def main():
args = parse_arguments()
file_list = []
for item in args.infile:
if '*' in item:
file_list += list(glob.glob(item))
else:
file_list.append(item)
if args.snapshots:
snapshot_indices = read_snapshot_indices(args.snapshots)
assert len(file_list) == 1, ("Should only run 1 file at a time with "
"particular snapshots specified!")
else:
snapshot_indices = None
for mae_file in file_list:
s = load_structure_reader(mae_file)
lig_out = '%d_lig.maegz'
rcp_out = '%d_rcp.maegz'
if args.lig_fmt:
assert re.search(r'.*%d.*\.maegz', args.lig_fmt)
lig_out = args.lig_fmt
if args.rcp_fmt:
assert re.search(r'.*%d.*\.maegz', args.rcp_fmt)
rcp_out = args.rcp_fmt
rcp_asl = args.rcp_asl
lig_asl = args.lig_asl
j = 1
for i, lig_st in enumerate(s):
if snapshot_indices and i not in snapshot_indices:
continue
rcp_st = lig_st.copy()
lig_trj = lig_out % j
rcp_trj = rcp_out % j
# Select and delete lig from rcp structure
lig_selection = structureutil.evaluate_asl(rcp_st, lig_asl)
rcp_st.deleteAtoms(lig_selection)
rcp_st.append(rcp_trj)
# Select and delete rcp from lig structure
rcp_selection = structureutil.evaluate_asl(lig_st, rcp_asl)
lig_st.deleteAtoms(rcp_selection)
lig_st.append(lig_trj)
j += 1
def writeRecStructureFile(self):
receptor_file = self.keywords.get('RECEPTOR_FILE')
if not receptor_file:
msg = "bedam_prep: No receptor file specified in the input file"
self.exit(msg)
if not os.path.exists(receptor_file):
msg = 'File does not exist: %s' % receptor_file
self.exit(msg)
receptor_asl = self.keywords.get('REST_RECEPTOR_ASL')
try:
st = structure.StructureReader(receptor_file).next()
except:
print "Cannot open Maestro structure file %s" % receptor_file
sys.exit(-1)
atoms = []
if receptor_asl:
atoms = structureutil.evaluate_asl(st, receptor_asl)
# all atoms free
for atom in st.atom:
atom.property['i_i_constraint'] = 0
# buffer the ones in ASL
for iat in atoms:
st.atom[iat].property['i_i_constraint'] = 2
receptor_file_restr = self.jobname + '_rcpt_restr' + '.maegz'
st.write(receptor_file_restr)
self.receptor_file_restr = receptor_file_restr
self.ligand_file_restr = self.ligidxfile
def __init__(self, receptor):
self.POLAR_RESIDUES = ["ARG","ASP","GLU","HIS","ASN","GLN","LYS","SER","THR","ARN","ASH","GLH","HID","HIE","LYN"]
self.HYDROPHOBIC_RESIDUES =["PHE","LEU","ILE","TYR","TRP","VAL","MET","PRO","CYS","ALA","CYX"]
self.AROMATIC_RESIDUES = ["PHE","TYR","TRP","TYO"]
self.CHARGED_RESIDUES = ["ARG","ASP","GLU","LYS","HIP","CYT","SRO","TYO","THO"]
# Bit positions for each interaction:
self.bit_pos = {'CONTACT' : 0, 'BACKBONE' : 1, 'SIDECHAIN' : 2, 'POLAR' : 3, 'HYDROPHOBIC' : 4, 'H_ACCEPTOR' : 5, 'H_DONOR' : 6, 'AROMATIC' : 7, 'CHARGED' : 8}
self.active_bits = ['CONTACT', 'BACKBONE', 'SIDECHAIN', 'POLAR', 'HYDROPHOBIC', 'H_ACCEPTOR', 'H_DONOR', 'AROMATIC', 'CHARGED']
self.max_bits_per_residue = 9
self.used_bits_number = len(self.active_bits)
