Package Bio :: Package PDB :: Module parse_pdb_header'
[hide private]
[frames] | no frames]

Source Code for Module Bio.PDB.parse_pdb_header'

  1  #!/usr/bin/env python 
  2  # 
  3  # parse_pdb_header.py 
  4  # parses header of PDB files into a python dictionary. 
  5  # emerged from the Columba database project www.columba-db.de. 
  6  #  
  7  # author: Kristian Rother 
  8  #  
  9  # license: same as BioPython, read LICENSE.TXT from current BioPython release. 
 10  #  
 11  # last modified: 9.2.2004 
 12  # 
 13  # Added some small changes: the whole PDB file is not read in anymore, but just 
 14  # until the first ATOM record (faster). I also split parse_pdb_header into  
 15  # parse_pdb_header and parse_pdb_header_list, because parse_pdb_header_list 
 16  # can be more easily reused in PDBParser. 
 17  # 
 18  # Thomas, 19/03/04 
 19  # 
 20  # Renamed some clearly private functions to _something (ie. parse_pdb_header_list 
 21  # is now _parse_pdb_header_list) 
 22  # Thomas 9/05/04 
 23   
 24  __doc__="Parse the header of a PDB file." 
 25   
 26  import sys 
 27  import os, re 
 28  import urllib 
 29  import types 
 30   
 31   
32 -def _get_journal(inl):
33 # JRNL AUTH L.CHEN,M.DOI,F.S.MATHEWS,A.Y.CHISTOSERDOV, 2BBK 7 34 journal="" 35 for l in inl: 36 if re.search("\AJRNL",l): 37 journal+=l[19:72].lower() 38 journal=re.sub("\s\s+"," ",journal) 39 return journal
40
41 -def _get_references(inl):
42 # REMARK 1 REFERENCE 1 1CSE 11 43 # REMARK 1 AUTH W.BODE,E.PAPAMOKOS,D.MUSIL 1CSE 12 44 references=[] 45 actref="" 46 for l in inl: 47 if re.search("\AREMARK 1",l): 48 if re.search("\AREMARK 1 REFERENCE",l): 49 if actref!="": 50 actref=re.sub("\s\s+"," ",actref) 51 if actref!=" ": 52 references.append(actref) 53 actref="" 54 else: 55 actref+=l[19:72].lower() 56 57 if actref!="": 58 actref=re.sub("\s\s+"," ",actref) 59 if actref!=" ": 60 references.append(actref) 61 return references
62 63 64 # bring dates to format: 1909-01-08
65 -def _format_date(pdb_date):
66 """Converts dates from DD-Mon-YY to YYYY-MM-DD format.""" 67 date="" 68 year=int(pdb_date[7:]) 69 if year<50: 70 century=2000 71 else: 72 century=1900 73 date=str(century+year)+"-" 74 all_months=['xxx','Jan','Feb','Mar','Apr','May','Jun','Jul',\ 75 'Aug','Sep','Oct','Nov','Dec'] 76 month=str(all_months.index(pdb_date[3:6])) 77 if len(month)==1: 78 month = '0'+month 79 date = date+month+'-'+pdb_date[:2] 80 return date
81 82
83 -def _chop_end_codes(line):
84 """Chops lines ending with ' 1CSA 14' and the like.""" 85 import re 86 return re.sub("\s\s\s\s+[\w]{4}.\s+\d*\Z","",line)
87
88 -def _chop_end_misc(line):
89 """Chops lines ending with ' 14-JUL-97 1CSA' and the like.""" 90 import re 91 return re.sub("\s\s\s\s+.*\Z","",line)
92
93 -def _nice_case(line):
94 """Makes A Lowercase String With Capitals.""" 95 l=line.lower() 96 s="" 97 i=0 98 nextCap=1 99 while i<len(l): 100 c=l[i] 101 if c>='a' and c<='z' and nextCap: 102 c=c.upper() 103 nextCap=0 104 elif c==' ' or c=='.' or c==',' or c==';' or c==':' or c=='\t' or\ 105 c=='-' or c=='_': 106 nextCap=1 107 s+=c 108 i+=1 109 return s
110
111 -def parse_pdb_header(file):
112 """ 113 Returns the header lines of a pdb file as a dictionary. 114 115 Dictionary keys are: head, deposition_date, release_date, structure_method, 116 resolution, structure_reference, journal_reference, author and 117 compound. 118 """ 119 header=[] 120 if type(file)==types.StringType: 121 f=open(file,'r') 122 else: 123 f=file 124 for l in f: 125 record_type=l[0:6] 126 if record_type=='ATOM ' or record_type=='HETATM' or record_type=='MODEL ': 127 break 128 else: 129 header.append(l) 130 f.close() 131 return _parse_pdb_header_list(header)
132
133 -def _parse_pdb_header_list(header):
