def _parseatoms(self, pdb):
atoms = []
# translate Bio.PDB atom objects to MDAnalysis Atom.
for iatom, atom in enumerate(pdb.get_atoms()):
residue = atom.parent
chain_id = residue.parent.id
atomname = atom.name
atomtype = guess_atom_type(atomname)
resname = residue.resname
resid = int(residue.id[1])
# no empty segids (or Universe throws IndexError)
segid = residue.get_segid().strip() or chain_id or "SYSTEM"
mass = guess_atom_mass(atomname)
charge = guess_atom_charge(atomname)
bfactor = atom.bfactor
# occupancy = atom.occupancy
atoms.append(
Atom(iatom,
atomname,
atomtype,
resname,
resid,
segid,
mass,
charge,
bfactor=bfactor))
return atoms
def _parseatoms(self, pdb):
atoms = []
# translate list of atoms to MDAnalysis Atom.
for iatom, atom in enumerate(pdb._atoms):
# ATOM
if len(atom.__dict__) == 10:
atomname = atom.name
atomtype = atom.element or guess_atom_type(atomname)
resname = atom.resName
resid = int(atom.resSeq)
chain = atom.chainID.strip()
# no empty segids (or Universe throws IndexError)
segid = atom.segID.strip() or chain or "SYSTEM"
mass = guess_atom_mass(atomname)
charge = guess_atom_charge(atomname)
bfactor = atom.tempFactor
# occupancy = atom.occupancy
altLoc = atom.altLoc
atoms.append(Atom(iatom, atomname, atomtype, resname, resid,
segid, mass, charge,
bfactor=bfactor, serial=atom.serial,
altLoc=altLoc))
# TER atoms
#elif len(atom.__dict__) == 5:
# pass
# #atoms.append(None)
self.structure["_atoms"] = atoms
def parse(self):
"""Parse CRD file *filename* and return the dict `structure`.
Only reads the list of atoms.
:Returns: MDAnalysis internal *structure* dict
.. SeeAlso:: The *structure* dict is defined in
`MDAnalysis.topology`
"""
extformat = FORTRANReader(
'2I10,2X,A8,2X,A8,3F20.10,2X,A8,2X,A8,F20.10')
stdformat = FORTRANReader('2I5,1X,A4,1X,A4,3F10.5,1X,A4,1X,A4,F10.5')
atoms = []
atom_serial = 0
with openany(self.filename) as crd:
for linenum, line in enumerate(crd):
# reading header
if line.split()[0] == '*':
continue
elif line.split()[-1] == 'EXT' and bool(int(
line.split()[0])) is True:
r = extformat
continue
elif line.split()[0] == line.split(
)[-1] and line.split()[0] != '*':
r = stdformat
continue
# anything else should be an atom
try:
serial, TotRes, resName, name, x, y, z, chainID, resSeq, tempFactor = r.read(
line)
except:
raise ValueError("Check CRD format at line {}: {}".format(
linenum, line.rstrip()))
atomtype = guess_atom_type(name)
mass = guess_atom_mass(name)
charge = guess_atom_charge(name)
atoms.append(
Atom(atom_serial, name, atomtype, resName, TotRes, chainID,
mass, charge))
atom_serial += 1
structure = {}
structure["_atoms"] = atoms
return structure
def parse(self):
"""Parse CRD file *filename* and return the dict `structure`.
Only reads the list of atoms.
:Returns: MDAnalysis internal *structure* dict
.. SeeAlso:: The *structure* dict is defined in
`MDAnalysis.topology`
"""
extformat = FORTRANReader('2I10,2X,A8,2X,A8,3F20.10,2X,A8,2X,A8,F20.10')
stdformat = FORTRANReader('2I5,1X,A4,1X,A4,3F10.5,1X,A4,1X,A4,F10.5')
atoms = []
atom_serial = 0
with openany(self.filename) as crd:
for linenum, line in enumerate(crd):
# reading header
if line.split()[0] == '*':
continue
elif line.split()[-1] == 'EXT' and bool(int(line.split()[0])) is True:
r = extformat
continue
elif line.split()[0] == line.split()[-1] and line.split()[0] != '*':
r = stdformat
continue
# anything else should be an atom
try:
serial, TotRes, resName, name, x, y, z, chainID, resSeq, tempFactor = r.read(line)
except:
raise ValueError("Check CRD format at line {}: {}".format(
linenum, line.rstrip()))
atomtype = guess_atom_type(name)
mass = guess_atom_mass(name)
charge = guess_atom_charge(name)
atoms.append(Atom(atom_serial, name, atomtype, resName, TotRes, chainID, mass, charge))
atom_serial += 1
structure = {}
structure["_atoms"] = atoms
return structure
def _parseatoms(self, pqr):
atoms = []