self.sifts = {}
self.receptor_st = receptor
backbone_list = structureutil.evaluate_asl( self.receptor_st,
"backbone")
self.backbone_set = set(backbone_list)
"""
structureutil.evaluate_asl
Search for substructures matching the ASL (Atom Specification Language)
the atom index is returned
"""
polar_list = structureutil.evaluate_asl( self.receptor_st,
"res. " + ', '.join(self.POLAR_RESIDUES ))
self.polar_set = set( polar_list )
hydrophobic_list = structureutil.evaluate_asl( self.receptor_st,
"res. " + ', '.join(self.HYDROPHOBIC_RESIDUES))
self.hydrophobic_set = set( hydrophobic_list )
aromatic_list = structureutil.evaluate_asl( self.receptor_st,
"res. " + ', '.join(self.AROMATIC_RESIDUES))
self.aromatic_set = set(aromatic_list)
charged_list = structureutil.evaluate_asl ( self.receptor_st,
"res. " + ', '.join(self.CHARGED_RESIDUES))
self.charged_set = set( charged_list )
def find_close_residues(self, ligand, cutoff=4.0):
lig_name = ligand.title
d_ = defaultdict(list)
for atom in ligand.atom:
# the distance to original point?
lig_dist = self.measure_dist([atom.xyz])
# range should refer to the node level in the red black tree
close_atoms = self.distances.find_nodes_in_range(lig_dist - cutoff, lig_dist + cutoff)
for close in close_atoms:
atom_atom_dist = self.measure_dist([atom.xyz], close.data.coords)
if atom_atom_dist <= cutoff:
self.update_min_max(close.data.res_num)
d_[close.data.atom].append((ligand.title, atom))
# Just take into account the special case, H_DONOR and H_RECEPTOR
self.fingerprints.add_sift_chunk_special_case(close.data.atom, atom, ligand.title, atom_atom_dist)
max_atom_nums = max([len(atom_list) for atom_list in d_.values()])
# modified definition of interaction
# print "max in-range atoms count ",max_atom_nums
strong_c = 0
exists_c = 0
for close_atom, lig_names in d_.items():
# print lig_names
if len(lig_names) >= 3:
# print "mark as strong"
self.fingerprints.add_sift_chunk_ordinary(
close_atom, lig_names[0][0], "strong"
) # we have only on ligand per complex
strong_c += 1
else:
# print "mark as exists"
self.fingerprints.add_sift_chunk_ordinary(close_atom, lig_names[0][0], "exists")
exists_c += 1
print sorted([len(lig_names) for lig_names in d_.values()])
####### start of environmental descriptors ########
# print "strong count",strong_c ,"exists count",exists_c
# addition of surrounding environment
POLAR_RESIDUES = [
"ARG",
"ASP",
"GLU",
"HIS",
"ASN",
"GLN",
"LYS",
"SER",
"THR",
"ARN",
"ASH",
"GLH",
"HID",
"HIE",
"LYN",
]
HYDROPHOBIC_RESIDUES = ["PHE", "LEU", "ILE", "TYR", "TRP", "VAL", "MET", "PRO", "CYS", "ALA", "CYX"]
AROMATIC_RESIDUES = ["PHE", "TYR", "TRP", "TYO"]
CHARGED_RESIDUES = ["ARG", "ASP", "GLU", "LYS", "HIP", "CYT", "SRO", "TYO", "THO"]
polar_list = structureutil.evaluate_asl(ligand, "res. " + ", ".join(POLAR_RESIDUES))
polar_set = set(polar_list)
hydrophobic_list = structureutil.evaluate_asl(ligand, "res. " + ", ".join(HYDROPHOBIC_RESIDUES))
