Source Code for Module biana.BianaParser.keggKOParser

  1  """ 
  2      BIANA: Biologic Interactions and Network Analysis 
  3      Copyright (C) 2009  Javier Garcia-Garcia, Emre Guney, Baldo Oliva 
  4   
  5      This program is free software: you can redistribute it and/or modify 
  6      it under the terms of the GNU General Public License as published by 
  7      the Free Software Foundation, either version 3 of the License, or 
  8      (at your option) any later version. 
  9   
 10      This program is distributed in the hope that it will be useful, 
 11      but WITHOUT ANY WARRANTY; without even the implied warranty of 
 12      MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the 
 13      GNU General Public License for more details. 
 14   
 15      You should have received a copy of the GNU General Public License 
 16      along with this program.  If not, see <http://www.gnu.org/licenses/>. 
 17   
 18  """ 
 19   
 20  """ 
 21  File        : keggkoParser.py 
 22  Author      : Javier Garcia Garcia 
 23  Creation    : January 2008 
 24  Contents    : fills up tables in database biana with information from kegg ko database 
 25  Called from :  
 26   
 27  ======================================================================================================= 
 28   
 29  This file implements a program that fills up tables in database biana with information of kegg ko databases 
 30   
 31  """ 
 32   
 33  from bianaParser import * 
 34   
 35   
 36 -class KeggKOParser(BianaParser): 
 37      """ 
 38   
 39      """ 
 40   
 41      name = "kegg_kO" 
 42      description = "This file implements a program that fills up tables in database biana with information of kegg KO Database" 
 43      external_entity_definition = "A external entity represents a KEGG KO" 
 44      external_entity_relations = "" 
 45   
 46 -    def __init__(self): 
 47   
 48          # Start with the default values 
 49   
 50          BianaParser.__init__(self, default_db_description = "KEGG KO database", 
 51                               default_script_name = "keggKOParser.py", 
 52                               default_script_description = KeggKOParser.description ) 
 53          self.default_eE_attribute = "keggCode" 
 54          self.initialize_input_file_descriptor() 
 55   
 56 -    def parse_database(self): 
 57          """ 
 58          """ 
 59   
 60          self.initialize_input_file_descriptor() 
 61   
 62          # General regex 
 63          continue_field_regex = re.compile("^\s{3,}([^;]+);*$") 
 64          field_regex = re.compile("^(\w+)\s+([^;]+);*$") 
 65          pathway_regex = re.compile("PATH\:\s+(map|rn)(\d+)\s+(.+)$") 
 66   
 67          space_regex = re.compile("\s+") 
 68          parenthesis_regex = re.compile("\(.+\)")  # used to eliminate extra information in sequence 
 69           
 70   
 71          entry_regex = re.compile("^ENTRY\s+(\w+)\s+KO") 
 72   
 73          dblink_split_regex = re.compile("(\w+)\:") 
 74   
 75          kegg_ko_object = None 
 76   
 77          temp_value = []           # List used to store the information of those fields that can have more than a single line 
 78          current_field = None 
 79           
 80          for line in self.input_file_fd: 
 81   
 82              m = entry_regex.match(line) 
 83   
 84              if m: 
 85                  if kegg_ko_object is not None: 
 86                      self.biana_access.insert_new_external_entity( externalEntity = kegg_ko_object ) 
 87                   
 88                  kegg_ko_object = ExternalEntityRelation( source_database = self.database, relation_type="cluster" ) 
 89                  kegg_ko_object.add_attribute( ExternalEntityAttribute( attribute_identifier = "keggCode", value=m.group(1), type = "unique" ) ) 
 90                  continue 
 91   
 92   
 93              new_field = field_regex.match(line) 
 94              if new_field: 
 95                  if current_field == "DBLINK": 
 96                      all_db_links = " ".join(temp_value) 
 97                      list_db_links = [ x.strip() for x in dblink_split_regex.split(all_db_links) ] 
 98   
 99                      for actual_position in xrange(len(list_db_links)): 
100                          if list_db_links[actual_position] == "EC": 
101                              [ kegg_ko_object.add_attribute(ExternalEntityAttribute( attribute_identifier = "EC", value=x, type="cross-reference")) for x in list_db_links[actual_position+1].split(" ") ] 
102                               
103                          elif list_db_links[actual_position] == "COG": 
104                              [ kegg_ko_object.add_attribute(ExternalEntityAttribute( attribute_identifier = "COG", value=x,type="cross-reference")) for x in list_db_links[actual_position+1].split(" ") ] 
105                               
106                          elif list_db_links[actual_position] == "GO": 
107                              [ kegg_ko_object.add_attribute(ExternalEntityAttribute( attribute_identifier = "GO", value=x,type="cross-reference")) for x in list_db_links[actual_position+1].split(" ") ] 
108                               
109                  current_field = new_field.group(1) 
110                  temp_value = [new_field.group(2)] 
111              else: 
112                  cont_value = continue_field_regex.match(line) 
113                  if cont_value: 
114                      temp_value.append(cont_value.group(1)) 
115   
116                   
117          # Insert the last one 
118          if kegg_ko_object is not None: 
119              self.biana_access.insert_new_external_entity( externalEntity = kegg_ko_object ) 
120