = False
# This checks that there is no ACETYL atoms and removes them it does.
for ii in insulin.Full_Structure.index[(insulin.Full_Structure.aaid == amino_acid_number) &\
(insulin.Full_Structure.ent_id == entity_number) &\
(insulin.Full_Structure.chain == chain)]:
if insulin.Full_Structure.loc[ii,'component'] == 'ACETY':
del_terminal = True
term_res = 'ACE'
break
if del_terminal:
insulin.delete_aa(amino_acid_number,entity_number,chain,term_res)
if amino_acid_number == max_aa:
if term_res != 'CTER':
del_terminal = False
for ii in insulin.Full_Structure.index[(insulin.Full_Structure.aaid == amino_acid_number) &\
(insulin.Full_Structure.ent_id == entity_number) &\
(insulin.Full_Structure.chain == chain)]:
if insulin.Full_Structure.loc[ii,'component'] == 'CTERM':
del_terminal = True
term_res = 'CTER'
break
if del_terminal:
insulin.delete_aa(amino_acid_number,entity_number,chain,term_res)
else:
pass
file_name = os.path.basename(options.out).split('.')[0]
dir_path = os.path.dirname(options.out)
# Super Structure needs to know about models.
num_models = len(range(((insulin.Full_Structure.shape[1]-20)/5)))+1
insulin.models = [str(i) for i in range(1,num_models+1)]
insulin.write_csv(dir_path,file_name)
IO.write_pdb(insulin, dir_path, file_name, 'all')
else:
print('ERROR: del_residue.py only works with natural aminoacids and ACE and CTER terminals.')
sys.exit(1)
else:
message += 'The number of entries in the instruction field, followed by -o or --rem, is not right.\n'
message += 'Type -h or --help for instructions\n'
print(message)
def mmgbsa_CA_bindingMatrix(self, options):
#################################### READ FILES #######################################################
#directory, filename = os.path.split()
crd_file = crd(self.dirpath + '/' + options.crd)
psf_file = psf(self.dirpath + '/' + options.psf)
out_file = os.path.basename(options.crd).split('.')[0]
###################### After reading files, Generate and Index and Super Structure ##############################
params = CP.read_charmm_FF(param_path)
insu = SS.Super_Structure(params, self.dirpath, 'charmm_input')
# At this point, a XPLOR psf could only have been creted from a complete structure, so no worries of gaps.
insu.create_super_structure_df_from_CRD_PSF(crd_file, psf_file)
################################################################################################################
insu.add_column_to_super_structure_df_from_CRD_PSF(
'gb_z0', self.dirpath + '/' + options.gb)
insu.add_column_to_super_structure_df_from_CRD_PSF(
'gb_z500', self.dirpath + '/' + options.gbz)
insu.add_column_to_super_structure_df_from_CRD_PSF(
'sa_z0', self.dirpath + '/' + options.sa)
insu.add_column_to_super_structure_df_from_CRD_PSF(
'sa_z500', self.dirpath + '/' + options.saz)
insu.create_column_with_ztranslated('z1_d', 'z1', "A", float(500))
###################### After reading files, Generate and Index a Super Structure ##############################
idx_ss = SS.index_super_structure(insu.Full_Structure)
idx_ss.generate_indexes_from_Full_Structure()
idx_ss.sort_comp_index_by_aaid_within_chains()
new_index = []
for i in range(len(idx_ss.comp_indx)):
if idx_ss.comp_indx[i][3] == 'PRO':
if idx_ss.comp_indx[i][4] == 'AMIN':
temp_list = range(idx_ss.comp_indx[i][5],
idx_ss.comp_indx[i][6] + 1)
temp_list = [temp_list[j] for j in [0, 4, 5, 1, 2, 3]]
for j in temp_list[3:]:
insu.Full_Structure.loc[j, 'component'] = 'SIDE2'
new_index = new_index + temp_list
elif idx_ss.comp_indx[i][4] == 'SIDE':
new_index = new_index + range(idx_ss.comp_indx[i][5],
idx_ss.comp_indx[i][6] + 1)
elif idx_ss.comp_indx[i][4] == 'CARB':
new_index = new_index + range(idx_ss.comp_indx[i][5],
idx_ss.comp_indx[i][6] + 1)
else:
new_index = new_index + range(idx_ss.comp_indx[i][5],
idx_ss.comp_indx[i][6] + 1)
insu.Full_Structure = insu.Full_Structure.loc[new_index]
insu.Full_Structure.component[(insu.Full_Structure.aa == 'PRO') & (
insu.Full_Structure.atmtyp1 == 'CB')] = 'SIDE3'
insu.Full_Structure.component[(insu.Full_Structure.aa == 'PRO') & (
insu.Full_Structure.atmtyp1 == 'HB1')] = 'SIDE3'
insu.Full_Structure.component[(insu.Full_Structure.aa == 'PRO') & (
insu.Full_Structure.atmtyp1 == 'HB2')] = 'SIDE3'
insu.Full_Structure.component[(insu.Full_Structure.aa == 'PRO') & (
insu.Full_Structure.atmtyp1 == 'CG')] = 'SIDE4'
insu.Full_Structure.component[(insu.Full_Structure.aa == 'PRO') & (
insu.Full_Structure.atmtyp1 == 'HG1')] = 'SIDE4'
insu.Full_Structure.component[(insu.Full_Structure.aa == 'PRO') & (
insu.Full_Structure.atmtyp1 == 'HG2')] = 'SIDE4'
insu.Full_Structure = insu.Full_Structure.reset_index(drop=True)
idx_ss = SS.index_super_structure(insu.Full_Structure)
idx_ss.generate_indexes_from_Full_Structure()
idx_ss.sort_comp_index_by_aaid_within_chains()
