def load_rigimol_inp(filename, oname, _self=cmd):
'''
DESCRIPTION
Load the structures from a RigiMOL "inp" file.
'''
import shlex
from chempy import Atom, Bond, models
parsestate = ''
model = None
state = 0
for line in open(filename):
lex = shlex.shlex(line, posix=True)
lex.whitespace_split = True
lex.quotes = '"'
a = list(lex)
if not a:
continue
if a[0] == '+':
if a[1] == 'NAME':
assert a[2:] == [
'RESI', 'RESN', 'CHAIN', 'SEGI', 'X', 'Y', 'Z', 'B'
]
parsestate = 'ATOMS'
model = models.Indexed()
elif a[1] == 'FROM':
assert a[2:] == ['TO']
parsestate = 'BONDS'
else:
parsestate = ''
elif a[0] == '|':
if parsestate == 'ATOMS':
atom = Atom()
atom.coord = [float(x) for x in a[6:9]]
atom.name = a[1]
atom.resi = a[2]
atom.resn = a[3]
atom.chain = a[4]
atom.segi = a[5]
atom.b = a[9]
atom.symbol = ''
model.add_atom(atom)
elif parsestate == 'BONDS':
bnd = Bond()
bnd.index = [int(a[1]), int(a[2])]
model.add_bond(bnd)
elif a[0] == 'end':
if parsestate in ('ATOMS', 'BONDS'):
state += 1
_self.load_model(model, oname, state=state)
parsestate = ''
def __init__(self,
object,
name='MMTK_model',
configuration=None,
b_values=None):
self.object = object
self.universe = object.universe()
self.name = name
self.model = Indexed()
self.index_map = {}
self.atoms = []
chain_id_number = 0
in_chain = True
for o in object.bondedUnits():
if Utility.isSequenceObject(o):
groups = [(g, g.name) for g in o]
if not in_chain:
chain_id_number = (chain_id_number + 1) % len(
self.chain_ids)
in_chain = True
else:
groups = [(o, o.name)]
in_chain = False
residue_number = 1
for g, g_name in groups:
for a in g.atomList():
atom = Atom()
atom.symbol = a.symbol
atom.name = a.name
atom.resi_number = residue_number
atom.chain = self.chain_ids[chain_id_number]
if b_values is not None:
atom.b = b_values[a]
atom.resn = g_name
self.model.atom.append(atom)
self.index_map[a] = len(self.atoms)
self.atoms.append(a)
residue_number = residue_number + 1
if in_chain:
chain_id_number = (chain_id_number + 1) % len(self.chain_ids)
try:
bonds = o.bonds
except AttributeError:
bonds = []
for b in bonds:
bond = Bond()
bond.index = [self.index_map[b.a1], self.index_map[b.a2]]
self.model.bond.append(bond)
self._setCoordinates(configuration)
def __init__(self, object, name = 'MMTK_model', configuration = None,
b_values = None):
self.object = object
self.universe = object.universe()
self.name = name
self.model = Indexed()
self.index_map = {}
self.atoms = []
chain_id_number = 0
in_chain = True
for o in object.bondedUnits():
if Utility.isSequenceObject(o):
groups = [(g, g.name) for g in o]
if not in_chain:
chain_id_number = (chain_id_number+1) % len(self.chain_ids)
in_chain = True
else:
groups = [(o, o.name)]
in_chain = False
residue_number = 1
for g, g_name in groups:
for a in g.atomList():
atom = Atom()
atom.symbol = a.symbol
atom.name = a.name
atom.resi_number = residue_number
atom.chain = self.chain_ids[chain_id_number]
if b_values is not None:
atom.b = b_values[a]
atom.resn = g_name
self.model.atom.append(atom)
self.index_map[a] = len(self.atoms)
self.atoms.append(a)
residue_number = residue_number + 1
if in_chain:
chain_id_number = (chain_id_number+1) % len(self.chain_ids)
try:
bonds = o.bonds
except AttributeError:
bonds = []
for b in bonds:
bond = Bond()
bond.index = [self.index_map[b.a1], self.index_map[b.a2]]
self.model.bond.append(bond)
self._setCoordinates(configuration)
def fromList(self,list): # currently no handling of conect records
model = Indexed()
# read atoms
cnt = 0
at = None
for rec in list:
if (rec[0:4]=='ATOM') or (rec[0:6]=='HETATM'):
at = Atom()
if rec[0]=='A': at.hetatm=0 # default is 1
at.index = cnt
at.name = string.strip(rec[12:16])
at.alt = string.strip(rec[16:17])
at.resn = string.strip(rec[17:20])
at.chain = string.strip(rec[21:22])
at.resi = string.strip(rec[22:27]) # note: insertion is part of resi
at.resi_number = int(rec[22:26])
at.coord = [float(rec[30:38]),
float(rec[38:46]),
float(rec[46:54])]
try:
at.q = float(rec[54:60])
except ValueError:
at.q = 1.0
try:
at.b = float(rec[60:66])
except ValueError:
at.b = 0.0
at.segi = string.strip(rec[72:76])
at.symbol = string.strip(rec[76:78])
if not len(at.symbol):
at.symbol = at.name[0:1]
if at.symbol in '012345678':
at.symbol = at.name[1:2]
cnt = cnt + 1
model.add_atom(at)
elif (rec[0:3]=='TER'):
if at:
at.ter=1
return(model)
def fromList(self,list): # currently no handling of conect records
model = Indexed()
# read atoms
cnt = 0
at = None
for rec in list:
if (rec[0:4]=='ATOM') or (rec[0:6]=='HETATM'):
at = Atom()
if rec[0]=='A': at.hetatm=0 # default is 1
at.index = cnt
at.name = string.strip(rec[12:16])
at.alt = string.strip(rec[16:17])
at.resn = string.strip(rec[17:20])
at.chain = string.strip(rec[21:22])
at.resi = string.strip(rec[22:27]) # note: insertion is part of resi
at.resi_number = int(rec[22:26])
at.coord = [float(rec[30:38]),
float(rec[38:46]),
float(rec[46:54])]
try:
at.q = float(rec[54:60])
except ValueError:
at.q = 1.0
try:
at.b = float(rec[60:66])
except ValueError:
at.b = 0.0
at.segi = string.strip(rec[72:76])
at.symbol = string.strip(rec[76:78])
if not len(at.symbol):
at.symbol = at.name[0:1]
if at.symbol in '012345678':
at.symbol = at.name[1:2]
cnt = cnt + 1
model.add_atom(at)
elif (rec[0:3]=='TER'):
if at:
at.ter=1
return(model)
def convert_to_chempy_model(atom_array):
"""
Convert an :class:`AtomArray` into a :class:`chempy.models.Indexed`
object.