# translate list of atoms to MDAnalysis Atom.
for iatom, atom in enumerate(pqr._atoms):
atomname = atom.name
atomtype = guess_atom_type(atomname)
resname = atom.resName
resid = int(atom.resSeq)
chain = atom.chainID.strip()
# no empty segids (or Universe throws IndexError)
segid = atom.segID.strip() or chain or "SYSTEM"
mass = guess_atom_mass(atomname)
charge = float(atom.charge)
radius = atom.radius
atoms.append(Atom(iatom, atomname, atomtype, resname, resid,
segid, mass, charge, radius=radius))
return atoms
def _parseatoms(self, pdb):
atoms = []
# translate Bio.PDB atom objects to MDAnalysis Atom.
for iatom, atom in enumerate(pdb.get_atoms()):
residue = atom.parent
chain_id = residue.parent.id
atomname = atom.name
atomtype = guess_atom_type(atomname)
resname = residue.resname
resid = int(residue.id[1])
# no empty segids (or Universe throws IndexError)
segid = residue.get_segid().strip() or chain_id or "SYSTEM"
mass = guess_atom_mass(atomname)
charge = guess_atom_charge(atomname)
bfactor = atom.bfactor
# occupancy = atom.occupancy
atoms.append(Atom(iatom, atomname, atomtype, resname, resid, segid,
mass, charge, bfactor=bfactor))
return atoms
def _parseatoms(self, pdb):
atoms = []
# translate list of atoms to MDAnalysis Atom.
for iatom, atom in enumerate(pdb._atoms):
# ATOM
if len(atom.__dict__) == 10:
atomname = atom.name
atomtype = atom.element or guess_atom_type(atomname)
resname = atom.resName
resid = int(atom.resSeq)
chain = atom.chainID.strip()
# no empty segids (or Universe throws IndexError)
segid = atom.segID.strip() or chain or "SYSTEM"
mass = guess_atom_mass(atomname)
charge = guess_atom_charge(atomname)
bfactor = atom.tempFactor
# occupancy = atom.occupancy
altLoc = atom.altLoc
atoms.append(
Atom(iatom,
atomname,
atomtype,
resname,
resid,
segid,
mass,
charge,
bfactor=bfactor,
serial=atom.serial,
altLoc=altLoc))
# TER atoms
#elif len(atom.__dict__) == 5:
# pass
# #atoms.append(None)
self.structure["_atoms"] = atoms
def parse(self):
"""Parse DMS file *filename* and return the dict `structure`.
Only reads the list of atoms.
:Returns: MDAnalysis internal *structure* dict, which contains
Atom and Bond objects
.. SeeAlso:: The *structure* dict is defined in
`MDAnalysis.topology`.
"""
# Fix by SB: Needed because sqlite3.connect does not raise anything if file is not there
if not os.path.isfile(self.filename):
raise IOError("No such file: {}".format(self.filename))
def dict_factory(cursor, row):
"""
Fetch SQL records as dictionaries, rather than the default tuples.