# CHARMM_Test/gbsw_ab_nbxmod5.dat
# ELECB,ELECU, GBENB, GBENU, VDWAB, VDWAU, ASPB, ASPU
# -379.654,-332.641,-495.252, -595.916, 5672.62, 2338.65, 37.9186, 48.284
ch_AB = insu.Full_Structure.loc[idx_ss.chn_indx[0][0]:idx_ss.chn_indx[-1][-1],['charg','epsilon',\
'rmin_half','x1','y1','z1','gb_z0','gb_z500','z1_d','sa_z0','sa_z500','atmtyp1','chain',\
'component','aa','aaid']].values
EE_K = -327.90
EE_K2 = -163.95
#Surface tension coefficient (sgamma) = 0.010 [kcal/mol/Angs**2]
SA_K = 0.010
l = len(idx_ss.comp_indx)
idxFile = open(self.dirpath + '/' + out_file + '_MMGBSA.idx', 'w')
if options.ver:
outFile = open(
self.dirpath + '/' + out_file + '_MMGBSA_verbose.out', 'w')
else:
outFile = open(self.dirpath + '/' + out_file + '_MMGBSA.out', 'w')
tempGB = 0
tempGB_Z = 0
tempSA = 0
tempSA_Z = 0
tempEE = 0
tempEE_Z = 0
tempVDW = 0
tempVDW_Z = 0
pairGB = 0
pairGB_Z = 0
DGB = 0.0
DGB_Z = 0.0
DEE = 0.0
DEE_Z = 0.0
DVW = 0.0
DVW_Z = 0.0
DSA = 0.0
DSA_Z = 0.0
for i in range(0, l):
idxFile.write(str(i)+" "+str(idx_ss.comp_indx[i][2])+" "+idx_ss.comp_indx[i][0]+" "+idx_ss.comp_indx[i][3]+\
" "+idx_ss.comp_indx[i][4]+"\n")
for k in range(idx_ss.comp_indx[i][5], idx_ss.comp_indx[i][6]):
temp5 = EE_K2 * ch_AB[k][0] * ch_AB[k][0] / ch_AB[k][6]
tempGB += temp5
temp5_Z = EE_K2 * ch_AB[k][0] * ch_AB[k][0] / ch_AB[k][7]
tempGB_Z += temp5_Z
tempSA += SA_K * ch_AB[k][9]
tempSA_Z += SA_K * ch_AB[k][10]
for kk in range(k + 1, idx_ss.comp_indx[i][6] + 1):
last = kk
d1 = np.linalg.norm(ch_AB[k][3:6] - ch_AB[kk][3:6])
r2 = d1 * d1
d1_Z = np.linalg.norm(ch_AB[k][3:6] - ch_AB[kk][3:6])
r2_Z = d1_Z * d1_Z
temp3 = ch_AB[k][6] * ch_AB[kk][6]
temp4 = (-1 * r2) / (4 * temp3)
temp3_Z = ch_AB[k][7] * ch_AB[kk][7]
temp4_Z = (-1 * r2_Z) / (4 * temp3_Z)
denm = np.sqrt(r2 + (temp3 * np.exp(temp4)))
denm_Z = np.sqrt(r2_Z + (temp3_Z * np.exp(temp4_Z)))
Kqq = EE_K * ch_AB[k][0] * ch_AB[kk][0]
temp5 = Kqq / denm
temp5_Z = Kqq / denm_Z
tempGB += temp5
tempGB_Z += temp5_Z
tempSA += SA_K * ch_AB[last][9]
tempSA_Z += SA_K * ch_AB[last][10]
temp5 = EE_K2 * ch_AB[last][0] * ch_AB[last][0] / ch_AB[last][6]
tempGB += temp5
temp5_Z = EE_K2 * ch_AB[last][0] * ch_AB[last][0] / ch_AB[last][7]
tempGB_Z += temp5_Z
DGB += tempGB
DGB_Z += tempGB_Z
DSA += tempSA
DSA_Z += tempSA_Z
if options.ver:
# Output for comparison with GB_Comp_Analy.prl
outFile.write("SELF "+str(i+1)+" "+str(i+1)+" "+idx_ss.comp_indx[i][0]+" "+str(idx_ss.comp_indx[i][2])\
+" "+str(idx_ss.comp_indx[i][3])+" "+str(idx_ss.comp_indx[i][4])+"\n GB "+\
str(tempGB)+"\n GBZ "+str(tempGB_Z)+"\n SA "+str(tempSA)+"\n SAZ "+\
str(tempSA_Z)+"\n")
else:
outFile.write(
str(i) + " " + str(i) + " " + str(tempGB - tempGB_Z) +
" " + str(tempSA - tempSA_Z) + "\n")
tempSA = 0.0
tempSA_Z = 0.0
for j in range(i + 1, l):
for m in range(idx_ss.comp_indx[i][5],
idx_ss.comp_indx[i][6] + 1):
for n in range(idx_ss.comp_indx[j][5],
idx_ss.comp_indx[j][6] + 1):
r = np.linalg.norm(ch_AB[m][3:6] - ch_AB[n][3:6])
r2 = r * r
r_Z = np.linalg.norm(np.array((ch_AB[m][3],ch_AB[m][4],ch_AB[m][8]))-\
np.array((ch_AB[n][3],ch_AB[n][4],ch_AB[n][8])))
r2_Z = r_Z * r_Z
temp = (332.06 * ch_AB[m][0] * ch_AB[n][0]) / r
temp_Z = (332.06 * ch_AB[m][0] * ch_AB[n][0]) / r_Z
tempEE += temp
tempEE_Z += temp_Z
Eps = np.sqrt(ch_AB[m][1] * ch_AB[n][1])
Rmin = ch_AB[m][2] + ch_AB[n][2]
A = Rmin / r
A_Z = Rmin / r_Z
A2 = A * A
A2_Z = A_Z * A_Z
A6 = A2 * A2 * A2
A6_Z = A2_Z * A2_Z * A2_Z
A12 = A6 * A6
A12_Z = A6_Z * A6_Z
tempVDW += Eps * (A12 - (2 * A6))
tempVDW_Z += Eps * (A12_Z - (2 * A6_Z))
temp3 = ch_AB[m][6] * ch_AB[n][6]
temp4 = (-1 * r2) / (4 * temp3)
temp3_Z = ch_AB[m][7] * ch_AB[n][7]
temp4_Z = (-1 * r2_Z) / (4 * temp3_Z)
denm = np.sqrt(r2 + (temp3 * np.exp(temp4)))
denm_Z = np.sqrt(r2_Z + (temp3_Z * np.exp(temp4_Z)))
temp5 = (EE_K * ch_AB[m][0] * ch_AB[n][0]) / denm
temp5_Z = (EE_K * ch_AB[m][0] * ch_AB[n][0]) / denm_Z
pairGB += temp5
pairGB_Z += temp5_Z
DGB += pairGB
DGB_Z += pairGB_Z
DEE += tempEE
DEE_Z += tempEE_Z
DVW += tempVDW
DVW_Z += tempVDW_Z
if options.ver:
# Output for comparison with GB_Comp_Analy.prl
outFile.write("PAIR " + str(i + 1) + " " + str(j + 1) +
" ")
outFile.write(idx_ss.comp_indx[i][0]+" "+str(idx_ss.comp_indx[i][2])+" "+str(idx_ss.comp_indx[i][3])\
+" "+str(idx_ss.comp_indx[i][4])+" - "+idx_ss.comp_indx[j][0]+" "+\
str(idx_ss.comp_indx[j][2])+" "+str(idx_ss.comp_indx[j][3])+" "+\
str(idx_ss.comp_indx[j][4])+"\n EE "+str(tempEE)+"\n EEZ "+str(tempEE_Z)+\
"\n VDW "+str(tempVDW)+"\n VDWZ "+str(tempVDW_Z)+"\n pGB "+str(pairGB)\
+"\n pGBZ "+str(pairGB_Z)+"\n")
else:
outFile.write(str(i)+" "+str(j)+" "+str(tempEE-tempEE_Z)+" "+str(tempVDW-tempVDW_Z)+" "+\
str(pairGB-pairGB_Z)+"\n")
pairGB = 0
pairGB_Z = 0
tempEE = 0
tempEE_Z = 0
tempVDW = 0
tempVDW_Z = 0
tempGB = 0
tempGB_Z = 0
tempSA = 0
tempSA_Z = 0
def main():
usage = "usage: %prog [options]"