Returns
-------
chempy_model : Indexed
The converted structure.
"""
model = IndexedModel()
annot_cat = atom_array.get_annotation_categories()
for i in range(atom_array.array_length()):
atom = Atom()
atom.segi = atom_array.chain_id[i]
atom.chain = atom_array.chain_id[i]
atom.resi_number = atom_array.res_id[i]
atom.ins_code = atom_array.ins_code[i]
res_name = atom_array.res_name[i]
atom.resn = res_name
if len(res_name) == 1:
atom.resn_code = res_name
else:
try:
atom.resn_code = ProteinSequence.convert_letter_3to1(res_name)
except KeyError:
atom.resn_code = "X"
atom.hetatm = 1 if atom_array.hetero[i] else 0
atom.name = atom_array.atom_name[i]
atom.symbol = atom_array.element[i]
if "b_factor" in annot_cat:
atom.b = atom_array.b_factor[i]
if "occupancy" in annot_cat:
atom.q = atom_array.occupancy[i]
if "charge" in annot_cat:
atom.formal_charge = atom_array.charge[i]
atom.coord = tuple(atom_array.coord[..., i, :])
atom.index = i + 1
model.add_atom(atom)
if atom_array.bonds is not None:
for i, j, bond_type in atom_array.bonds.as_array():
bond = Bond()
bond.order = BOND_ORDER[bond_type]
bond.index = [i, j]
model.add_bond(bond)
else:
warnings.warn(
"The given atom array (stack) has no associated bond information")
return model
def load_pdbml(filename, object='', discrete=0, multiplex=1, zoom=-1,
quiet=1, _self=cmd):
'''
DESCRIPTION
Load a PDBML formatted structure file
'''
from chempy import Atom, models
from collections import defaultdict
multiplex, discrete = int(multiplex), int(discrete)
try:
root = etree.fromstring(_self.file_read(filename))
PDBxNS = root.tag.rstrip('datablock')
atom_site_list = root.findall('.'
'/' + PDBxNS + 'atom_siteCategory'
'/' + PDBxNS + 'atom_site')
except etree.XMLSyntaxError:
raise CmdException("File doesn't look like XML")
except etree.XPathEvalError:
raise CmdException("XML file doesn't look like a PDBML file")
if not atom_site_list:
raise CmdException("no PDBx:atom_site nodes found in XML file")
# state -> model dictionary
model_dict = defaultdict(models.Indexed)
# atoms
for atom_site in atom_site_list:
atom = Atom()
atom.coord = [None, None, None]
model_num = 1
for child in atom_site:
tag = child.tag
if tag == PDBxNS + 'Cartn_x':
atom.coord[0] = float(child.text)
elif tag == PDBxNS + 'Cartn_y':
atom.coord[1] = float(child.text)
elif tag == PDBxNS + 'Cartn_z':
atom.coord[2] = float(child.text)
elif tag == PDBxNS + 'B_iso_or_equiv':
atom.b = float(child.text)
elif tag == PDBxNS + 'auth_asym_id':
atom.chain = child.text or ''
elif tag == PDBxNS + 'auth_atom_id':
atom.name = child.text or ''
elif tag == PDBxNS + 'auth_comp_id':
atom.resn = child.text or ''
elif tag == PDBxNS + 'auth_seq_id':
atom.resi = child.text or ''
elif tag == PDBxNS + 'label_alt_id':
atom.resi = child.text or ''
elif tag == PDBxNS + 'label_asym_id':
atom.segi = child.text or ''
elif tag == PDBxNS + 'label_atom_id':
if not atom.name:
atom.name = child.text or ''
elif tag == PDBxNS + 'label_comp_id':
if not atom.resn:
atom.resn = child.text or ''
elif tag == PDBxNS + 'label_seq_id':
if not atom.resi:
atom.resi = child.text or ''
elif tag == PDBxNS + 'label_entity_id':
atom.custom = child.text or ''
elif tag == PDBxNS + 'occupancy':
atom.q = float(child.text)
elif tag == PDBxNS + 'pdbx_PDB_model_num':
model_num = int(child.text)
elif tag == PDBxNS + 'type_symbol':
atom.symbol = child.text or ''
elif tag == PDBxNS + 'group_PDB':
atom.hetatm = (child.text == 'HETATM')
if None not in atom.coord:
model_dict[model_num].add_atom(atom)
# symmetry and cell
try:
node = root.findall('.'
