Source Code for Module Bio.SCOP.Cla

  1  # Copyright 2001 by Gavin E. Crooks.  All rights reserved. 
  2  # Modifications Copyright 2010 Jeffrey Finkelstein. All rights reserved. 
  3  # 
  4  # This code is part of the Biopython distribution and governed by its 
  5  # license.  Please see the LICENSE file that should have been included 
  6  # as part of this package. 
  7   
  8  """ Handle the SCOP CLAssification file, which describes SCOP domains. 
  9   
 10  The file format is described in the scop 
 11  "release notes.":http://scop.mrc-lmb.cam.ac.uk/scop/release-notes.html 
 12  The latest CLA file can be found 
 13  "elsewhere at SCOP.":http://scop.mrc-lmb.cam.ac.uk/scop/parse/ 
 14   
 15  "Release 1.73": http://scop.mrc-lmb.cam.ac.uk/scop/parse/dir.cla.scop.txt_1.73 
 16  (July 2008) 
 17   
 18  """ 
 19   
 20  import Residues 
 21   
 22   
 23 -class Record(object): 
 24      """Holds information for one SCOP domain. 
 25   
 26      sid         --  SCOP identifier. e.g. d1danl2 
 27   
 28      residues    --  The domain definition as a Residues object 
 29   
 30      sccs        --  SCOP concise classification strings.  e.g. b.1.2.1 
 31   
 32      sunid       --  SCOP unique identifier for this domain 
 33   
 34      hierarchy   --  A dictionary, keys are nodetype, values are sunid, 
 35                      describing the location of this domain in the SCOP 
 36                      hierarchy. See the Scop module for a description of 
 37                      nodetypes. This used to be a list of (key,value) tuples 
 38                      in older versions of Biopython (see Bug 3109). 
 39      """ 
 40 -    def __init__(self, line=None): 
 41          self.sid = '' 
 42          self.residues = None 
 43          self.sccs = '' 
 44          self.sunid ='' 
 45          self.hierarchy = {} 
 46          if line: 
 47              self._process(line) 
 48   
 49 -    def _process(self, line): 
 50          line = line.rstrip()         # no trailing whitespace 
 51          columns = line.split('\t')   # separate the tab-delineated cols 
 52          if len(columns) != 6: 
 53              raise ValueError("I don't understand the format of %s" % line) 
 54   
 55          self.sid, pdbid, residues, self.sccs, self.sunid, hierarchy = columns 
 56          self.residues = Residues.Residues(residues) 
 57          self.residues.pdbid = pdbid 
 58          self.sunid = int(self.sunid) 
 59   
 60          for ht in hierarchy.split(","): 
 61              key, value = ht.split('=') 
 62              self.hierarchy[key] = int(value) 
 63   
 64 -    def __str__(self): 
 65          s = [] 
 66          s.append(self.sid) 
 67          s += str(self.residues).split(" ") 
 68          s.append(self.sccs) 
 69          s.append(self.sunid) 
 70   
 71          s.append(','.join('='.join((key, str(value))) for key, value 
 72                            in self.hierarchy.iteritems())) 
 73   
 74          return "\t".join(map(str,s)) + "\n" 
 75   
 76   
 77 -def parse(handle): 
 78      """Iterates over a CLA file, returning a Cla record for each line 
 79      in the file. 
 80   
 81      Arguments: 
 82   
 83          handle -- file-like object. 
 84      """ 
 85      for line in handle: 
 86          if line.startswith('#'): 
 87              continue 
 88          yield Record(line) 
 89   
 90   
 91 -class Index(dict): 
 92      """A CLA file indexed by SCOP identifiers, allowing rapid 
 93         random access into a file.""" 
 94 -    def __init__(self, filename): 
 95          """ 
 96          Arguments: 
 97   
 98            filename  -- The file to index 
 99          """ 
100          dict.__init__(self) 
101          self.filename = filename 
102          f = open(self.filename, "rU") 
103          try: 
104              position = 0 
105              while True: 
106                  line = f.readline() 
107                  if not line: 
108                      break 
109                  if line.startswith('#'): 
110                      continue 
111                  record = Record(line) 
112                  key = record.sid 
113                  if key is not None: 
114                      self[key] = position 
115                  position = f.tell() 
116          finally: 
117              f.close() 
118   
119 -    def __getitem__(self, key): 
120          """Return an item from the indexed file.""" 
121          position = dict.__getitem__(self,key) 
122   
123          f = open(self.filename, "rU") 
124          try: 
125              f.seek(position) 
126              line = f.readline() 
127              record = Record(line) 
128          finally: 
129              f.close() 
130          return record 
131