コード例 #1
0
ファイル: PDBParser.py プロジェクト: cbirdlab/sap
    def __init__(self, PERMISSIVE=True, get_header=False,
                 structure_builder=None, QUIET=False):
        """
        The PDB parser call a number of standard methods in an aggregated
        StructureBuilder object. Normally this object is instanciated by the
        PDBParser object itself, but if the user provides his/her own
        StructureBuilder object, the latter is used instead.

        Arguments:

        o PERMISSIVE - Evaluated as a Boolean. If false, exceptions in
        constructing the SMCRA data structure are fatal. If true (DEFAULT),
        the exceptions are caught, but some residues or atoms will be missing.
        THESE EXCEPTIONS ARE DUE TO PROBLEMS IN THE PDB FILE!.

        o structure_builder - an optional user implemented StructureBuilder class.

        o QUIET - Evaluated as a Boolean. If true, warnings issued in constructing
        the SMCRA data will be suppressed. If false (DEFAULT), they will be shown.
        These warnings might be indicative of problems in the PDB file!
        """
        if structure_builder is not None:
            self.structure_builder = structure_builder
        else:
            self.structure_builder = StructureBuilder()
        self.header = None
        self.trailer = None
        self.line_counter = 0
        self.PERMISSIVE = bool(PERMISSIVE)
        self.QUIET = bool(QUIET)
コード例 #2
0
ファイル: PDBParser.py プロジェクト: kaspermunch/sap
    def __init__(self, PERMISSIVE=True, get_header=False, structure_builder=None, QUIET=False):
        """
        The PDB parser call a number of standard methods in an aggregated
        StructureBuilder object. Normally this object is instanciated by the
        PDBParser object itself, but if the user provides his/her own
        StructureBuilder object, the latter is used instead.

        Arguments:

        o PERMISSIVE - Evaluated as a Boolean. If false, exceptions in
        constructing the SMCRA data structure are fatal. If true (DEFAULT),
        the exceptions are caught, but some residues or atoms will be missing.
        THESE EXCEPTIONS ARE DUE TO PROBLEMS IN THE PDB FILE!.

        o structure_builder - an optional user implemented StructureBuilder class.

        o QUIET - Evaluated as a Boolean. If true, warnings issued in constructing
        the SMCRA data will be suppressed. If false (DEFAULT), they will be shown.
        These warnings might be indicative of problems in the PDB file!
        """
        if structure_builder is not None:
            self.structure_builder = structure_builder
        else:
            self.structure_builder = StructureBuilder()
        self.header = None
        self.trailer = None
        self.line_counter = 0
        self.PERMISSIVE = bool(PERMISSIVE)
        self.QUIET = bool(QUIET)
コード例 #3
0
ファイル: PDBIO.py プロジェクト: kaspermunch/sap
    def set_structure(self, pdb_object):
        # Check what the user is providing and build a structure appropriately
        if pdb_object.level == "S":
            structure = pdb_object
        else:
            sb = StructureBuilder()
            sb.init_structure('pdb')
            sb.init_seg(' ')
            # Build parts as necessary
            if pdb_object.level == "M":
                sb.structure.add(pdb_object)
                self.structure = sb.structure
            else:
                sb.init_model(0)
                if pdb_object.level == "C":
                    sb.structure[0].add(pdb_object)
                else:
                    sb.init_chain('A')
                    if pdb_object.level == "R":
                        try:
                            parent_id = pdb_object.parent.id
                            sb.structure[0]['A'].id = parent_id
                        except Exception:
                            pass
                        sb.structure[0]['A'].add(pdb_object)
                    else:
                        # Atom
                        sb.init_residue('DUM', ' ', 1, ' ')
                        try:
                            parent_id = pdb_object.parent.parent.id
                            sb.structure[0]['A'].id = parent_id
                        except Exception:
                            pass
                        sb.structure[0]['A'].child_list[0].add(pdb_object)

            # Return structure
            structure = sb.structure
        self.structure=structure
コード例 #4
0
ファイル: PDBParser.py プロジェクト: kaspermunch/sap
class PDBParser(object):
    """
    Parse a PDB file and return a Structure object.
    """

    def __init__(self, PERMISSIVE=True, get_header=False, structure_builder=None, QUIET=False):
        """
        The PDB parser call a number of standard methods in an aggregated
        StructureBuilder object. Normally this object is instanciated by the
        PDBParser object itself, but if the user provides his/her own
        StructureBuilder object, the latter is used instead.