'/' + PDBxNS + 'cellCategory'
'/' + PDBxNS + 'cell')[0]
cell = [
float(node.findall('./' + PDBxNS + a)[0].text)
for a in [
'length_a', 'length_b', 'length_c', 'angle_alpha',
'angle_beta', 'angle_gamma'
]
]
spacegroup = root.findall('.'
'/' + PDBxNS + 'symmetryCategory'
'/' + PDBxNS + 'symmetry'
'/' + PDBxNS + 'space_group_name_H-M')[0].text
except IndexError:
cell = None
spacegroup = ''
# object name
if not object:
object = os.path.basename(filename).split('.', 1)[0]
# only multiplex if more than one model/state
multiplex = multiplex and len(model_dict) > 1
# load models as objects or states
for model_num in sorted(model_dict):
if model_num < 1:
print(" Error: model_num < 1 not supported")
continue
model = model_dict[model_num]
model.connect_mode = 3
if cell:
model.cell = cell
model.spacegroup = spacegroup
if multiplex:
oname = '%s_%04d' % (object, model_num)
model_num = 1
else:
oname = object
_self.load_model(model, oname,
state=model_num, zoom=zoom, discrete=discrete)
def load_3d(filename, object=''):
'''
DESCRIPTION
Load a survex 3d cave survey as "molecule"
http://survex.com
http://trac.survex.com/browser/trunk/doc/3dformat.htm
'''
from chempy import Atom, Bond, models
from struct import unpack
if object == '':
object = os.path.splitext(os.path.basename(filename))[0]
f = open(filename, 'rb')
line = f.readline() # File ID
if not line.startswith('Survex 3D Image File'):
print " Error: not a Survex 3D File"
raise CmdException
line = f.readline() # File format version
assert line[0] == 'v'
ff_version = int(line[1:])
line = unicode(f.readline(), 'latin1') # Survex title
line = f.readline() # Timestamp
class Station:
def __init__(self):
self.labels = []
self.adjacent = []
self.lrud = None
self.flag = 0
def connect(self, other):
self.adjacent.append(other)
def is_surface(self):
return self.flag & 0x01
def is_underground(self):
return self.flag & 0x02
def is_entrance(self):
return self.flag & 0x04
def is_exported(self):
return self.flag & 0x08
def is_fixed(self):
return self.flag & 0x10
class Survey(dict):
def __init__(self):
self.prev = None
self.curr_label = ''
self.labelmap = {}
def get(self, xyz):
return dict.setdefault(self, tuple(xyz), Station())
def line(self, xyz):
s = self.get(xyz)
self.prev.connect(s)
self.prev = s
def move(self, xyz):
s = self.get(xyz)
self.prev = s
def label(self, xyz, flag=0):
s = self.get(xyz)
s.labels.append(self.curr_label)
self.labelmap[s.labels[-1]] = s
if flag > 0:
s.flag = flag
def lrud(self, lrud):
s = self.labelmap[self.curr_label]
s.lrud = lrud
survey = Survey()
def read_xyz():
return unpack('<iii', f.read(12))
def read_len():
len = read_byte()
if len == 0xfe:
len += unpack('<H', f.read(2))[0]
elif len == 0xff:
len += unpack('<I', f.read(4))[0]
return len
def read_label():
len = read_len()
if len > 0:
survey.curr_label += skip_bytes(len)
def skip_bytes(n):
return f.read(n)
def read_byte():
byte = f.read(1)
if len(byte) != 1:
return -1
return ord(byte)
while 1:
byte = read_byte()
if byte == -1:
break
if byte == 0x00:
# STOP
survey.curr_label = ''
elif byte <= 0x0e:
# TRIM
# FIXME: according to doc, trim 16 bytes, but img.c does 17!
(i,n) = (-17,0)
while n < byte:
i -= 1
if survey.curr_label[i] == '.': n += 1
survey.curr_label = survey.curr_label[:i + 1]
elif byte <= 0x0f:
# MOVE
xyz = read_xyz()
survey.move(xyz)
elif byte <= 0x1f:
# TRIM
survey.curr_label = survey.curr_label[:15 - byte]
elif byte <= 0x20:
# DATE
if ff_version < 7:
skip_bytes(4)
else:
skip_bytes(2)
elif byte <= 0x21:
# DATE
if ff_version < 7:
skip_bytes(8)
else:
skip_bytes(3)
elif byte <= 0x22:
# Error info
skip_bytes(5 * 4)
elif byte <= 0x23:
# DATE
skip_bytes(4)
elif byte <= 0x24:
# DATE
continue
elif byte <= 0x2f:
# Reserved
continue
elif byte <= 0x31:
# XSECT
read_label()
lrud = unpack('<hhhh', f.read(8))
survey.lrud(lrud)
elif byte <= 0x33:
# XSECT
read_label()
lrud = unpack('<iiii', f.read(16))
survey.lrud(lrud)
elif byte <= 0x3f:
# Reserved
continue
elif byte <= 0x7f:
# LABEL
read_label()
xyz = read_xyz()
survey.label(xyz, byte & 0x3f)
elif byte <= 0xbf:
# LINE
read_label()
xyz = read_xyz()
survey.line(xyz)
elif byte <= 0xff:
# Reserved
continue
model = models.Indexed()
for (xyz,s) in survey.iteritems():
l0, _, l1 = s.labels[0].rpartition('.')
