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] 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 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)