        Arguments:

        o PERMISSIVE - Evaluated as a Boolean. If false, exceptions in
        constructing the SMCRA data structure are fatal. If true (DEFAULT),
        the exceptions are caught, but some residues or atoms will be missing.
        THESE EXCEPTIONS ARE DUE TO PROBLEMS IN THE PDB FILE!.

        o structure_builder - an optional user implemented StructureBuilder class.

        o QUIET - Evaluated as a Boolean. If true, warnings issued in constructing
        the SMCRA data will be suppressed. If false (DEFAULT), they will be shown.
        These warnings might be indicative of problems in the PDB file!
        """
        if structure_builder is not None:
            self.structure_builder = structure_builder
        else:
            self.structure_builder = StructureBuilder()
        self.header = None
        self.trailer = None
        self.line_counter = 0
        self.PERMISSIVE = bool(PERMISSIVE)
        self.QUIET = bool(QUIET)

    # Public methods

    def get_structure(self, id, file):
        """Return the structure.

        Arguments:
        o id - string, the id that will be used for the structure
        o file - name of the PDB file OR an open filehandle
        """
        with warnings.catch_warnings():
            if self.QUIET:
                warnings.filterwarnings("ignore", category=PDBConstructionWarning)

            self.header = None
            self.trailer = None
            # Make a StructureBuilder instance (pass id of structure as parameter)
            self.structure_builder.init_structure(id)

            with as_handle(file) as handle:
                self._parse(handle.readlines())

            self.structure_builder.set_header(self.header)
            # Return the Structure instance
            structure = self.structure_builder.get_structure()

        return structure

    def get_header(self):
        "Return the header."
        return self.header

    def get_trailer(self):
        "Return the trailer."
        return self.trailer

    # Private methods

    def _parse(self, header_coords_trailer):
        "Parse the PDB file."
        # Extract the header; return the rest of the file
        self.header, coords_trailer = self._get_header(header_coords_trailer)
        # Parse the atomic data; return the PDB file trailer
        self.trailer = self._parse_coordinates(coords_trailer)

    def _get_header(self, header_coords_trailer):
        "Get the header of the PDB file, return the rest."
        structure_builder = self.structure_builder
        i = 0
        for i in range(0, len(header_coords_trailer)):
            structure_builder.set_line_counter(i + 1)
            line = header_coords_trailer[i]
            record_type = line[0:6]
            if record_type == "ATOM  " or record_type == "HETATM" or record_type == "MODEL ":
                break
        header = header_coords_trailer[0:i]
        # Return the rest of the coords+trailer for further processing
        self.line_counter = i
        coords_trailer = header_coords_trailer[i:]
        header_dict = _parse_pdb_header_list(header)
        return header_dict, coords_trailer