resi, name = l1[:5], l1[5:]
segi, chain, resn = l0[-8:-4], l0[-4:-3], l0[-3:]
atom = Atom()
atom.coord = [i/100.0 for i in xyz]
atom.segi = segi
atom.chain = chain
atom.resn = resn
atom.name = name
atom.resi = resi
atom.b = atom.coord[2]
atom.label = s.labels[0]
if s.lrud is not None:
atom.vdw = sum(s.lrud)/400.0
model.add_atom(atom)
s2i = dict((s,i) for (i,s) in enumerate(survey.itervalues()))
for (s,i) in s2i.iteritems():
for o in s.adjacent:
bnd = Bond()
bnd.index = [i, s2i[o]]
model.add_bond(bnd)
cmd.load_model(model, object, 1)
cmd.show_as('lines', object)
cmd.spectrum('b', 'rainbow', object)
def fromList(self, MMODList):
model = Connected()
# get header information
nAtom = int(MMODList[0][1:6])
model.molecule.title = string.strip(MMODList[0][8:])
irec = 1
# loop through atoms
cnt = 0
for a in range(nAtom):
model.bond.append([])
for a in range(nAtom):
at = Atom()
at.numeric_type = int(MMODList[irec][1:4])
# extract connectivity information
tokens = string.splitfields(MMODList[irec][5:52])
at.neighbor = []
at.bondorder = []
for i in range(6):
if tokens[2 * i] != "0":
a2 = int(tokens[2 * i]) - 1
if (a2 > cnt):
b = Bond()
b.index = [cnt, a2]
b.order = int(tokens[2 * i + 1])
model.bond[b.index[0]].append(
b) # note two refs to same object
model.bond[b.index[1]].append(
b) # note two refs to same object
else:
break
# extract other information
at.coord = [
float(MMODList[irec][53:64]),
float(MMODList[irec][65:76]),
float(MMODList[irec][77:88])
]
at.resi = string.strip(MMODList[irec][89:94])
at.resi_number = int(at.resi)
resn_code = string.strip(MMODList[irec][94:95])
if len(resn_code): at.resn_code = resn_code
color_code = string.strip(MMODList[irec][96:100])
if color_code != '':
at.color_code = int(color_code)
else:
at.color_code = 0
chain = string.strip(MMODList[irec][95:96])
if len(chain): at.chain = chain
at.partial_charge = float(MMODList[irec][100:109])
at.resn = MMODList[irec][119:123]
name = string.strip(MMODList[irec][124:128])
if len(name): at.name = name
model.atom.append(at)
irec = irec + 1
cnt = cnt + 1
# fill in remaining datatypes
cnt = 1
for a in model.atom:
a.text_type = MMOD_atom_data[a.numeric_type][0]
a.symbol = MMOD_atom_data[a.numeric_type][1]
a.formal_charge = MMOD_atom_data[a.numeric_type][4]
cnt = cnt + 1
return (model.convert_to_indexed())
def fromList(self,MMODList):
model = Connected()
# get header information
nAtom = int(MMODList[0][1:6])
model.molecule.title = string.strip(MMODList[0][8:])
irec = 1
# loop through atoms
cnt = 0
for a in range(nAtom):
model.bond.append([])
for a in range(nAtom):
at = Atom()
at.numeric_type = int(MMODList[irec][1:4])
# extract connectivity information
tokens = string.splitfields(MMODList[irec][5:52])
at.neighbor = []
at.bondorder = []
for i in range(6):
if tokens[2*i] != "0":
a2 = int(tokens[2*i])-1
if (a2>cnt):
b = Bond()
b.index = [cnt,a2]
b.order = int(tokens[2*i+1])
model.bond[b.index[0]].append(b) # note two refs to same object
model.bond[b.index[1]].append(b) # note two refs to same object
else:
break
# extract other information
at.coord = [float(MMODList[irec][53:64]),
float(MMODList[irec][65:76]), float(MMODList[irec][77:88])]
at.resi = string.strip(MMODList[irec][89:94])
at.resi_number = int(at.resi)
resn_code = string.strip(MMODList[irec][94:95])
if len(resn_code): at.resn_code = resn_code
color_code = string.strip(MMODList[irec][96:100])
if color_code!='':
at.color_code = int(color_code)
else:
at.color_code = 0
chain = string.strip(MMODList[irec][95:96])
if len(chain): at.chain = chain
at.partial_charge = float(MMODList[irec][100:109])
at.resn = MMODList[irec][119:123]