d = "It takes two pdbs and aligns the structures in different orientations. A center-pdb is placed at (0,0,0),and \
a rotate-pdb is place around center at given angle intervals and distances. You can choose to join chains to make \
a fusion peptide, or, if no, you can just align proteins for searching protein-protein interactions."
option_parser = optparse.OptionParser(usage, description=d)
option_parser.add_option("--center", type="str", \
help="Path to pdb structure to place at (0,0,0) according to its center of mass.")
option_parser.add_option("--rotate", type="str", \
help="Path to pdb structure to rotated around center structure at X degree intervals. \
Angle intervals must be multiple of 360." )
option_parser.add_option("--angle", type="int", \
help="Angle intervals used to place rotated structured around center one.")
option_parser.add_option("--distance", type="float",\
help="Center of mass distance between rotated and centered structures.")
option_parser.add_option(
'--id',
type="str",
action="store",
default='s',
help="A prefix id to identify the output structures. Default is s.")
option_parser.add_option('--link', type="str",action = "store", default = 'x', \
help = "Option only works with a 'yes' value from the --join option. It will join chains \
in the following format, for example: 'A.r,B.f:B.f,A.f'. This --link option can be read as\
the centered structure's chain 'A' amino acid sequence numbering is reversed and joind to \
the rotated structure's chain B with amino acid sequenced numbers not reversed, or kept \
forward. The values after the column are read similarly. Default if missing: 'x'" )
option_parser.add_option('--join', type="str",action = "store", default = 'no', \
help = "Only two possible options 'CR' and 'RC'. In both options C stands for Center, and \
R for rotate. CR means that the numbering starts at the first amino acid of --center \
structure up to its end, and it is followed by the first amino acid of --rotate up to its \
end. This option might overide the directions (r or f) of amino acid numbering for each\
chain in --link option. This gives total manipulation flexibility for joining chains." )
options, args = option_parser.parse_args()
if not os.path.exists(options.center):
print "Error: File path for molecule to be centered does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
if not os.path.exists(options.rotate):
print "Error: File path for molecule to be rotated does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
if options.map.lower() == 'yes' and options.link.lower() == 'x':
print("Error: when option map is equal to yes, option link must be")
print(
" a series of chains identifiers for association as described"
)
print(" in the link paramater help.")
sys.exit(1)
pdb_parser = PDBParser(QUIET=True)
Angle = options.angle
distance = options.distance
#directory = os.path.dirname(options.center)
filepath1 = options.center
filepath2 = options.rotate
join_o = options.join
link_o = options.link
###########################################################################
# Uncomment to test from spyder IDE
#pdb_parser = PDBParser(QUIET = True)
#Angle = 45
#distance = 45
#file_name = os.path.basename(options.out).split('.')[0]
#directory = "/home/noel/Projects/Protein_design/ccl_lectures/Lecture_4/"
#filepath1 = directory+'2hiu_1rr.pdb'
#filepath2 = directory+'2zta_1rr.pdb'
#param_path = "/home/noel/Projects/Protein_design/EntropyMaxima/params/charmm27.ff/"
#join_o = "RC"
#link_o = "A.f,A.f:B.f,B.f"
####################################################################################################################
# Process strig that the determines how the centered and rotated structures will be connected.
params = CP.read_charmm_FF()
cmc = md.CenterOfMassCalculator(params)
rig = MRM.Molecular_Rigid_Manipulation(params)
####################################################################################################################
# Check that the structures only have one model, and Place the structures' center of mass at (0,0,0) to give an idea
# of their location in the cartesian coordinate system
s1 = pdb_parser.get_structure('Centered', filepath1)
countS1 = 0
modelS1 = -1
for i in s1.get_models():
countS1 += 1
modelS1 = i.id
if countS1 != 1:
print(
"ERROR: Number of models cannot be different from 1. Models found:"
+ str(countS1))
print(" Make sure Centered PDBs have only one model.")
sys.exit(1)
rig.translate_molecule(s1, modelS1,
rig.center_molecule(cmc.get_center_of_mass(s1)))
s2 = pdb_parser.get_structure('Rotated', filepath2)
countS2 = 0
modelS2 = -1
for i in s2.get_models():
countS2 += 1
modelS2 = i.id
if countS2 != 1:
print(
"ERROR: Number of models cannot be different from 1. Models found:"
+ str(countS2))
print(" Make sure Rotated PDBs have only one model.")