    def _parse_coordinates(self, coords_trailer):
        "Parse the atomic data in the PDB file."
        local_line_counter = 0
        structure_builder = self.structure_builder
        current_model_id = 0
        # Flag we have an open model
        model_open = 0
        current_chain_id = None
        current_segid = None
        current_residue_id = None
        current_resname = None
        for i in range(0, len(coords_trailer)):
            line = coords_trailer[i]
            record_type = line[0:6]
            global_line_counter = self.line_counter + local_line_counter + 1
            structure_builder.set_line_counter(global_line_counter)
            if record_type == "ATOM  " or record_type == "HETATM":
                # Initialize the Model - there was no explicit MODEL record
                if not model_open:
                    structure_builder.init_model(current_model_id)
                    current_model_id += 1
                    model_open = 1
                fullname = line[12:16]
                # get rid of whitespace in atom names
                split_list = fullname.split()
                if len(split_list) != 1:
                    # atom name has internal spaces, e.g. " N B ", so
                    # we do not strip spaces
                    name = fullname
                else:
                    # atom name is like " CA ", so we can strip spaces
                    name = split_list[0]
                altloc = line[16]
                resname = line[17:20]
                chainid = line[21]
                try:
                    serial_number = int(line[6:11])
                except:
                    serial_number = 0
                resseq = int(line[22:26].split()[0])  # sequence identifier
                icode = line[26]  # insertion code
                if record_type == "HETATM":  # hetero atom flag
                    if resname == "HOH" or resname == "WAT":
                        hetero_flag = "W"
                    else:
                        hetero_flag = "H"
                else:
                    hetero_flag = " "
                residue_id = (hetero_flag, resseq, icode)
                # atomic coordinates
                try:
                    x = float(line[30:38])
                    y = float(line[38:46])
                    z = float(line[46:54])
                except:
                    # Should we allow parsing to continue in permissive mode?
                    # If so, what coordinates should we default to?  Easier to abort!
                    raise PDBConstructionException("Invalid or missing coordinate(s) at line %i." % global_line_counter)
                coord = numpy.array((x, y, z), "f")
                # occupancy & B factor
                try:
                    occupancy = float(line[54:60])
                except:
                    self._handle_PDB_exception("Invalid or missing occupancy", global_line_counter)
                    occupancy = None  # Rather than arbitrary zero or one
                try:
                    bfactor = float(line[60:66])
                except:
                    self._handle_PDB_exception("Invalid or missing B factor", global_line_counter)
                    bfactor = 0.0  # The PDB use a default of zero if the data is missing
                segid = line[72:76]
                element = line[76:78].strip()
                if current_segid != segid:
                    current_segid = segid
                    structure_builder.init_seg(current_segid)
                if current_chain_id != chainid:
                    current_chain_id = chainid
                    structure_builder.init_chain(current_chain_id)
                    current_residue_id = residue_id
                    current_resname = resname
                    try:
                        structure_builder.init_residue(resname, hetero_flag, resseq, icode)
                    except PDBConstructionException as message:
                        self._handle_PDB_exception(message, global_line_counter)
                elif current_residue_id != residue_id or current_resname != resname:
                    current_residue_id = residue_id
                    current_resname = resname
                    try:
                        structure_builder.init_residue(resname, hetero_flag, resseq, icode)
                    except PDBConstructionException as message:
                        self._handle_PDB_exception(message, global_line_counter)
                # init atom
                try:
                    structure_builder.init_atom(
                        name, coord, bfactor, occupancy, altloc, fullname, serial_number, element
                    )
                except PDBConstructionException as message:
                    self._handle_PDB_exception(message, global_line_counter)
            elif record_type == "ANISOU":
                anisou = [
                    float(x) for x in (line[28:35], line[35:42], line[43:49], line[49:56], line[56:63], line[63:70])
                ]
                # U's are scaled by 10^4
                anisou_array = (numpy.array(anisou, "f") / 10000.0).astype("f")
                structure_builder.set_anisou(anisou_array)
            elif record_type == "MODEL ":
                try:
                    serial_num = int(line[10:14])
                except:
                    self._handle_PDB_exception("Invalid or missing model serial number", global_line_counter)
                    serial_num = 0
                structure_builder.init_model(current_model_id, serial_num)
                current_model_id += 1
                model_open = 1
                current_chain_id = None
                current_residue_id = None
            elif record_type == "END   " or record_type == "CONECT":
                # End of atomic data, return the trailer
                self.line_counter += local_line_counter
                return coords_trailer[local_line_counter:]
            elif record_type == "ENDMDL":
                model_open = 0
                current_chain_id = None
                current_residue_id = None
            elif record_type == "SIGUIJ":
                # standard deviation of anisotropic B factor
                siguij = [
                    float(x) for x in (line[28:35], line[35:42], line[42:49], line[49:56], line[56:63], line[63:70])
                ]
                # U sigma's are scaled by 10^4
                siguij_array = (numpy.array(siguij, "f") / 10000.0).astype("f")
                structure_builder.set_siguij(siguij_array)
            elif record_type == "SIGATM":
                # standard deviation of atomic positions
                sigatm = [float(x) for x in (line[30:38], line[38:45], line[46:54], line[54:60], line[60:66])]
                sigatm_array = numpy.array(sigatm, "f")
                structure_builder.set_sigatm(sigatm_array)
            local_line_counter += 1
        # EOF (does not end in END or CONECT)
        self.line_counter = self.line_counter + local_line_counter
        return []

    def _handle_PDB_exception(self, message, line_counter):
        """
        This method catches an exception that occurs in the StructureBuilder
        object (if PERMISSIVE), or raises it again, this time adding the
        PDB line number to the error message.
        """
        message = "%s at line %i." % (message, line_counter)
        if self.PERMISSIVE:
            # just print a warning - some residues/atoms may be missing
            warnings.warn(
                "PDBConstructionException: %s\n"
                "Exception ignored.\n"
                "Some atoms or residues may be missing in the data structure." % message,
                PDBConstructionWarning,
            )
        else:
            # exceptions are fatal - raise again with new message (including line nr)
            raise PDBConstructionException(message)
コード例 #5
0
ファイル: MMCIFParser.py プロジェクト: cbirdlab/sap
 def get_structure(self, structure_id, filename):
     self._mmcif_dict = MMCIF2Dict(filename)
     self._structure_builder = StructureBuilder()
     self._build_structure(structure_id)
     return self._structure_builder.get_structure()
コード例 #6
0
ファイル: MMCIFParser.py プロジェクト: cbirdlab/sap
class MMCIFParser(object):
    def get_structure(self, structure_id, filename):
        self._mmcif_dict = MMCIF2Dict(filename)
        self._structure_builder = StructureBuilder()
        self._build_structure(structure_id)
        return self._structure_builder.get_structure()