hydrophobic_set = set(hydrophobic_list)
aromatic_list = structureutil.evaluate_asl(ligand, "res. " + ", ".join(AROMATIC_RESIDUES))
aromatic_set = set(aromatic_list)
charged_list = structureutil.evaluate_asl(ligand, "res. " + ", ".join(CHARGED_RESIDUES))
charged_set = set(charged_list)
# print polar_set, hydrophobic_set ,aromatic_set , charged_set
env_d_ = dict()
for close_atom, lig_names in d_.items():
for lig_atom in [l[1] for l in lig_names]:
lig_atom = int(lig_atom)
if close_atom not in env_d_.keys():
env_d_[close_atom] = defaultdict(int)
if lig_atom in polar_set:
# print "it is polar"
env_d_[close_atom]["ENV_POLAR"] += 1
elif lig_atom in aromatic_set:
# print "it is aromatic"
env_d_[close_atom]["ENV_AROMATIC"] += 1
elif lig_atom in hydrophobic_set:
# print "it is h"
env_d_[close_atom]["ENV_HYDROPHOBIC"] += 1
elif lig_atom in charged_set:
# print "it is c"
env_d_[close_atom]["ENV_CHARGED"] += 1
atom_nums = list()
for a, fp in env_d_.items():
atom_nums += fp.values()
max_atom_nums = max(atom_nums)
env_stat = {"p_list": [0], "a_list": [0], "h_list": [0], "c_list": [0]}
env_stat["p_list"] += [fp["ENV_POLAR"] for a, fp in env_d_.items() if fp["ENV_POLAR"] != 0]
env_stat["a_list"] += [fp["ENV_AROMATIC"] for a, fp in env_d_.items() if fp["ENV_AROMATIC"] != 0]
env_stat["h_list"] += [fp["ENV_HYDROPHOBIC"] for a, fp in env_d_.items() if fp["ENV_HYDROPHOBIC"] != 0]
env_stat["c_list"] += [fp["ENV_CHARGED"] for a, fp in env_d_.items() if fp["ENV_CHARGED"] != 0]
print env_stat
print max(env_stat["p_list"]), max(env_stat["a_list"]), max(env_stat["h_list"]), max(env_stat["c_list"])
strong_c = 0
exists_c = 0
for a, fp in env_d_.items():
for t, c in fp.items():
if c >= 3:
self.fingerprints.add_sift_chunk_env(a, lig_name, t, "strong")
strong_c += 1
else:
self.fingerprints.add_sift_chunk_env(a, lig_name, t, "exists")
exists_c += 1
def __init__(self, receptor):
self.POLAR_RESIDUES = [
"ARG",
"ASP",
"GLU",
"HIS",
"ASN",
"GLN",
"LYS",
"SER",
"THR",
"ARN",
"ASH",
"GLH",
"HID",
"HIE",
"LYN",
]
self.HYDROPHOBIC_RESIDUES = ["PHE", "LEU", "ILE", "TYR", "TRP", "VAL", "MET", "PRO", "CYS", "ALA", "CYX"]
self.AROMATIC_RESIDUES = ["PHE", "TYR", "TRP", "TYO"]
self.CHARGED_RESIDUES = ["ARG", "ASP", "GLU", "LYS", "HIP", "CYT", "SRO", "TYO", "THO"]
# Bit positions for each interaction:
self.bit_pos = {
"CONTACT": 0,
"BACKBONE": 1,
"SIDECHAIN": 2,
"POLAR": 3,
"HYDROPHOBIC": 4,
"H_ACCEPTOR": 5,
"H_DONOR": 6,
"AROMATIC": 7,
"CHARGED": 8,
"ENV_POLAR": 9,
"ENV_HYDROPHOBIC": 10,
"ENV_AROMATIC": 11,
"ENV_CHARGED": 12,
}
self.active_bits = [
"CONTACT",
"BACKBONE",
"SIDECHAIN",
"POLAR",
"HYDROPHOBIC",
"H_ACCEPTOR",
"H_DONOR",
"AROMATIC",
"CHARGED",
"ENV_POLAR",
"ENV_HYDROPHOBIC",
"ENV_AROMATIC",
"ENV_CHARGED",
]
self.max_bits_per_residue = 13
self.used_bits_number = len(self.active_bits)
self.sifts = {}
self.receptor_st = receptor