"""
d = {}
for idx, col in enumerate(cursor.description):
d[col[0]] = row[idx]
return d
with sqlite3.connect(self.filename) as con:
try:
# This will return dictionaries instead of tuples,
# when calling cur.fetch() or fetchall()
con.row_factory = dict_factory
cur = con.cursor()
cur.execute('SELECT * FROM particle')
particles = cur.fetchall()
except sqlite3.DatabaseError:
raise IOError("Failed reading the atoms from DMS Database")
else:
# p["anum"] contains the atomic number
try:
atoms = [
Atom(p["id"], p["name"].strip(),
guess_atom_type(p["name"].strip()),
p["resname"].strip(), p["resid"],
p["segid"].strip(), p["mass"], p["charge"])
for p in particles
]
except KeyError:
raise ValueError("Failed reading atom information")
try:
cur.execute('SELECT * FROM bond')
bonds = cur.fetchall()
except sqlite3.DatabaseError:
raise IOError("Failed reading the bonds from DMS Database")
else:
bondlist = []
bondorder = {}
for b in bonds:
desc = tuple(sorted([b['p0'], b['p1']]))
bondlist.append(desc)
bondorder[desc] = b['order']
# All the records below besides donors and acceptors can be contained in a DMS file.
# In addition to the coordinates and bonds, DMS may contain the entire force-field
# information (terms+parameters),
structure = {
"_atoms": atoms,
"_bonds": tuple(bondlist),
"_bondorder": bondorder
}
return structure
def guess_types(names):
"""GRO doesn't supply types, this returns an Attr"""
return Atomtypes(
np.array([guess_atom_type(name) for name in names], dtype=object))
def check_atom_type(atype, aname):
assert_equal(guess_atom_type(aname), atype)
def check_atom_type(atype, aname):
assert_equal(guess_atom_type(aname), atype)
def parse(self, filename=None):
"""Parse MOL2 file *filename* and return the dict `structure`.
Only reads the list of atoms.
:Returns: MDAnalysis internal *structure* dict
.. SeeAlso:: The *structure* dict is defined in
:func:`MDAnalysis.topology.PSFParser.PSFParser`.
"""
if not filename:
filename = self.filename
blocks = []
with openany(filename) as f:
for i, line in enumerate(f):
# found new molecules
if "@<TRIPOS>MOLECULE" in line:
if len(blocks):
break
blocks.append({"start_line": i, "lines": []})
blocks[-1]["lines"].append(line)
if not len(blocks):
raise ValueError("The mol2 file '{}' needs to have at least one"
" @<TRIPOS>MOLECULE block".format(filename))
block = blocks[0]
sections = {}
cursor = None
for line in block["lines"]:
if "@<TRIPOS>" in line:
cursor = line.split("@<TRIPOS>")[1].strip().lower()
sections[cursor] = []
continue
elif line.startswith("#") or line == "\n":
continue
sections[cursor].append(line)
atom_lines, bond_lines = sections["atom"], sections["bond"]
if not len(atom_lines):
raise ValueError("The mol2 block ({}:{}) has no atoms".format(
os.path.basename(filename), block["start_line"]))
if not len(bond_lines):
raise ValueError("The mol2 block ({}:{}) has no bonds".format(
os.path.basename(filename), block["start_line"]))
atoms = []
for a in atom_lines:
aid, name, x, y, z, atom_type, resid, resname, charge = a.split()
aid = int(aid) - 1
#x, y, z = float(x), float(y), float(z)
resid = int(resid)
charge = float(charge)
element = guess_atom_type(name)
mass = guess_atom_mass(element)
# atom type is sybl atom type
atoms.append(
Atom(aid, name, atom_type, resname, resid, "X", mass, charge))
#guess_atom_type(a.split()[1]
bonds = []
bondorder = {}
for b in bond_lines:
# bond_type can be: 1, 2, am, ar
bid, a0, a1, bond_type = b.split()
a0, a1 = int(a0) - 1, int(a1) - 1
bond = tuple(sorted([a0, a1]))
bondorder[bond] = bond_type
bonds.append(bond)
structure = {"_atoms": atoms, "_bonds": bonds, "_bondorder": bondorder}
return structure
def guess_types(names):
"""GRO doesn't supply types, this returns an Attr"""
return Atomtypes(np.array([guess_atom_type(name) for name in names], dtype=object))
def parse(self):
"""Parse DMS file *filename* and return the dict `structure`.
Only reads the list of atoms.
:Returns: MDAnalysis internal *structure* dict, which contains
Atom and Bond objects
.. SeeAlso:: The *structure* dict is defined in
`MDAnalysis.topology`.