sys.exit(1)
rig.translate_molecule(s2, modelS2,
rig.center_molecule(cmc.get_center_of_mass(s2)))
####################################################################################################################
# TODO: This works only for angles between 0 and 90 not including 0, and 90 and will generate angles in all 8
# quadrants of the cartesian coordinate system (I do not see why using an angle other than 45 for now.)
# TODO: quaternions might work better.
# The location list has a list of normalized vectors releative to (0,0,0) that will be use to place s2's in the
# right orientation relative to s1
locations = []
angles = []
for h in range(0, 3):
for i in range(0, 360, Angle):
for j in range(0, 90 / Angle - 1):
if h == 0:
z = 0
angles.append(str(i) + "_" + str(0))
elif h == 1:
z = np.cos(45 * np.pi / 180)
angles.append(str(i) + "_" + str(45))
elif h == 2:
z = -1 * np.cos(45 * np.pi / 180)
angles.append(str(i) + "_" + str(315))
locations.append(
[np.cos(i * np.pi / 180),
np.sin(i * np.pi / 180), z])
locations.append([0, 0, 1])
angles.append(str(0) + "_" + str(90))
locations.append([0, 0, -1])
angles.append(str(0) + "_" + str(270))
for i in range(0, len(locations)):
locations[i] = list(locations[i] / np.linalg.norm(locations[i]))
####################################################################################################################
# This works when you do not want to link two proteins but instead you want to place one around the other and check
# protein-protein interaction's binding affinity.
if join_o.lower() == 'no' and link_o.lower() == 'x':
ids = {}
for i in string.ascii_uppercase:
ids[i] = False
# First We used model 0 of structure 1 and turn ids for chains to True identifier to True.
for i in s1[0]:
ids[i.id] = True
# Now we go through structure 2 and if there are any chains with the same id as those found in structure 1,
# we will change the chain ids to something else becase if chain ids are repeated in the same structure
# it will be consider one chain when they are actually separated.
for i in s2[0]:
if ids[i.id]:
id_found = False
for j in string.ascii_uppercase:
if not ids[j]:
i.id = j
ids[j] = True
# ID founds means id is unused so far and will be reserved and changed on S2
id_found = True
break
# If the number of chains and identifiers exceeds letters in the alphabet,
# It is necessary to modify the code. Until then, let's check this won't happen by exiting.
if not id_found:
print(
"ERROR: Number of chains in both structures exceeds letters in the alphabet. No ID \
identifiers available. Program will exit without output. Fix the code. June 24, 2016"
)
sys.exit(1)
s3 = copy.deepcopy(s1)
s3.id = 'Ensamble'
for i in s2.get_chains():
s3[0].add(i)
###############################################################################
structure_id = 0
for i in locations:
if structure_id == 0:
ccc = md.ChargeCalculator(params)
cm = ccc.calculate_center_of_charge(s2)
cm = cm / np.linalg.norm(
cm) # ix orientation of LZ here only the first time.
RM = rig.alignVectors(i, cm)
for j in s2.get_atoms():
v2 = [j.get_coord()[0], j.get_coord()[1], j.get_coord()[2]]
jj = np.dot(v2, RM)
j.set_coord(jj)
last_direction = i
# After aligning along centerofcharge/dipolemoment, the structure
# is flipped to have the cterm closest to insulin. This only works with
# LZ because it is a homodimer with both helices aligned in parallel.
# For any other structure, this might not work.
# TODO This Vector needs to be picked by the user from information from the pdb_cif.py --summary
#
# For some reason after modifications to flower the next comented out code in this if step does not seem necessery
#m = rotaxis(np.pi, Vector(0, 1, 0))
#for j2 in s2.get_atoms():
# v2 = Vector([j2.get_coord()[0],j2.get_coord()[1],j2.get_coord()[2]])
# v3 = v2.left_multiply(m)
# j2.set_coord(v3.get_array())
else:
# FIX: Something is wrong with rig.alignVectors([0,0,1],[0,0,-1]) it could just be a trigonometry case that
# gives some sort of singularity. The following code inside the if statement patch just avoids the problem
# but does not explain it. Find out.
if i == [0.0, 0.0, -1.0] and last_direction == [0.0, 0.0, 1.0]:
RM = rig.alignVectors([0.0, 1.0, 1.0], [0.0, 0.0, 1.0])
for j in s2.get_atoms():
v2 = [j.get_coord()[0], j.get_coord()[1], j.get_coord()[2]]
jj = np.dot(v2, RM)
j.set_coord(jj)
last_direction = [0.0, 1.0, 1.0]
RM = rig.alignVectors(i, last_direction)
for j in s2.get_atoms():
v2 = [j.get_coord()[0], j.get_coord()[1], j.get_coord()[2]]
jj = np.dot(v2, RM)
j.set_coord(jj)
last_direction = i
# This 45 changes to explore different distaces between insulin and LZ
ii = [k * distance for k in i]
rig.translate_molecule(s2, modelS2, ii)
###
# We want to join the two structures into one structure, with one model
# and the chains of structure 1 and 2 joined and named consistently.
if join_o.lower() == 'no' and link_o.lower() == 'x':
pass
else:
s3 = struct.StructureBuilder.Structure('newrot')
s3.add(struct.Model.Model(0))
lnk_label = join_chains(s1, s2, s3, join_o, link_o)
io = PDBIO()
io.set_structure(s3)
io.save('s' + '_' + angles[structure_id] + "_" + str(structure_id) +
lnk_label + '.pdb')
#io.save(directory+options.id+'_'+angles[structure_id]+"_"+str(structure_id)+lnk_label+'.pdb')
structure_id = structure_id + 1
rig.translate_molecule(s2, modelS2,
rig.center_molecule(cmc.get_center_of_mass(s2)))
def main():
usage = "usage: %prog [options] arg"
d = "This program reads a CSV file that has been generated by Super_Structure.\
The file corresponds to a Super Structure of a Protein. \
Multiple residues can be added at the time, No terminal will be added..\n \
This program can only add residues or terminals that are in the parameter file."
opt_parser = optparse.OptionParser(usage, description=d)
opt_parser.add_option(
"--apn",
type="str",
help="Enter Instruction for where to append residues in hard '\"'\n \
quotes. Place: Amino Acid Number, Entity ID, Chain ID and \n \
the direction to add residues separated by comas. Add. The \
direction to add residues is either Ndir or Cdir. This means \
that if a residue is added in residue 10, it could be toward \
the N or C terminal. This is important so that the program \
knows if the new residue is placed before or after the residue.\
Example \"1,1,A,Ndir\" or \"20,2,A,Cdir\". \n \
Chain ID, amino acid or terminal name are not case sensitive \
and do not need to go in quotes.\n"
)
opt_parser.add_option(
"-r",
"--res",
type="str",
help="Enter list of amino acids to be added in hard quotes.'\"'\n\
Example: \"ALA,VAL,ASP,ASN,GLU\"."