name = string.strip(MMODList[irec][124:128])
if len(name): at.name = name
model.atom.append(at)
irec = irec + 1
cnt = cnt + 1
# fill in remaining datatypes
cnt = 1
for a in model.atom:
a.text_type = MMOD_atom_data[a.numeric_type][0]
a.symbol = MMOD_atom_data[a.numeric_type][1]
a.formal_charge = MMOD_atom_data[a.numeric_type][4]
cnt = cnt + 1
return(model.convert_to_indexed())
def load_3d(filename, object=''):
'''
DESCRIPTION
Load a survex 3d cave survey as "molecule"
http://survex.com
http://trac.survex.com/browser/trunk/doc/3dformat.htm
'''
from chempy import Atom, Bond, models
from struct import unpack
if object == '':
object = os.path.splitext(os.path.basename(filename))[0]
f = open(filename, 'rb')
line = f.readline() # File ID
if not line.startswith('Survex 3D Image File'):
print(" Error: not a Survex 3D File")
raise CmdException
line = f.readline() # File format version
assert line[0] == 'v'
ff_version = int(line[1:])
line = f.readline().decode('latin1') # Survex title
line = f.readline() # Timestamp
class Station:
def __init__(self):
self.labels = []
self.adjacent = []
self.lrud = None
self.flag = 0
def connect(self, other):
self.adjacent.append(other)
def is_surface(self):
return self.flag & 0x01
def is_underground(self):
return self.flag & 0x02
def is_entrance(self):
return self.flag & 0x04
def is_exported(self):
return self.flag & 0x08
def is_fixed(self):
return self.flag & 0x10
class Survey(dict):
def __init__(self):
self.prev = None
self.curr_label = ''
self.labelmap = {}
def get(self, xyz):
return dict.setdefault(self, tuple(xyz), Station())
def line(self, xyz):
s = self.get(xyz)
self.prev.connect(s)
self.prev = s
def move(self, xyz):
s = self.get(xyz)
self.prev = s
def label(self, xyz, flag=0):
s = self.get(xyz)
s.labels.append(self.curr_label)
self.labelmap[s.labels[-1]] = s
if flag > 0:
s.flag = flag
def lrud(self, lrud):
s = self.labelmap[self.curr_label]
s.lrud = lrud
survey = Survey()
def read_xyz():
return unpack('<iii', f.read(12))
def read_len():
len = read_byte()
if len == 0xfe:
len += unpack('<H', f.read(2))[0]
elif len == 0xff:
len += unpack('<I', f.read(4))[0]
return len
def read_label():
len = read_len()
if len > 0:
survey.curr_label += skip_bytes(len)
def skip_bytes(n):
return f.read(n)
def read_byte():
byte = f.read(1)
if len(byte) != 1:
return -1
return ord(byte)
while 1:
byte = read_byte()
if byte == -1:
break
if byte == 0x00:
# STOP
survey.curr_label = ''
elif byte <= 0x0e:
# TRIM
# FIXME: according to doc, trim 16 bytes, but img.c does 17!
(i, n) = (-17, 0)
while n < byte:
i -= 1
if survey.curr_label[i] == '.': n += 1
survey.curr_label = survey.curr_label[:i + 1]
elif byte <= 0x0f:
# MOVE
xyz = read_xyz()
survey.move(xyz)
elif byte <= 0x1f:
# TRIM
survey.curr_label = survey.curr_label[:15 - byte]
elif byte <= 0x20:
# DATE
if ff_version < 7:
skip_bytes(4)
else:
skip_bytes(2)
elif byte <= 0x21:
# DATE
if ff_version < 7:
skip_bytes(8)
else:
skip_bytes(3)
elif byte <= 0x22:
# Error info
skip_bytes(5 * 4)
elif byte <= 0x23:
# DATE
skip_bytes(4)
elif byte <= 0x24:
# DATE
continue
elif byte <= 0x2f:
# Reserved
continue
elif byte <= 0x31:
# XSECT
read_label()
lrud = unpack('<hhhh', f.read(8))
survey.lrud(lrud)
elif byte <= 0x33:
# XSECT
read_label()
lrud = unpack('<iiii', f.read(16))
survey.lrud(lrud)
elif byte <= 0x3f:
# Reserved
continue
elif byte <= 0x7f:
# LABEL
read_label()
xyz = read_xyz()
survey.label(xyz, byte & 0x3f)
elif byte <= 0xbf:
# LINE
read_label()
xyz = read_xyz()
survey.line(xyz)
elif byte <= 0xff:
# Reserved
continue
model = models.Indexed()
for (xyz, s) in survey.items():
l0, _, l1 = s.labels[0].rpartition('.')