    def _build_structure(self, structure_id):
        mmcif_dict = self._mmcif_dict
        atom_id_list = mmcif_dict["_atom_site.label_atom_id"]
        residue_id_list = mmcif_dict["_atom_site.label_comp_id"]
        try:
            element_list = mmcif_dict["_atom_site.type_symbol"]
        except KeyError:
            element_list = None
        seq_id_list = mmcif_dict["_atom_site.label_seq_id"]
        chain_id_list = mmcif_dict["_atom_site.label_asym_id"]
        x_list = [float(x) for x in mmcif_dict["_atom_site.Cartn_x"]]
        y_list = [float(x) for x in mmcif_dict["_atom_site.Cartn_y"]]
        z_list = [float(x) for x in mmcif_dict["_atom_site.Cartn_z"]]
        alt_list = mmcif_dict["_atom_site.label_alt_id"]
        b_factor_list = mmcif_dict["_atom_site.B_iso_or_equiv"]
        occupancy_list = mmcif_dict["_atom_site.occupancy"]
        fieldname_list = mmcif_dict["_atom_site.group_PDB"]
        try:
            serial_list = [
                int(n) for n in mmcif_dict["_atom_site.pdbx_PDB_model_num"]
            ]
        except KeyError:
            # No model number column
            serial_list = None
        except ValueError:
            # Invalid model number (malformed file)
            raise PDBConstructionException("Invalid model number")
        try:
            aniso_u11 = mmcif_dict["_atom_site.aniso_U[1][1]"]
            aniso_u12 = mmcif_dict["_atom_site.aniso_U[1][2]"]
            aniso_u13 = mmcif_dict["_atom_site.aniso_U[1][3]"]
            aniso_u22 = mmcif_dict["_atom_site.aniso_U[2][2]"]
            aniso_u23 = mmcif_dict["_atom_site.aniso_U[2][3]"]
            aniso_u33 = mmcif_dict["_atom_site.aniso_U[3][3]"]
            aniso_flag = 1
        except KeyError:
            # no anisotropic B factors
            aniso_flag = 0
        # if auth_seq_id is present, we use this.
        # Otherwise label_seq_id is used.
        if "_atom_site.auth_seq_id" in mmcif_dict:
            seq_id_list = mmcif_dict["_atom_site.auth_seq_id"]
        else:
            seq_id_list = mmcif_dict["_atom_site.label_seq_id"]
        # Now loop over atoms and build the structure
        current_chain_id = None
        current_residue_id = None
        structure_builder = self._structure_builder
        structure_builder.init_structure(structure_id)
        structure_builder.init_seg(" ")
        # Historically, Biopython PDB parser uses model_id to mean array index
        # so serial_id means the Model ID specified in the file
        current_model_id = 0
        current_serial_id = 0
        for i in range(0, len(atom_id_list)):
            x = x_list[i]
            y = y_list[i]
            z = z_list[i]
            resname = residue_id_list[i]
            chainid = chain_id_list[i]
            altloc = alt_list[i]
            if altloc == ".":
                altloc = " "
            resseq = seq_id_list[i]
            name = atom_id_list[i]
            # occupancy & B factor
            try:
                tempfactor = float(b_factor_list[i])
            except ValueError:
                raise PDBConstructionException("Invalid or missing B factor")
            try:
                occupancy = float(occupancy_list[i])
            except ValueError:
                raise PDBConstructionException("Invalid or missing occupancy")
            fieldname = fieldname_list[i]
            if fieldname == "HETATM":
                hetatm_flag = "H"
            else:
                hetatm_flag = " "
            if serial_list is not None:
                # model column exists; use it
                serial_id = serial_list[i]
                if current_serial_id != serial_id:
                    # if serial changes, update it and start new model
                    current_serial_id = serial_id
                    structure_builder.init_model(current_model_id,
                                                 current_serial_id)
                    current_model_id += 1
            else:
                # no explicit model column; initialize single model
                structure_builder.init_model(current_model_id)
            if current_chain_id != chainid:
                current_chain_id = chainid
                structure_builder.init_chain(current_chain_id)
                current_residue_id = resseq
                icode, int_resseq = self._get_icode(resseq)
                structure_builder.init_residue(resname, hetatm_flag,
                                               int_resseq, icode)
            elif current_residue_id != resseq:
                current_residue_id = resseq
                icode, int_resseq = self._get_icode(resseq)
                structure_builder.init_residue(resname, hetatm_flag,
                                               int_resseq, icode)
            coord = numpy.array((x, y, z), 'f')
            element = element_list[i] if element_list else None
            structure_builder.init_atom(name,
                                        coord,
                                        tempfactor,
                                        occupancy,
                                        altloc,
                                        name,
                                        element=element)
            if aniso_flag == 1:
                u = (aniso_u11[i], aniso_u12[i], aniso_u13[i], aniso_u22[i],
                     aniso_u23[i], aniso_u33[i])
                mapped_anisou = [float(x) for x in u]
                anisou_array = numpy.array(mapped_anisou, 'f')
                structure_builder.set_anisou(anisou_array)
        # Now try to set the cell
        try:
            a = float(mmcif_dict["_cell.length_a"])
            b = float(mmcif_dict["_cell.length_b"])
            c = float(mmcif_dict["_cell.length_c"])
            alpha = float(mmcif_dict["_cell.angle_alpha"])
            beta = float(mmcif_dict["_cell.angle_beta"])
            gamma = float(mmcif_dict["_cell.angle_gamma"])
            cell = numpy.array((a, b, c, alpha, beta, gamma), 'f')
            spacegroup = mmcif_dict["_symmetry.space_group_name_H-M"]
            spacegroup = spacegroup[1:-1]  # get rid of quotes!!
            if spacegroup is None:
                raise Exception
            structure_builder.set_symmetry(spacegroup, cell)
        except:
            pass  # no cell found, so just ignore