backbone_list = structureutil.evaluate_asl(self.receptor_st, "backbone")
self.backbone_set = set(backbone_list)
"""
structureutil.evaluate_asl
Search for substructures matching the ASL (Atom Specification Language)
the atom index is returned
"""
polar_list = structureutil.evaluate_asl(self.receptor_st, "res. " + ", ".join(self.POLAR_RESIDUES))
self.polar_set = set(polar_list)
hydrophobic_list = structureutil.evaluate_asl(self.receptor_st, "res. " + ", ".join(self.HYDROPHOBIC_RESIDUES))
self.hydrophobic_set = set(hydrophobic_list)
aromatic_list = structureutil.evaluate_asl(self.receptor_st, "res. " + ", ".join(self.AROMATIC_RESIDUES))
self.aromatic_set = set(aromatic_list)
charged_list = structureutil.evaluate_asl(self.receptor_st, "res. " + ", ".join(self.CHARGED_RESIDUES))
self.charged_set = set(charged_list)
def writeRestraintFile(self):
# check that structure files with internal indexes have been generated
if self.recidxfile is None or self.ligidxfile is None:
msg = "writeRestraintFile: Internal error: structure files not found"
self.exit(msg)
receptor_asl = self.keywords.get('REST_LIGAND_CMRECASL')
if not receptor_asl:
msg = "bedam_prep: No ASL specified for receptor site center"
self.exit(msg)
ligand_asl = self.keywords.get('REST_LIGAND_CMLIGASL')
if not ligand_asl:
ligand_asl = '( all)'
try:
st = structure.StructureReader(self.recidxfile).next()
except:
print "Cannot open Maestro structure file %s" % self.recidxfile
sys.exit(-1)
nrecatoms = len(st.atom)
atoms = structureutil.evaluate_asl(st, receptor_asl)
rec_atoms = [];
for iat in atoms:
rec_atoms.append(st.atom[iat].property['i_i_internal_atom_index'])
#computes receptor CM
cmrx = cmry = cmrz = 0.
for iat in atoms:
cmrx += st.atom[iat].x
cmry += st.atom[iat].y
cmrz += st.atom[iat].z
n = len(atoms)
cmrx /= float(n)
cmry /= float(n)
cmrz /= float(n)
try:
st = structure.StructureReader(self.ligidxfile).next()
except:
print "Cannot open Maestro structure file %s" % self.ligidxfile
sys.exit(-1)
nligatoms = len(st.atom)
atoms = structureutil.evaluate_asl(st, ligand_asl)
lig_atoms = [];
for iat in atoms:
lig_atoms.append(st.atom[iat].property['i_i_internal_atom_index']-nrecatoms)
#computes ligand CM
cmlx = cmly = cmlz = 0.
for iat in atoms:
cmlx += st.atom[iat].x
cmly += st.atom[iat].y
cmlz += st.atom[iat].z
n = len(atoms)
cmlx /= float(n)
cmly /= float(n)
cmlz /= float(n)
#computes and reports the distance btw CM's
d = sqrt((cmlx - cmrx)*(cmlx - cmrx) + (cmly - cmry)*(cmly - cmry) + (cmlz - cmrz)*(cmlz - cmrz))
print "CM-CM distance = %f" % d
self.restraint_file = self.jobname + '_cmrestraint.dat'
f = open(self.restraint_file,"w")
f.write("Receptor\n")
f.write("%d\n" % 1)
f.write("%d\n" % len(rec_atoms) )
for i in rec_atoms:
f.write("%d\n" % i)
f.write("Ligand\n")
f.write("%d\n" % 2)
f.write("%d\n" % len(lig_atoms) )
for i in lig_atoms:
f.write("%d\n" % i)
f.close()