resi, name = l1[:5], l1[5:]
segi, chain, resn = l0[-8:-4], l0[-4:-3], l0[-3:]
atom = Atom()
atom.coord = [i / 100.0 for i in xyz]
atom.segi = segi
atom.chain = chain
atom.resn = resn
atom.name = name
atom.resi = resi
atom.b = atom.coord[2]
atom.label = s.labels[0]
if s.lrud is not None:
atom.vdw = sum(s.lrud) / 400.0
model.add_atom(atom)
s2i = dict((s, i) for (i, s) in enumerate(survey.values()))
for (s, i) in s2i.items():
for o in s.adjacent:
bnd = Bond()
bnd.index = [i, s2i[o]]
model.add_bond(bnd)
cmd.load_model(model, object, 1)
cmd.show_as('lines', object)
cmd.spectrum('b', 'rainbow', object)
def load_3d(filename, object=''):
'''
DESCRIPTION
Load a survex 3d cave survey as "molecule"
http://survex.com
'''
from chempy import Atom, Bond, models
if object == '':
import os
object = os.path.splitext(os.path.basename(filename))[0]
f = open(filename, 'rb')
line = f.readline() # File ID
if not line.startswith('Survex 3D Image File'):
print " Error: not a Survex 3D File"
raise CmdException
line = f.readline() # File format version
assert line[0] == 'v'
ff_version = int(line[1:])
line = unicode(f.readline(), 'latin1') # Survex title
line = f.readline() # Timestamp
class Station(tuple):
def __new__(cls, xyz):
return tuple.__new__(cls, xyz)
def __init__(self, xyz):
self.labels = []
self.adjacent = []
def connect(self, other):
self.adjacent.append(other)
class Survey(dict):
def __init__(self):
self.prev = None
self.curr_label = ''
def get(self, xyz):
s = Station(xyz)
return dict.setdefault(self, s, s)
def line(self, xyz):
s = self.get(xyz)
self.prev.connect(s)
self.prev = s
def move(self, xyz):
s = self.get(xyz)
self.prev = s
def label(self, xyz):
s = survey.get(xyz)
s.labels.append(self.curr_label)
def __repr__(self):
return 'Survey(' + repr(self.keys())[1:-1] + ')'
survey = Survey()
def read_xyz():
x = read_int(4, 1)
y = read_int(4, 1)
z = read_int(4, 1)
return [ x, y, z ]
def read_int(len, sign):
int = 0
for i in range(len):
int |= read_byte() << (8 * i)
if sign and (int >> (8 * len - 1)):
int -= (1 << 8 * len)
return int
def read_len():
len = read_byte()
if len == 0xfe:
len += read_int(2, 0)
elif len == 0xff:
len = read_int(4, 0)
return len
def read_label():
len = read_len()
if len > 0:
survey.curr_label += skip_bytes(len)
def skip_bytes(n):
return f.read(n)
def read_byte():
byte = f.read(1)
if len(byte) != 1:
return -1
return ord(byte)
while 1:
byte = read_byte()
if byte == -1:
break
if byte == 0x00:
# STOP
survey.curr_label = ''
elif byte <= 0x0e:
# TRIM
(i,n) = (-16,0)
while n < byte:
i -= 1
if survey.curr_label[i] == '.': n += 1
survey.curr_label = survey.curr_label[:i + 1]
elif byte <= 0x0f:
# MOVE
xyz = read_xyz()
survey.move(xyz)
elif byte <= 0x1f:
# TRIM
survey.curr_label = survey.curr_label[:15 - byte]
elif byte <= 0x20:
# DATE
if ff_version < 7:
skip_bytes(4)
else:
skip_bytes(2)
elif byte <= 0x21:
# DATE
if ff_version < 7:
skip_bytes(8)
else:
skip_bytes(3)
elif byte <= 0x22:
# Error info
skip_bytes(5 * 4)
elif byte <= 0x23:
# DATE
skip_bytes(4)
elif byte <= 0x24:
# DATE
continue
elif byte <= 0x2f:
# Reserved
continue
elif byte <= 0x31:
# XSECT
read_label()
skip_bytes(4 * 2)
elif byte <= 0x33:
# XSECT
read_label()
skip_bytes(4 * 4)
elif byte <= 0x3f:
# Reserved
continue
elif byte <= 0x7f:
# LABEL
read_label()
xyz = read_xyz()
survey.label(xyz)
elif byte <= 0xbf:
# LINE
read_label()
xyz = read_xyz()
survey.line(xyz)
elif byte <= 0xff:
# Reserved
continue
model = models.Indexed()
for s in survey:
l0, _, l1 = s.labels[0].rpartition('.')
resi, name = l1[:5], l1[5:]
# segi, chain, resn = l0[:4],l0[-4:-3], l0[-3:]
segi, chain, resn = l0[-8:-4], l0[-4:-3], l0[-3:]
atom = Atom()
atom.coord = [i/100.0 for i in s]
atom.segi = segi
atom.chain = chain
atom.resn = resn
atom.name = name
atom.resi = resi
atom.b = atom.coord[2]
model.add_atom(atom)
s2i = dict((s,i) for (i,s) in enumerate(survey))
for s in survey:
for o in s.adjacent:
bnd = Bond()
bnd.index = [s2i[s], s2i[o]]
model.add_bond(bnd)
cmd.load_model(model, object, 1)
cmd.show_as('lines', object)
cmd.spectrum('b', 'rainbow', object)
def get_model(proc, info,number): vmodel=Indexed() atoms=info["AtomsPerMolecule"][number] atomsread=0 ratoms=True rcoords=False rbfactors=False rss=False first=True while(True): v=proc.stdout.readline() if "Molname" in v and ratoms: ad=json.loads(v) at=Atom() at.name=ad["Name"] at.symbol=ad["Symbol"] at.chain=ad["Chain"] at.id=ad["ID"] at.resi_number=ad["MolID"] at.resn=ad["Molname"] vmodel.atom.append(at) atomsread=atomsread+1 if atomsread==atoms: ratoms=False rcoords=True atomsread=0 continue if "Coords" in v and not "Molname" in v and rcoords: coords=json.loads(v) vmodel.atom[atomsread].coord=coords["Coords"] atomsread=atomsread+1 if atomsread==atoms: rcoords=False atomsread=0 if info["Bfactors"]: rbfactors=True if info["SS"]: rss=True continue #In many cases this part will not be needed if "Bfactors" in v: bf=json.loads(v) vmodel.atom[atomsread].b=bf["Bfactors"] atomsread=atomsread+1 if atomsread==atoms: atomsread=0 rbfactors=False if info["SS"]: rss=True continue #This one should be needed only seldomly if "SS" in v: SS=json.loads(v) vmodel.atom[atomsread].ss=SS["SS"] ++atomsread if atomsread==atoms: atomsread=0 rss=False continue # print "me fui con una deuda de 500" break return vmodel
def _to_chempy(data, use_auth=True):
'''
Construct a "chempy" model (molecule) from decoded MMTF data.