"""
# Fix by SB: Needed because sqlite3.connect does not raise anything if file is not there
if not os.path.isfile(self.filename):
raise IOError("No such file: {}".format(self.filename))
def dict_factory(cursor, row):
"""
Fetch SQL records as dictionaries, rather than the default tuples.
"""
d = {}
for idx, col in enumerate(cursor.description):
d[col[0]] = row[idx]
return d
with sqlite3.connect(self.filename) as con:
try:
# This will return dictionaries instead of tuples,
# when calling cur.fetch() or fetchall()
con.row_factory = dict_factory
cur = con.cursor()
cur.execute('SELECT * FROM particle')
particles = cur.fetchall()
except sqlite3.DatabaseError:
raise IOError("Failed reading the atoms from DMS Database")
else:
# p["anum"] contains the atomic number
try:
atoms = [Atom(p["id"], p["name"].strip(),
guess_atom_type(p["name"].strip()),
p["resname"].strip(), p["resid"],
p["segid"].strip(), p["mass"], p["charge"])
for p in particles]
except KeyError:
raise ValueError("Failed reading atom information")
try:
cur.execute('SELECT * FROM bond')
bonds = cur.fetchall()
except sqlite3.DatabaseError:
raise IOError("Failed reading the bonds from DMS Database")
else:
bondlist = []
bondorder = {}
for b in bonds:
desc = tuple(sorted([b['p0'], b['p1']]))
bondlist.append(desc)
bondorder[desc] = b['order']
# All the records below besides donors and acceptors can be contained in a DMS file.
# In addition to the coordinates and bonds, DMS may contain the entire force-field
# information (terms+parameters),
structure = {"_atoms": atoms, "_bonds": tuple(bondlist), "_bondorder": bondorder}
return structure
def parse(self, filename=None):
"""Parse MOL2 file *filename* and return the dict `structure`.
Only reads the list of atoms.
:Returns: MDAnalysis internal *structure* dict
.. SeeAlso:: The *structure* dict is defined in
:func:`MDAnalysis.topology.PSFParser.PSFParser`.
"""
if not filename:
filename = self.filename
blocks = []
with openany(filename) as f:
for i, line in enumerate(f):
# found new molecules
if "@<TRIPOS>MOLECULE" in line:
if len(blocks):
break
blocks.append({"start_line": i, "lines": []})
blocks[-1]["lines"].append(line)
if not len(blocks):
raise ValueError("The mol2 file '{}' needs to have at least one"
" @<TRIPOS>MOLECULE block".format(filename))
block = blocks[0]
sections = {}
cursor = None
for line in block["lines"]:
if "@<TRIPOS>" in line:
cursor = line.split("@<TRIPOS>")[1].strip().lower()
sections[cursor] = []
continue
elif line.startswith("#") or line == "\n":
continue
sections[cursor].append(line)
atom_lines, bond_lines = sections["atom"], sections["bond"]
if not len(atom_lines):
raise ValueError("The mol2 block ({}:{}) has no atoms".format(
os.path.basename(filename), block["start_line"]))
if not len(bond_lines):
raise ValueError("The mol2 block ({}:{}) has no bonds".format(
os.path.basename(filename), block["start_line"]))
atoms = []
for a in atom_lines:
aid, name, x, y, z, atom_type, resid, resname, charge = a.split()
aid = int(aid) - 1
#x, y, z = float(x), float(y), float(z)
resid = int(resid)
charge = float(charge)
element = guess_atom_type(name)
mass = guess_atom_mass(element)
# atom type is sybl atom type
atoms.append(Atom(aid, name, atom_type, resname,
resid, "X", mass, charge))
#guess_atom_type(a.split()[1]
bonds = []
bondorder = {}
for b in bond_lines:
# bond_type can be: 1, 2, am, ar
bid, a0, a1, bond_type = b.split()
a0, a1 = int(a0) - 1, int(a1) - 1
bond = tuple(sorted([a0, a1]))
bondorder[bond] = bond_type
bonds.append(bond)
structure = {"_atoms": atoms,
"_bonds": bonds,
"_bondorder": bondorder}
return structure