)
opt_parser.add_option("--inp",
type="str",
help="Path to CSV file for adding residue.")
opt_parser.add_option(
"--out",
type="str",
help="Path and name to CSV and PDB outputs with added residues.")
opt_parser.add_option("--pep", type="str", help="Path to peptide file.")
opt_parser.add_option("--par",
type="str",
help="Path to Charmm parameter folder.")
options, args = opt_parser.parse_args()
if not os.path.exists(options.inp):
print "Error: File path Super Structure CSV file does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
########################## Init Setup #####################################
# Comment out the next four lines to test in Spyder.
directory, filename = os.path.split(options.inp)
params = CP.read_charmm_FF(options.par)
insulin = SS.Super_Structure(params, options.inp, 'add_linker')
parse_list = options.apn.split(',')
if options.res.find(',') == -1:
aa_add = [i for i in options.res]
aa_add = [ut.utilities.residueDict1_1[i] for i in aa_add]
else:
aa_add = options.res.split(',')
parser2 = PDBParser()
pep_file = parser2.get_structure('Peptides', options.pep)
# Uncomment the next four lines to test
#file_path = '/home/noel/Projects/Protein_design/EntropyMaxima/examples/Linker_minimization/2hiu.csv'
#insulin = SS.Super_Structure(params, file_path,'add_linker')
#parse_list = "1,1,A,Ndir".split(',')
#aa_add = "ALA".split(',')
###############################################
insulin.build_pep_and_anchers(pep_file)
############### Begin processing parse_list and aa_add ####################
message = ''
print(parse_list, len(parse_list))
if len(parse_list) == 4 and len(aa_add) > 0:
aaid_add = int(parse_list[0])
ent_id_add = int(parse_list[1])
chain_add = str(parse_list[2]).upper()
term_dir = str(parse_list[3])
# So far this only works with natural aminoacids and ACE and CTER
if term_dir in ['Ndir', 'Cdir']:
message += 'Adding residues ' + str(
aa_add) + ' in th ' + term_dir + ' at amino acid ' + str(
aaid_add) + ', ' + 'entity '
message += str(ent_id_add) + ' and direction ' + term_dir + '.'
print(message)
# TODO: counting atoms do not seem necessary. Consider deleting.
#count_atoms_added = 0
#for i in aa_add:
# for j in insulin.params.AA[i].atoms:
# for k in j:
# count_atoms_added += 1
#count_aa_added = len(aa_add)
###################################################################
# So we now create the link dataframe and follow the prosses in
# Super_Structures to populate its fields.
link = pd.DataFrame()
aa = []
aaid = []
entity_id = []
chain_id = []
atmtyp1 = []
atmtyp2 = []
charg = []
component = []
snum = 1
for res in aa_add:
chrm = res
pdbx = res
if chrm in insulin.params.AA:
comp = 1
for k in insulin.params.AA[chrm].atoms:
for l in k:
aa.append(pdbx)
aaid.append(snum)
entity_id.append(ent_id_add)
chain_id.append(chain_add)
atmtyp1.append(insulin.corrections(chrm, l))
atmtyp2.append(insulin.params.AA[chrm].atom_type[
insulin.corrections(chrm, l)])
charg.append(insulin.params.AA[chrm].atom_chrg[
insulin.corrections(chrm, l)])
if comp == 1:
component.append('AMINO')
else:
if l in ['C', 'O']:
component.append('CARBO')
else:
component.append(('SIDE' + str(comp)))
comp += 1
snum += 1
else:
print('Warning: Amino Acid identifier', chrm,
' is not found in parameters.')
sys.exit(1)
link['aa'] = pd.Series(aa)
link['aaid'] = pd.Series(aaid)
link['ent_id'] = pd.Series(entity_id)
link['chain'] = pd.Series(chain_id)
link['atmtyp1'] = pd.Series(atmtyp1)
link['atmtyp2'] = pd.Series(atmtyp2)
link['component'] = pd.Series(component)
link['charg'] = pd.Series(charg)
###########################################################################
# Add atomtyp, masses and atmNumber to each atom type
mass = []
atmNum = []
atmtyp3 = []
epsilon = []
rmin_half = []
atminfo = []
aainfo = []
for i in link['atmtyp2']:
atmNum.append(params.am.MASS[i][0])
mass.append(params.am.MASS[i][1])
atmtyp3.append(params.am.MASS[i][2])
epsilon.append(params.NONBONDED[i][1])
rmin_half.append(params.NONBONDED[i][2])
atminfo.append(True)
aainfo.append(False)
link['epsilon'] = pd.Series(epsilon)
link['rmin_half'] = pd.Series(rmin_half)
link['atmtyp3'] = pd.Series(atmtyp3)
link['mass'] = pd.Series(mass)
link['atmNum'] = pd.Series(atmNum)
###########################################################################
# DF Type correction.
link['aaid'] = link['aaid'].apply(int)
link['ent_id'] = link['ent_id'].apply(int)
link['mass'] = link['mass'].apply(float)
link['epsilon'] = link['epsilon'].apply(float)
link['rmin_half'] = link['rmin_half'].apply(float)
link['atmNum'] = link['atmNum'].apply(int)
# We now fill out the number of columns in the DataFrame with nan
for i in insulin.Full_Structure.columns:
if i not in list(link.columns):
if i[0:6] == 'aainfo':
link[i] = pd.Series(aainfo)
elif i[0:7] == 'atminfo':
link[i] = pd.Series(atminfo)
else:
link[i] = pd.Series(
[float('nan') for j in range(len(link))])
if term_dir == 'Ndir':
beg_insert = min(insulin.Full_Structure.index[(insulin.Full_Structure.aaid == aaid_add) &\
(insulin.Full_Structure.ent_id == ent_id_add) &\
(insulin.Full_Structure.chain == chain_add)])
end_insert = beg_insert + link.shape[0]
elif term_dir == 'Cdir':
print(
'WARNING: The code has not been design and tested for insertions in the CTER.'