'''
from itertools import islice
from chempy import models, Atom, Bond
def add_bond(i1, i2, order, offset=0):
bond = Bond()
bond.order = order
bond.index = [i1 + offset, i2 + offset]
model.add_bond(bond)
coord_iter = data.get_table_iter([
'xCoordList',
'yCoordList',
'zCoordList',
])
atom_iter = data.get_table_iter([
'bFactorList',
'occupancyList',
'altLocList',
'atomIdList',
], [0.0, 1.0, '', -1])
group_iter = data.get_table_iter([
'groupTypeList',
'sequenceIndexList',
'groupIdList',
'insCodeList',
'secStructList',
])
chain_list_iter = enumerate(
data.get_table_iter([
'chainIdList',
'chainNameList',
'groupsPerChain',
]))
groupList = data.get('groupList')
symmetry = (
data.get('unitCell', None),
as_str(data.get('spaceGroup', '')),
)
model_output = []
for n_chains in data.get_iter('chainsPerModel'):
model = models.Indexed()
model_output.append(model)
if symmetry[0] is not None:
model.cell, model.spacegroup = symmetry
for (chain_idx, (segi, chain,
n_groups)) in islice(chain_list_iter, n_chains):
for (groupType, label_seq_id, auth_seq_id, ins_code, ss_info) in \
islice(group_iter, n_groups):
group = groupList[groupType]
resn = as_str(group[b'groupName'])
group_bond_iter = izip(
group[b'bondAtomList'][0::2],
group[b'bondAtomList'][1::2],
group[b'bondOrderList'],
)
offset = len(model.atom)
for (i1, i2, order) in group_bond_iter:
add_bond(i1, i2, order, offset)
group_atom_iter = izip(
group[b'atomNameList'],
group[b'elementList'],
group[b'formalChargeList'],
)
for (name, elem, formal_charge) in group_atom_iter:
atom = Atom()
(atom.b, atom.q, atom.alt, atom.id) = next(atom_iter)
atom.coord = next(coord_iter)
atom.symbol = as_str(elem)
atom.name = as_str(name)
atom.resn = resn
atom.hetatm = label_seq_id == -1
atom.formal_charge = formal_charge
atom.segi = segi
atom.chain = chain
atom.ss = ss_map.get(ss_info, '')
if use_auth or label_seq_id is None:
atom.resi = auth_seq_id
atom.ins_code = ins_code or ''
else:
atom.resi = label_seq_id + 1
model.add_atom(atom)
model_atom_max = 0
model_atom_min = 0
model_iter = iter(model_output)
bondAtomList_iter = data.get_iter('bondAtomList')
for order in data.get_iter('bondOrderList'):
i1 = next(bondAtomList_iter)
i2 = next(bondAtomList_iter)
if i1 >= model_atom_max or i2 >= model_atom_max:
model = next(model_iter)
model_atom_min = model_atom_max
model_atom_max += len(model.atom)
add_bond(i1, i2, order, -model_atom_min)
return model_output
def load_pdbml(filename, object='', discrete=0, multiplex=1, zoom=-1,
quiet=1, _self=cmd):
'''
DESCRIPTION
Load a PDBML formatted structure file
'''
from chempy import Atom, models
from collections import defaultdict
multiplex, discrete = int(multiplex), int(discrete)
try:
root = etree.fromstring(_self.file_read(filename))
PDBxNS = root.tag.rstrip('datablock')
atom_site_list = root.findall('.'
