view lib/sequence.py @ 158:2ed0c183867a
fix. delete sequence.pdb_files on pdb_unload. however memory leak remains
| author |
boris <bnagaev@gmail.com> |
| date |
Thu, 28 Oct 2010 08:14:23 +0400 |
| parents |
9a42bc8d5198 |
| children |
fa6ef3e58128 |
line source
2 # -*- coding: utf-8 -*-
4 from monomer import AminoAcidType
5 from Bio.PDB import CaPPBuilder, PDBIO
6 from Bio.PDB.DSSP import make_dssp_dict
7 from allpy_pdb import std_id, pdb_id_parse, get_structure
13 from tempfile import NamedTemporaryFile
17 class Sequence(object):
18 """ Sequence of Monomers
21 * name -- str with the name of sequence
22 * description -- str with description of the sequence
23 * monomers -- list of monomer objects (aminoacids or nucleotides)
24 * pdb_chains -- list of Bio.PDB.Chain's
25 * pdb_files -- dictionary like {Bio.PDB.Chain: file_obj}
27 * pdb_residues -- dictionary like {Bio.PDB.Chain: {Monomer: Bio.PDB.Residue}}
28 * pdb_secstr -- dictionary like {Bio.PDB.Chain: {Monomer: 'Secondary structure'}}
29 Code Secondary structure
31 B Isolated ??-bridge residue
40 ?TODO: global pdb_structures
42 def __init__(self, monomers=None, name='', description=""):
46 self.description = description
47 self.monomers = monomers
50 self.pdb_residues = {}
54 return len(self.monomers)
57 """ Returns sequence in one-letter code """
58 return ''.join([monomer.type.code1 for monomer in self.monomers])
60 def __eq__(self, other):
61 """ Returns if all corresponding monomers of this sequences are equal
63 If lengths of sequences are not equal, returns False
65 return len(self) == len(other) and \
66 all([a==b for a, b in zip(self.monomers, other.monomers)])
68 def __ne__(self, other):
69 return not (self == other)
71 def pdb_chain_add(self, pdb_file, pdb_id, pdb_chain, pdb_model=0):
72 """ Reads Pdb chain from file
74 and align each Monomer with PDB.Residue (TODO)
76 name = std_id(pdb_id, pdb_chain, pdb_model)
77 structure = get_structure(pdb_file, name)
78 chain = structure[pdb_model][pdb_chain]
79 self.pdb_chains.append(chain)
80 self.pdb_residues[chain] = {}
81 self.pdb_secstr[chain] = {}
82 pdb_sequence = Sequence.from_pdb_chain(chain)
83 alignment = project.Project.from_sequences(self, pdb_sequence)
84 alignment.muscle_align()
85 for monomer, pdb_monomer in alignment.column(sequence=pdb_sequence, original=self):
86 if pdb_sequence.pdb_has(chain, pdb_monomer):
87 residue = pdb_sequence.pdb_residues[chain][pdb_monomer]
88 self.pdb_residues[chain][monomer] = residue
89 self.pdb_files[chain] = pdb_file
92 """ Delete all pdb-connected links """
94 self.pdb_residues = {}
95 self.pdb_secstr = {} # FIXME
96 self.pdb_files = {} # FIXME
99 def from_str(fasta_str, name='', description='', monomer_kind=AminoAcidType):
100 """ Import data from one-letter code
102 monomer_kind is class, inherited from MonomerType
104 monomers = [monomer_kind.from_code1(aa).instance() for aa in fasta_str]
105 return Sequence(monomers, name, description)
108 def from_pdb_chain(chain):
109 """ Returns Sequence with Monomers with link to Bio.PDB.Residue
111 chain is Bio.PDB.Chain
113 cappbuilder = CaPPBuilder()
114 peptides = cappbuilder.build_peptides(chain)
115 sequence = Sequence()
116 sequence.pdb_chains = [chain]
117 sequence.pdb_residues[chain] = {}
118 sequence.pdb_secstr[chain] = {}
119 for peptide in peptides:
120 for ca_atom in peptide.get_ca_list():
121 residue = ca_atom.get_parent()
122 monomer = AminoAcidType.from_pdb_residue(residue).instance()
123 sequence.pdb_residues[chain][monomer] = residue
124 sequence.monomers.append(monomer)
127 def pdb_auto_add(self, conformity_info=None, pdb_directory='./pdb'):
128 """ Adds pdb information to each monomer
130 Returns if information has been successfully added
131 TODO: conformity_file
135 if not conformity_info:
136 path = os.path.join(pdb_directory, self.name)
137 if os.path.exists(path):
138 match = pdb_id_parse(self.name)
139 self.pdb_chain_add(open(path), match['code'],
140 match['chain'], match['model'])
142 match = pdb_id_parse(self.name)
145 pdb_filename = config.pdb_dir % code
146 if not os.path.exists(pdb_filename):
147 url = config.pdb_url % code
148 print "Download %s" % url
149 pdb_file = open(pdb_filename, 'w')
150 data = urllib2.urlopen(url).read()
153 print "Save %s" % pdb_filename
154 pdb_file = open(pdb_filename)
155 self.pdb_chain_add(pdb_file, code, match['chain'], match['model'])
157 def pdb_save(self, out_filename, pdb_chain):
158 """ Saves pdb_chain to out_file """
159 class GlySelect(Select):
160 def accept_chain(self, chain):
161 if chain == pdb_chain:
166 structure = chain.get_parent()
167 io.set_structure(structure)
168 io.save(out_filename, GlySelect())
171 def pdb_add_sec_str(self, pdb_chain):
172 """ Add secondary structure data """
173 tmp_file = NamedTemporaryFile(delete=False)
175 pdb_file = self.pdb_files[pdb_chain].name
176 os.system("dsspcmbi %(pdb)s %(tmp)s" % {'pdb': pdb_file, 'tmp': tmp_file.name})
177 dssp, keys = make_dssp_dict(tmp_file.name)
178 for monomer in self.monomers:
179 if self.pdb_has(pdb_chain, monomer):
180 residue = self.pdb_residues[pdb_chain][monomer]
182 d = dssp[(pdb_chain.get_id(), residue.get_id())]
183 self.pdb_secstr[pdb_chain][monomer] = d[1]
185 print "No dssp information about %s at %s" % (monomer, pdb_chain)
186 os.unlink(tmp_file.name)
188 def pdb_has(self, chain, monomer):
189 return chain in self.pdb_residues and monomer in self.pdb_residues[chain]
191 def secstr_has(self, chain, monomer):
192 return chain in self.pdb_secstr and monomer in self.pdb_secstr[chain]