)
print('Exiting the program without finishing.')
sys.exit(1)
else:
print(
'ERROR: wrong terminal to insert link. Ndir and Cdir are the only choices. Exiting now.'
)
sys.exit(1)
joint_df = pd.DataFrame(columns=link.columns)
count = 0
insert = True
# When links are added , aaid needs to be fixed to reflect added residues
aaid_offset = 0
for i in insulin.Full_Structure.index:
if (i >= beg_insert) and (i < end_insert):
if insert:
for j in link.index:
joint_df.loc[count] = link.loc[j]
joint_df.loc[count, 'aaid'] = joint_df.loc[
count, 'aaid'] + aaid_offset
current_aaid = link.loc[j, 'aaid']
count += 1
insert = False
aaid_offset = aaid_offset + current_aaid
joint_df.loc[count] = insulin.Full_Structure.loc[i]
# So that only residues after the added link get increased in the given ent_id and chain
# Any other entity or chain in the molecules is not fixed.
if (joint_df.loc[count, 'ent_id'] == ent_id_add) & (
joint_df.loc[count, 'chain'] == chain_add):
joint_df.loc[count,
'aaid'] = joint_df.loc[count,
'aaid'] + aaid_offset
count += 1
# After adding residues, it all gets copied back to original dataframe.
for i in joint_df.index:
insulin.Full_Structure.loc[i] = joint_df.loc[i]
# The way to get number of models is very specific to the way this program
# stores data in DataFrame. Be careful if the data frame column structure changes.
# TODO: missing atom coordinates are added manually. It needs to be automated more.
num_models = len(
range(((insulin.Full_Structure.shape[1] - 20) / 5))) + 1
for i in range(1, num_models + 1):
for j in range(len(aa_add), 0, -1):
insulin.fit_coordinates(term_dir, j, ent_id_add, chain_add,
str(i), aa_add[j - 1])
# NOTE: insulin.models are not in the Super Structure Class, but it is added here.
# This works, but it does not seem the best way to do it. should models be a field of super
# structures and be initialized there?
insulin.models = [str(i) for i in range(1, num_models + 1)]
################ Write to outputs ####################
file_name = os.path.basename(options.out).split('.')[0]
dir_path = os.path.dirname(options.out)
insulin.write_csv(os.path.dirname(options.out), file_name)
IO.write_pdb(insulin, dir_path, file_name, 'all')
else:
print("ERROR: only two directions to add residues, Ndir and Cdir.")
print(" The entries are not case sensitive.")
else:
message += 'The number of entries in the instruction field, followed by -a or --apn, is not right.\n'
message += 'Type -h or --help for instructions\n'
print(message)
def main():
usage = "usage: %prog [options]"
d = "It takes two pdbs and aligns the structures in different \
orientations. A center-pdb is placed at (0,0,0),and a rotate-pdb \
is place around at given angle intervals and distances. \
The program assumes that the rotated protein's N-terminal will \
become the N-Terminal when joined to the center protein. \
WARNING: The programs needs to be extended to allow the reading \
of the dipole vectors of the two proteins, and align them in any \
way the user wants. As is, it centers the center-protein in whatever \
orientation it is, and it alings the center of charge vector with \
vector <0, 1, 0>. From that starting point, it rotates rotate-protein \
by the angle given as input."
option_parser = optparse.OptionParser(usage, description=d)
option_parser.add_option("--center", type="str", \
help="Path to pdb structure to place at (0,0,0) \
according to its center of mass." )
option_parser.add_option("--rotate", type="str", \
help="Path to pdb structure to rotated around \
center structure at X degree intervals. Angle \
intervals must be multiple of 360." )
option_parser.add_option("--angle", type="int", \
help="Angle intervals used to place rotated \
structured around center one." )
option_parser.add_option("--distance", type="float",\
help="Center of mass distance between rotated and \
centered structures." )
option_parser.add_option('--id',
type="str",
action="store",
default='s',
help="A prefix id to identify the \
output structures. Default is s.")
option_parser.add_option('--map',
type="str",
action="store",
default='no',
help="Optional arg if not present it \
is 'no', and it will add rotate with different chain \
identifiers. if it is 'yes' then it expects --link \
to be something different from 'X' and in the right format."
)
option_parser.add_option('--link',
type="str",
action="store",
default="X",
help="Optional arg if not present it \
is 'X', and it will add \'rotate\' structure with different chain \
identifiers. If it is 'yes', it will add the \'rotate\' structure \
with the same identifiers as \'center\' structure in the following \
format: A:A,B:B or chain A in \'rotate\' will link with A in \'center\' \
and B to B respectively. When both proteins have two chains, the only \
other option is A:B,B:A. There can be as many chains linked separated \
by commas but be careful because the program does not checks the \
physicality of this connections.")
option_parser.add_option("--par",
type="str",
help="Path to charmm parameters folder.")
options, args = option_parser.parse_args()
if not os.path.exists(options.center):
print "Error: File path for molecule to be centered does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
if not os.path.exists(options.rotate):
print "Error: File path for molecule to be rotated does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
if options.map.lower() == 'yes' and options.link.lower() == 'x':
print("Error: when option map is equal to yes, option link must be")
print(
" a series of chains identifiers for association as described"
)
print(" in the link paramater help.")