    def _get_icode(self, resseq):
        """Tries to return the icode. In MMCIF files this is just part of
        resseq! In PDB files, it's a separate field."""
        last_resseq_char = resseq[-1]
        if last_resseq_char in ascii_letters:
            icode = last_resseq_char
            int_resseq = int(resseq[0:-1])
        else:
            icode = " "
            int_resseq = int(resseq)
        return icode, int_resseq
コード例 #7
0
ファイル: MMCIFParser.py プロジェクト: kaspermunch/sap
 def get_structure(self, structure_id, filename):
     self._mmcif_dict=MMCIF2Dict(filename)
     self._structure_builder=StructureBuilder()
     self._build_structure(structure_id)
     return self._structure_builder.get_structure()
コード例 #8
0
ファイル: MMCIFParser.py プロジェクト: kaspermunch/sap
class MMCIFParser(object):
    def get_structure(self, structure_id, filename):
        self._mmcif_dict=MMCIF2Dict(filename)
        self._structure_builder=StructureBuilder()
        self._build_structure(structure_id)
        return self._structure_builder.get_structure()

    def _build_structure(self, structure_id):
        mmcif_dict=self._mmcif_dict
        atom_id_list=mmcif_dict["_atom_site.label_atom_id"]
        residue_id_list=mmcif_dict["_atom_site.label_comp_id"]
        try:
            element_list = mmcif_dict["_atom_site.type_symbol"]
        except KeyError:
            element_list = None
        seq_id_list=mmcif_dict["_atom_site.label_seq_id"]
        chain_id_list=mmcif_dict["_atom_site.label_asym_id"]
        x_list = [float(x) for x in mmcif_dict["_atom_site.Cartn_x"]]
        y_list = [float(x) for x in mmcif_dict["_atom_site.Cartn_y"]]
        z_list = [float(x) for x in mmcif_dict["_atom_site.Cartn_z"]]
        alt_list=mmcif_dict["_atom_site.label_alt_id"]
        b_factor_list=mmcif_dict["_atom_site.B_iso_or_equiv"]
        occupancy_list=mmcif_dict["_atom_site.occupancy"]
        fieldname_list=mmcif_dict["_atom_site.group_PDB"]
        try:
            serial_list = [int(n) for n in mmcif_dict["_atom_site.pdbx_PDB_model_num"]]
        except KeyError:
            # No model number column
            serial_list = None
        except ValueError:
            # Invalid model number (malformed file)
            raise PDBConstructionException("Invalid model number")
        try:
            aniso_u11=mmcif_dict["_atom_site.aniso_U[1][1]"]
            aniso_u12=mmcif_dict["_atom_site.aniso_U[1][2]"]
            aniso_u13=mmcif_dict["_atom_site.aniso_U[1][3]"]
            aniso_u22=mmcif_dict["_atom_site.aniso_U[2][2]"]
            aniso_u23=mmcif_dict["_atom_site.aniso_U[2][3]"]
            aniso_u33=mmcif_dict["_atom_site.aniso_U[3][3]"]
            aniso_flag=1
        except KeyError:
            # no anisotropic B factors
            aniso_flag=0
        # if auth_seq_id is present, we use this.
        # Otherwise label_seq_id is used.
        if "_atom_site.auth_seq_id" in mmcif_dict:
            seq_id_list=mmcif_dict["_atom_site.auth_seq_id"]
        else:
            seq_id_list=mmcif_dict["_atom_site.label_seq_id"]
        # Now loop over atoms and build the structure
        current_chain_id=None
        current_residue_id=None
        structure_builder=self._structure_builder
        structure_builder.init_structure(structure_id)
        structure_builder.init_seg(" ")
        # Historically, Biopython PDB parser uses model_id to mean array index
        # so serial_id means the Model ID specified in the file
        current_model_id = 0
        current_serial_id = 0
        for i in range(0, len(atom_id_list)):
            x=x_list[i]
            y=y_list[i]
            z=z_list[i]
            resname=residue_id_list[i]
            chainid=chain_id_list[i]
            altloc=alt_list[i]
            if altloc==".":
                altloc=" "
            resseq=seq_id_list[i]
            name=atom_id_list[i]
            # occupancy & B factor
            try:
                tempfactor=float(b_factor_list[i])
            except ValueError:
                raise PDBConstructionException("Invalid or missing B factor")
            try:
                occupancy=float(occupancy_list[i])
            except ValueError:
                raise PDBConstructionException("Invalid or missing occupancy")
            fieldname=fieldname_list[i]
            if fieldname=="HETATM":
                hetatm_flag="H"
            else:
                hetatm_flag=" "
            if serial_list is not None:
                # model column exists; use it
                serial_id = serial_list[i]
                if current_serial_id != serial_id:
                    # if serial changes, update it and start new model
                    current_serial_id = serial_id