'/' + PDBxNS + 'atom_siteCategory'
'/' + PDBxNS + 'atom_site')
except etree.XMLSyntaxError:
raise CmdException("File doesn't look like XML")
except etree.XPathEvalError:
raise CmdException("XML file doesn't look like a PDBML file")
if not atom_site_list:
raise CmdException("no PDBx:atom_site nodes found in XML file")
# state -> model dictionary
model_dict = defaultdict(models.Indexed)
# atoms
for atom_site in atom_site_list:
atom = Atom()
atom.coord = [None, None, None]
model_num = 1
for child in atom_site:
tag = child.tag
if tag == PDBxNS + 'Cartn_x':
atom.coord[0] = float(child.text)
elif tag == PDBxNS + 'Cartn_y':
atom.coord[1] = float(child.text)
elif tag == PDBxNS + 'Cartn_z':
atom.coord[2] = float(child.text)
elif tag == PDBxNS + 'B_iso_or_equiv':
atom.b = float(child.text)
elif tag == PDBxNS + 'auth_asym_id':
atom.chain = child.text or ''
elif tag == PDBxNS + 'auth_atom_id':
atom.name = child.text or ''
elif tag == PDBxNS + 'auth_comp_id':
atom.resn = child.text or ''
elif tag == PDBxNS + 'auth_seq_id':
atom.resi = child.text or ''
elif tag == PDBxNS + 'label_alt_id':
atom.resi = child.text or ''
elif tag == PDBxNS + 'label_asym_id':
atom.segi = child.text or ''
elif tag == PDBxNS + 'label_atom_id':
if not atom.name:
atom.name = child.text or ''
elif tag == PDBxNS + 'label_comp_id':
if not atom.resn:
atom.resn = child.text or ''
elif tag == PDBxNS + 'label_seq_id':
if not atom.resi:
atom.resi = child.text or ''
elif tag == PDBxNS + 'label_entity_id':
atom.custom = child.text or ''
elif tag == PDBxNS + 'occupancy':
atom.q = float(child.text)
elif tag == PDBxNS + 'pdbx_PDB_model_num':
model_num = int(child.text)
elif tag == PDBxNS + 'type_symbol':
atom.symbol = child.text or ''
elif tag == PDBxNS + 'group_PDB':
atom.hetatm = (child.text == 'HETATM')
if None not in atom.coord:
model_dict[model_num].add_atom(atom)
# symmetry and cell
try:
node = root.findall('.'
'/' + PDBxNS + 'cellCategory'
'/' + PDBxNS + 'cell')[0]
cell = [
float(node.findall('./' + PDBxNS + a)[0].text)
for a in [
'length_a', 'length_b', 'length_c', 'angle_alpha',
'angle_beta', 'angle_gamma'
]
]
spacegroup = root.findall('.'
'/' + PDBxNS + 'symmetryCategory'
'/' + PDBxNS + 'symmetry'
'/' + PDBxNS + 'space_group_name_H-M')[0].text
except IndexError:
cell = None
spacegroup = ''
# object name
if not object:
object = os.path.basename(filename).split('.', 1)[0]
# only multiplex if more than one model/state
multiplex = multiplex and len(model_dict) > 1
# load models as objects or states
for model_num in sorted(model_dict):
if model_num < 1:
print(" Error: model_num < 1 not supported")
continue
model = model_dict[model_num]
model.connect_mode = 3
if cell:
model.cell = cell
model.spacegroup = spacegroup
if multiplex:
oname = '%s_%04d' % (object, model_num)
model_num = 1
else:
oname = object
_self.load_model(model, oname,
state=model_num, zoom=zoom, discrete=discrete)
def GetModel(proc, info, number):
vmodel = Indexed()
atoms = info["AtomsPerMolecule"][number]
states = info["FramesPerMolecule"][number]
statecontainer = States(atoms)
atomsread = 0
statesread = 0
ratoms = True
rcoords = False
rbfactors = False
rss = False
first = True
while (True):
v = proc.stdout.readline().decode("utf-8")
if "Molname" in v and ratoms:
ad = json.loads(v)
at = Atom()
at.name = ad["Name"]
at.symbol = ad["Symbol"]
at.chain = ad["Chain"]
at.id = ad["ID"]
at.resi_number = ad["MolID"]
at.resn = ad["Molname"]
at.hetatm = 1
if at.resn in resncodes.keys():
at.hetatm = 0
at.resn_code = resncodes[ad["Molname"]]
vmodel.atom.append(at)
atomsread = atomsread + 1
if atomsread == atoms:
ratoms = False
rcoords = True
atomsread = 0
continue
if "Coords" in v and not "Molname" in v and rcoords:
coords = json.loads(v)
if statesread == 0:
vmodel.atom[atomsread].coord = coords["Coords"]
else:
statecontainer.NewCoords(coords["Coords"])
# print("coords added to state!!!") #############3
atomsread = atomsread + 1
if atomsread == atoms:
statesread += 1
atomsread = 0
if statesread == states:
rcoords = False
if info["Bfactors"]:
rbfactors = True
if info["SS"]:
rss = True
else:
statecontainer.NewState()
# print("NEW STATE!") ########3
continue
#In many cases this part will not be needed
if "Bfactors" in v:
bf = json.loads(v)
vmodel.atom[atomsread].b = bf["Bfactors"]
atomsread = atomsread + 1
if atomsread == atoms:
atomsread = 0
rbfactors = False
if info["SS"]:
rss = True
continue
#This one should be needed only seldom
if "SS" in v:
SS = json.loads(v)
vmodel.atom[atomsread].ss = SS["SS"]
++atomsread
if atomsread == atoms:
atomsread = 0
rss = False
continue
# print "me fui con una deuda de 500"
break
#i don't like this, but I don't want to break my other plugins by changing the signature of the function
#for them. Also, most pugins will only get one state anyway.
if statecontainer.States() < 1:
return vmodel
else:
return (vmodel, statecontainer)