sys.exit(1)
pdb_parser = PDBParser(QUIET=True)
Angle = options.angle
distance = options.distance
#directory = os.path.dirname(options.center)
filepath1 = options.center
filepath2 = options.rotate
map_o = options.map
lnk_o = options.link
param_path = options.par
###########################################################################
# Uncomment to test from spyder IDE
#pdb_parser = PDBParser(QUIET = True)
#Angle = 45
#distance = 45
##file_name = os.path.basename(options.out).split('.')[0]
#directory = "/home/noel/Projects/Protein_design/ccl_lectures/Lecture_4/"
#filepath1 = directory+'2hiu_1rr.pdb'
#filepath2 = directory+'2zta_1rr.pdb'
#param_path = "/home/noel/Projects/Protein_design/EntropyMaxima/params/charmm27.ff/"
#map_o = "yes"
#lnk_o = "A:A,B:B"
####################################################################################################################
# Process strig that the determines how the centered and rotated structures will be connected.
lnk_o = lnk_o.split(',')
lnk_o = [i.split(':') for i in lnk_o]
lnk_label = ''
for i in lnk_o:
lnk_label += '_'
for j in i:
lnk_label += j.lower()
params = CP.read_charmm_FF(param_path)
cmc = md.CenterOfMassCalculator(params)
rig = MRM.Molecular_Rigid_Manipulation(param_path)
####################################################################################################################
# Check that the structures only have one model, and Place the structures' center of mass at (0,0,0) to give an idea
# of their location in the cartesian coordinate system
s1 = pdb_parser.get_structure('Centered', filepath1)
countS1 = 0
modelS1 = -1
for i in s1.get_models():
countS1 += 1
modelS1 = i.id
if countS1 != 1:
print(
"ERROR: Number of models cannot be different from 1. Models found:"
+ str(countS1))
print(" Make sure Centered PDBs have only one model.")
sys.exit(1)
rig.translate_molecule(s1, modelS1,
rig.center_molecule(cmc.get_center_of_mass(s1)))
s2 = pdb_parser.get_structure('Rotated', filepath2)
countS2 = 0
modelS2 = -1
for i in s2.get_models():
countS2 += 1
modelS2 = i.id
if countS2 != 1:
print(
"ERROR: Number of models cannot be different from 1. Models found:"
+ str(countS2))
print(" Make sure Rotated PDBs have only one model.")
sys.exit(1)
rig.translate_molecule(s2, modelS2,
rig.center_molecule(cmc.get_center_of_mass(s2)))
####################################################################################################################
# TODO: This works only for angles between 0 and 90 not including 0, and 90 and will generate angles in all 8
# quadrants of the cartesian coordinate system (I do not see why using an angle other than 45 for now.)
# TODO: quaternions might work better.
# The location list has a list of normalized vectors releative to (0,0,0) that will be use to place s2's in the
# right orientation relative to s1
locations = []
angles = []
for h in range(0, 3):
for i in range(0, 360, Angle):
for j in range(0, 90 / Angle - 1):
if h == 0:
z = 0
angles.append(str(i) + "_" + str(0))
elif h == 1:
z = np.cos(45 * np.pi / 180)
angles.append(str(i) + "_" + str(45))
elif h == 2:
z = -1 * np.cos(45 * np.pi / 180)
angles.append(str(i) + "_" + str(315))
locations.append(
[np.cos(i * np.pi / 180),
np.sin(i * np.pi / 180), z])
locations.append([0, 0, 1])
angles.append(str(0) + "_" + str(90))
locations.append([0, 0, -1])
angles.append(str(0) + "_" + str(270))
for i in range(0, len(locations)):
locations[i] = list(locations[i] / np.linalg.norm(locations[i]))
####################################################################################################################
# This works when you do not want to link two proteins but instead you want to place one around the other and check
# protein-protein interaction's binding affinity.
if map_o.lower() == 'no' and options.link.lower() == 'x':
ids = {}
for i in string.ascii_uppercase:
ids[i] = False
# First We used model 0 of structure 1 and turn ids for chains to True identifier to True.
for i in s1[0]:
ids[i.id] = True
# Now we go through structure 2 and if there are any chains with the same id as those found in structure 1,
# we will change the chain ids to something else becase if chain id is repeated in the same structure
# it will be consider one chain when they are actually separated.
for i in s2[0]:
if ids[i.id]:
id_found = False
for j in string.ascii_uppercase:
if not ids[j]:
i.id = j
ids[j] = True
# ID founds means id is unused so far and will be reserved and changed on S2
id_found = True
break
# If the number of chains and identifiers exceeds letters in the alphabet,
# It is necessary to modify the code. Until then, let's check this won't happen by exiting.
if not id_found:
print(
"ERROR: Number of chains in both structures exceeds letters in the alphabet. No ID \
identifiers available. Program will exit without output. Fix the code. June 24, 2016"
)
sys.exit(1)
s3 = copy.deepcopy(s1)
s3.id = 'Ensamble'
for i in s2.get_chains():
s3[0].add(i)
else:
chain_info_s1 = get_chains_info(s1)
chain_info_s2 = get_chains_info(s2)
###############################################################################
structure_id = 0
for i in locations:
if structure_id == 0:
ccc = md.ChargeCalculator(params)
cm = ccc.calculate_center_of_charge(s2)
cm = cm / np.linalg.norm(
cm) # ix orientation of LZ here only the first time.
RM = rig.alignVectors(i, cm)
for j in s2.get_atoms():
v2 = [j.get_coord()[0], j.get_coord()[1], j.get_coord()[2]]
jj = np.dot(v2, RM)
j.set_coord(jj)