                    structure_builder.init_model(current_model_id, current_serial_id)
                    current_model_id += 1
            else:
                # no explicit model column; initialize single model
                structure_builder.init_model(current_model_id)
            if current_chain_id!=chainid:
                current_chain_id=chainid
                structure_builder.init_chain(current_chain_id)
                current_residue_id=resseq
                icode, int_resseq=self._get_icode(resseq)
                structure_builder.init_residue(resname, hetatm_flag, int_resseq,
                    icode)
            elif current_residue_id!=resseq:
                current_residue_id=resseq
                icode, int_resseq=self._get_icode(resseq)
                structure_builder.init_residue(resname, hetatm_flag, int_resseq,
                    icode)
            coord=numpy.array((x, y, z), 'f')
            element = element_list[i] if element_list else None
            structure_builder.init_atom(name, coord, tempfactor, occupancy, altloc,
                name, element=element)
            if aniso_flag==1:
                u=(aniso_u11[i], aniso_u12[i], aniso_u13[i],
                    aniso_u22[i], aniso_u23[i], aniso_u33[i])
                mapped_anisou = [float(x) for x in u]
                anisou_array=numpy.array(mapped_anisou, 'f')
                structure_builder.set_anisou(anisou_array)
        # Now try to set the cell
        try:
            a=float(mmcif_dict["_cell.length_a"])
            b=float(mmcif_dict["_cell.length_b"])
            c=float(mmcif_dict["_cell.length_c"])
            alpha=float(mmcif_dict["_cell.angle_alpha"])
            beta=float(mmcif_dict["_cell.angle_beta"])
            gamma=float(mmcif_dict["_cell.angle_gamma"])
            cell=numpy.array((a, b, c, alpha, beta, gamma), 'f')
            spacegroup=mmcif_dict["_symmetry.space_group_name_H-M"]
            spacegroup=spacegroup[1:-1] # get rid of quotes!!
            if spacegroup is None:
                raise Exception
            structure_builder.set_symmetry(spacegroup, cell)
        except:
            pass    # no cell found, so just ignore

    def _get_icode(self, resseq):
        """Tries to return the icode. In MMCIF files this is just part of
        resseq! In PDB files, it's a separate field."""
        last_resseq_char=resseq[-1]
        if last_resseq_char in ascii_letters:
            icode=last_resseq_char
            int_resseq=int(resseq[0:-1])
        else:
            icode=" "
            int_resseq=int(resseq)
        return icode, int_resseq
コード例 #9
0
ファイル: PDBIO.py プロジェクト: cbirdlab/sap
    def set_structure(self, pdb_object):
        # Check what the user is providing and build a structure appropriately
        if pdb_object.level == "S":
            structure = pdb_object
        else:
            sb = StructureBuilder()
            sb.init_structure('pdb')
            sb.init_seg(' ')
            # Build parts as necessary
            if pdb_object.level == "M":
                sb.structure.add(pdb_object)
                self.structure = sb.structure
            else:
                sb.init_model(0)
                if pdb_object.level == "C":
                    sb.structure[0].add(pdb_object)
                else:
                    sb.init_chain('A')
                    if pdb_object.level == "R":
                        try:
                            parent_id = pdb_object.parent.id
                            sb.structure[0]['A'].id = parent_id
                        except Exception:
                            pass
                        sb.structure[0]['A'].add(pdb_object)
                    else:
                        # Atom
                        sb.init_residue('DUM', ' ', 1, ' ')
                        try:
                            parent_id = pdb_object.parent.parent.id
                            sb.structure[0]['A'].id = parent_id
                        except Exception:
                            pass
                        sb.structure[0]['A'].child_list[0].add(pdb_object)

            # Return structure
            structure = sb.structure
        self.structure = structure
コード例 #10
0
ファイル: PDBParser.py プロジェクト: cbirdlab/sap
class PDBParser(object):
    """
    Parse a PDB file and return a Structure object.
    """

    def __init__(self, PERMISSIVE=True, get_header=False,
                 structure_builder=None, QUIET=False):
        """
        The PDB parser call a number of standard methods in an aggregated
        StructureBuilder object. Normally this object is instanciated by the
        PDBParser object itself, but if the user provides his/her own
        StructureBuilder object, the latter is used instead.