last_direction = i
# After aligning along centerofcharge/dipolemoment, the structure
# is flipped to have the cterm closest to insulin. This only works with
# LZ because it is a homodimer with both helices aligned in parallel.
# For any other structure, this might not work.
# TODO This Vector needs to be picked by the user from information from the pdb_cif.py --summary
#
# For some reason after modifications to flower the next comented out code in this if step does not seem necessery
#m = rotaxis(np.pi, Vector(0, 1, 0))
#for j2 in s2.get_atoms():
# v2 = Vector([j2.get_coord()[0],j2.get_coord()[1],j2.get_coord()[2]])
# v3 = v2.left_multiply(m)
# j2.set_coord(v3.get_array())
else:
# FIX: Something is wrong with rig.alignVectors([0,0,1],[0,0,-1]) it could just be a trigonometry case that
# gives some sort of singularity. The following code inside the if statement patch just avoids the problem
# but does not explain it. Find out.
if i == [0.0, 0.0, -1.0] and last_direction == [0.0, 0.0, 1.0]:
RM = rig.alignVectors([0.0, 1.0, 1.0], [0.0, 0.0, 1.0])
for j in s2.get_atoms():
v2 = [j.get_coord()[0], j.get_coord()[1], j.get_coord()[2]]
jj = np.dot(v2, RM)
j.set_coord(jj)
last_direction = [0.0, 1.0, 1.0]
RM = rig.alignVectors(i, last_direction)
for j in s2.get_atoms():
v2 = [j.get_coord()[0], j.get_coord()[1], j.get_coord()[2]]
jj = np.dot(v2, RM)
j.set_coord(jj)
last_direction = i
# This 45 changes to explore different distaces between insulin and LZ
ii = [k * distance for k in i]
rig.translate_molecule(s2, modelS2, ii)
# We want to join the two structures into one structure, with one model
# and the chains of structure 1 and 2 joined and name consistently.
if map_o.lower() == 'no' and lnk_o.lower() == 'x':
pass
else:
s3 = struct.StructureBuilder.Structure('newrot')
s3.add(struct.Model.Model(0))
join_range1 = [(chain_info_s2[lnk_o[0][0]]['min_res'],
chain_info_s2[lnk_o[0][0]]['max_res']),
(chain_info_s1[lnk_o[0][1]]['min_res'],
chain_info_s1[lnk_o[0][1]]['max_res'])]
join_chains(s3, s2, s1, lnk_o[0], join_range1)
join_range2 = [(chain_info_s2[lnk_o[1][0]]['min_res'],
chain_info_s2[lnk_o[1][0]]['max_res']),
(chain_info_s1[lnk_o[1][1]]['min_res'],
chain_info_s1[lnk_o[1][1]]['max_res'])]
join_chains(s3, s2, s1, lnk_o[1], join_range2)
io = PDBIO()
io.set_structure(s3)
io.save('s' + '_' + angles[structure_id] + "_" + str(structure_id) +
lnk_label + '.pdb')
#io.save(directory+options.id+'_'+angles[structure_id]+"_"+str(structure_id)+lnk_label+'.pdb')
structure_id = structure_id + 1
rig.translate_molecule(s2, modelS2,
rig.center_molecule(cmc.get_center_of_mass(s2)))
def main():
usage = "usage: %prog [options] arg"
d = "This program reads a CIF file and checks that all residues in the file\
are found in the CHARMM top_27 parameters. Residues found, but missing in \
the structure, are added to the structure. The full structure is outputed \
to a CSV file where Charmm, CIF and additional information is stored. \
Added residues are copied from a peptide structure with all amino acids \
present in the local CHARMM parameters files with fixed dihedral angles. \
Info in the CSV file should be all there is to explore the conformational \
space of added atoms."
opt_parser = optparse.OptionParser(usage, description=d)
group = optparse.OptionGroup(
opt_parser,
"Generates CSV and PDB files for each model from a CIF file.")
group.add_option("--fromcif",
action="store_true",
help="Flag to generate a CSV frile from a CIF file.")
group.add_option("-i", "--cif", type="str", help="Path to input cif file.")
group.add_option("-o", "--out1", type="str", help="Path to output csv.")
group.add_option("-p",
"--pep",
type="str",
help="Path to CHARMM peptide file.")
opt_parser.add_option_group(group)
group = optparse.OptionGroup(
opt_parser, "Generates a CSV file from CRD and PSF files.")
group.add_option(
"--frompsfcrd",
action="store_true",
help="Flag to generates a CSV frile from a CRD and PSF file.")
group.add_option("-f",
"--psf",
type="str",
help="Path to input PSF file in XPLOR format.")
group.add_option("-d", "--crd", type="str", help="Path to input CRD file.")
opt_parser.add_option_group(group)
options, args = opt_parser.parse_args()
############################################ Options Entered ##########################################################
if options.fromcif:
if options.frompsfcrd:
opt_parser.error(
"Two option flags can't be selected at the same time. Enter -h for help."
)
########################################################################################################################
if options.fromcif:
if not os.path.exists(options.cif):
print "Error: File path for input file does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
params = CP.read_charmm_FF()
parser2 = PDBParser(QUIET=True)
pep_file_path = pkg_resources.resource_filename(
'em', 'params/' + 'peptides.pdb')
p1 = parser2.get_structure('Peptides', pep_file_path)
###########################################################################
# The peptide construct is build with charmm so corrections for some atom
# names to PDB/Databank atom types is needed.
# TODO: this might not be necessary as the correction and inv_correction dictionary in Super Structure takes care of it.
# Check before removing the correction here.
for i in p1.get_models():
for j in i.get_chains():
for k in j.get_residues():
for l in k.get_atom():
if k.get_resname() == 'ILE' and l.get_id() == 'CD':
l.name = 'CD1'
l.id = 'CD1'
###########################################################################
# Create Super Structure
myCIF = SS.Super_Structure(params, options.cif, 'setup')
myCIF.build_pep_and_anchers(p1)
myCIF.read_dict_into_dataframes()
myCIF.check_models()
myCIF.create_super_structure_df()
###########################################################################
# Find missing residues to add to the Super Structure. Missing residues
# are group in lists of contiguous residues and aded to another list.
myCIF.build_missing_aa()
file_name = os.path.basename(options.cif).split('.')[0]
myCIF.write_csv('', file_name)
#outPDB = IO.pdb()
IO.write_pdb(myCIF, '', file_name, 'all')
if options.frompsfcrd:
if not os.path.exists(options.psf):
print "Error: File path for PSF file does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
if not os.path.exists(options.crd):
print "Error: File path for CRD file does not exist."
print("Type -h or --help for description and options.")
sys.exit(1)
directory, filename = os.path.split(options.crd)
crd_file = IO.crd(options.crd)
psf_file = IO.psf(options.psf)
file_name = filename.split('.')[0]
################################################################################################################
###################### After reading files, Generate and Index a Super Structure ##############################
params = CP.read_charmm_FF()
myCSV = SS.Super_Structure(params, directory, 'charmm_input')
# At this point, a XPLOR psf could only have been creted from a complete structure, so no worries of gaps.
myCSV.create_super_structure_df_from_CRD_PSF(crd_file, psf_file)
myCSV.write_csv(directory, file_name)