        Arguments:

        o PERMISSIVE - Evaluated as a Boolean. If false, exceptions in
        constructing the SMCRA data structure are fatal. If true (DEFAULT),
        the exceptions are caught, but some residues or atoms will be missing.
        THESE EXCEPTIONS ARE DUE TO PROBLEMS IN THE PDB FILE!.

        o structure_builder - an optional user implemented StructureBuilder class.

        o QUIET - Evaluated as a Boolean. If true, warnings issued in constructing
        the SMCRA data will be suppressed. If false (DEFAULT), they will be shown.
        These warnings might be indicative of problems in the PDB file!
        """
        if structure_builder is not None:
            self.structure_builder = structure_builder
        else:
            self.structure_builder = StructureBuilder()
        self.header = None
        self.trailer = None
        self.line_counter = 0
        self.PERMISSIVE = bool(PERMISSIVE)
        self.QUIET = bool(QUIET)

    # Public methods

    def get_structure(self, id, file):
        """Return the structure.

        Arguments:
        o id - string, the id that will be used for the structure
        o file - name of the PDB file OR an open filehandle
        """
        with warnings.catch_warnings():
            if self.QUIET:
                warnings.filterwarnings("ignore", category=PDBConstructionWarning)

            self.header = None
            self.trailer = None
            # Make a StructureBuilder instance (pass id of structure as parameter)
            self.structure_builder.init_structure(id)

            with as_handle(file) as handle:
                self._parse(handle.readlines())

            self.structure_builder.set_header(self.header)
            # Return the Structure instance
            structure = self.structure_builder.get_structure()

        return structure

    def get_header(self):
        "Return the header."
        return self.header

    def get_trailer(self):
        "Return the trailer."
        return self.trailer

    # Private methods

    def _parse(self, header_coords_trailer):
        "Parse the PDB file."
        # Extract the header; return the rest of the file
        self.header, coords_trailer = self._get_header(header_coords_trailer)
        # Parse the atomic data; return the PDB file trailer
        self.trailer = self._parse_coordinates(coords_trailer)

    def _get_header(self, header_coords_trailer):
        "Get the header of the PDB file, return the rest."
        structure_builder = self.structure_builder
        i = 0
        for i in range(0, len(header_coords_trailer)):
            structure_builder.set_line_counter(i + 1)
            line = header_coords_trailer[i]
            record_type = line[0:6]
            if record_type == "ATOM  " or record_type == "HETATM" or record_type == "MODEL ":
                break
        header = header_coords_trailer[0:i]
        # Return the rest of the coords+trailer for further processing
        self.line_counter = i
        coords_trailer = header_coords_trailer[i:]
        header_dict = _parse_pdb_header_list(header)
        return header_dict, coords_trailer

    def _parse_coordinates(self, coords_trailer):
        "Parse the atomic data in the PDB file."
        local_line_counter = 0
        structure_builder = self.structure_builder
        current_model_id = 0
        # Flag we have an open model
        model_open = 0
        current_chain_id = None
        current_segid = None
        current_residue_id = None
        current_resname = None
        for i in range(0, len(coords_trailer)):
            line = coords_trailer[i]
            record_type = line[0:6]
            global_line_counter = self.line_counter + local_line_counter + 1
            structure_builder.set_line_counter(global_line_counter)
            if record_type == "ATOM  " or record_type == "HETATM":
                # Initialize the Model - there was no explicit MODEL record
                if not model_open:
                    structure_builder.init_model(current_model_id)
                    current_model_id += 1
                    model_open = 1
                fullname = line[12:16]
                # get rid of whitespace in atom names
                split_list = fullname.split()
                if len(split_list) != 1:
                    # atom name has internal spaces, e.g. " N B ", so
                    # we do not strip spaces
                    name = fullname
                else:
                    # atom name is like " CA ", so we can strip spaces
                    name = split_list[0]
                altloc = line[16]
                resname = line[17:20]
                chainid = line[21]
                try:
                    serial_number = int(line[6:11])
                except:
                    serial_number = 0
                resseq = int(line[22:26].split()[0])  # sequence identifier
                icode = line[26]  # insertion code
                if record_type == "HETATM":  # hetero atom flag
                    if resname == "HOH" or resname == "WAT":
                        hetero_flag = "W"
                    else:
                        hetero_flag = "H"
                else:
                    hetero_flag = " "
                residue_id = (hetero_flag, resseq, icode)
                # atomic coordinates
                try:
                    x = float(line[30:38])
                    y = float(line[38:46])
                    z = float(line[46:54])
                except:
                    # Should we allow parsing to continue in permissive mode?
                    # If so, what coordinates should we default to?  Easier to abort!
                    raise PDBConstructionException("Invalid or missing coordinate(s) at line %i."
                                                   % global_line_counter)
                coord = numpy.array((x, y, z), "f")
                # occupancy & B factor
                try:
                    occupancy = float(line[54:60])
                except:
                    self._handle_PDB_exception("Invalid or missing occupancy",
                                               global_line_counter)
                    occupancy = None # Rather than arbitrary zero or one
                try:
                    bfactor = float(line[60:66])
                except:
                    self._handle_PDB_exception("Invalid or missing B factor",
                                               global_line_counter)
                    bfactor = 0.0  # The PDB use a default of zero if the data is missing
                segid = line[72:76]
                element = line[76:78].strip()
                if current_segid != segid:
                    current_segid = segid
                    structure_builder.init_seg(current_segid)
                if current_chain_id != chainid:
                    current_chain_id = chainid
                    structure_builder.init_chain(current_chain_id)
                    current_residue_id = residue_id
                    current_resname = resname
                    try:
                        structure_builder.init_residue(resname, hetero_flag, resseq, icode)
                    except PDBConstructionException as message:
                        self._handle_PDB_exception(message, global_line_counter)
                elif current_residue_id != residue_id or current_resname != resname:
                    current_residue_id = residue_id
                    current_resname = resname
                    try:
                        structure_builder.init_residue(resname, hetero_flag, resseq, icode)
                    except PDBConstructionException as message:
                        self._handle_PDB_exception(message, global_line_counter)
                # init atom
                try:
                    structure_builder.init_atom(name, coord, bfactor, occupancy, altloc,
                                                fullname, serial_number, element)
                except PDBConstructionException as message:
                    self._handle_PDB_exception(message, global_line_counter)
            elif record_type == "ANISOU":
                anisou = [float(x) for x in (line[28:35], line[35:42], line[43:49],
                                             line[49:56], line[56:63], line[63:70])]
                # U's are scaled by 10^4
                anisou_array = (numpy.array(anisou, "f") / 10000.0).astype("f")
                structure_builder.set_anisou(anisou_array)
            elif record_type == "MODEL ":
                try:
                    serial_num = int(line[10:14])
                except:
                    self._handle_PDB_exception("Invalid or missing model serial number",
                                               global_line_counter)
                    serial_num = 0
                structure_builder.init_model(current_model_id, serial_num)
                current_model_id += 1
                model_open = 1
                current_chain_id = None
                current_residue_id = None
            elif record_type == "END   " or record_type == "CONECT":
                # End of atomic data, return the trailer
                self.line_counter += local_line_counter
                return coords_trailer[local_line_counter:]
            elif record_type == "ENDMDL":
                model_open = 0
                current_chain_id = None
                current_residue_id = None
            elif record_type == "SIGUIJ":
                # standard deviation of anisotropic B factor
                siguij = [float(x) for x in (line[28:35], line[35:42], line[42:49],
                                             line[49:56], line[56:63], line[63:70])]
                # U sigma's are scaled by 10^4
                siguij_array = (numpy.array(siguij, "f") / 10000.0).astype("f")
                structure_builder.set_siguij(siguij_array)
            elif record_type == "SIGATM":
                # standard deviation of atomic positions
                sigatm = [float(x) for x in (line[30:38], line[38:45], line[46:54],
                                             line[54:60], line[60:66])]
                sigatm_array = numpy.array(sigatm, "f")
                structure_builder.set_sigatm(sigatm_array)
            local_line_counter += 1
        # EOF (does not end in END or CONECT)
        self.line_counter = self.line_counter + local_line_counter
        return []

    def _handle_PDB_exception(self, message, line_counter):
        """
        This method catches an exception that occurs in the StructureBuilder
        object (if PERMISSIVE), or raises it again, this time adding the
        PDB line number to the error message.
        """
        message = "%s at line %i." % (message, line_counter)
        if self.PERMISSIVE:
            # just print a warning - some residues/atoms may be missing
            warnings.warn("PDBConstructionException: %s\n"
                          "Exception ignored.\n"
                          "Some atoms or residues may be missing in the data structure."
                          % message, PDBConstructionWarning)
        else:
            # exceptions are fatal - raise again with new message (including line nr)
            raise PDBConstructionException(message)