134 # database fields 135 dict={'name':"", 136 'head':'', 137 'deposition_date' : "1909-01-08", 138 'release_date' : "1909-01-08", 139 'structure_method' : "unknown", 140 'resolution' : 0.0, 141 'structure_reference' : "unknown", 142 'journal_reference' : "unknown", 143 'author' : "", 144 'compound':{'1':{'misc':''}},'source':{'1':{'misc':''}}} 145 146 dict['structure_reference'] = _get_references(header) 147 dict['journal_reference'] = _get_journal(header) 148 comp_molid="1" 149 src_molid="1" 150 last_comp_key="misc" 151 last_src_key="misc" 152 153 for hh in header: 154 h=re.sub("[\s\n\r]*\Z","",hh) # chop linebreaks off 155 key=re.sub("\s.+\s*","",h) 156 tail=re.sub("\A\w+\s+\d*\s*","",h) 157 # print key+":"+tail 158 159 # From here, all the keys from the header are being parsed 160 if key=="TITLE": 161 name=_chop_end_codes(tail).lower() 162 if dict.has_key('name'): 163 dict['name'] += " "+name 164 else: 165 dict['name']=name 166 elif key=="HEADER": 167 rr=re.search("\d\d-\w\w\w-\d\d",tail) 168 if rr!=None: 169 dict['deposition_date']=_format_date(_nice_case(rr.group())) 170 head=_chop_end_misc(tail).lower() 171 dict['head']=head 172 elif key=="COMPND": 173 tt=re.sub("\;\s*\Z","",_chop_end_codes(tail)).lower() 174 # look for E.C. numbers in COMPND lines 175 rec = re.search('\d+\.\d+\.\d+\.\d+',tt) 176 if rec: 177 dict['compound'][comp_molid]['ec_number']=rec.group() 178 tt=re.sub("\((e\.c\.)*\d+\.\d+\.\d+\.\d+\)","",tt) 179 tok=tt.split(":") 180 if len(tok)>=2: 181 ckey=tok[0] 182 cval=re.sub("\A\s*","",tok[1]) 183 if ckey=='mol_id': 184 dict['compound'][cval]={'misc':''} 185 comp_molid=cval 186 last_comp_key="misc" 187 else: 188 dict['compound'][comp_molid][ckey]=cval 189 last_comp_key=ckey 190 else: 191 dict['compound'][comp_molid][last_comp_key]+=tok[0]+" " 192 elif key=="SOURCE": 193 tt=re.sub("\;\s*\Z","",_chop_end_codes(tail)).lower() 194 tok=tt.split(":") 195 # print tok 196 if len(tok)>=2: 197 ckey=tok[0] 198 cval=re.sub("\A\s*","",tok[1]) 199 if ckey=='mol_id': 200 dict['source'][cval]={'misc':''} 201 comp_molid=cval 202 last_src_key="misc" 203 else: 204 dict['source'][comp_molid][ckey]=cval 205 last_src_key=ckey 206 else: 207 dict['source'][comp_molid][last_src_key]+=tok[0]+" " 208 elif key=="KEYWDS": 209 kwd=_chop_end_codes(tail).lower() 210 if dict.has_key('keywords'): 211 dict['keywords']+=" "+kwd 212 else: 213 dict['keywords']=kwd 214 elif key=="EXPDTA": 215 expd=_chop_end_codes(tail) 216 # chop junk at end of lines for some structures 217 expd=re.sub('\s\s\s\s\s\s\s.*\Z','',expd) 218 # if re.search('\Anmr',expd,re.IGNORECASE): expd='nmr' 219 # if re.search('x-ray diffraction',expd,re.IGNORECASE): expd='x-ray diffraction' 220 dict['structure_method']=expd.lower() 221 elif key=="CAVEAT": 222 # make Annotation entries out of these!!! 223 pass 224 elif key=="REVDAT": 225 rr=re.search("\d\d-\w\w\w-\d\d",tail) 226 if rr!=None: 227 dict['release_date']=_format_date(_nice_case(rr.group())) 228 elif key=="JRNL": 229 # print key,tail 230 if dict.has_key('journal'): 231 dict['journal']+=tail 232 else: 233 dict['journal']=tail 234 elif key=="AUTHOR": 235 auth = _nice_case(_chop_end_codes(tail)) 236 if dict.has_key('author'): 237 dict['author']+=auth 238 else: 239 dict['author']=auth 240 elif key=="REMARK": 241 if re.search("REMARK 2 RESOLUTION.",hh): 242 r=_chop_end_codes(re.sub("REMARK 2 RESOLUTION.",'',hh)) 243 r=re.sub("\s+ANGSTROM.*","",r) 244 try: 245 dict['resolution']=float(r) 246 except: 247 #print 'nonstandard resolution',r 248 dict['resolution']=None 249 else: 250 # print key 251 pass 252 if dict['structure_method']=='unknown': 253 if dict['resolution']>0.0: dict['structure_method']='x-ray diffraction' 254 return dict
255 256 if __name__=='__main__': 257 """ 258 Reads a PDB file passed as argument, parses its header, extracts 259 some data and returns it as a dictionary. 260 """ 261 filename = sys.argv[1] 262 file = open(filename,'r') 263 dict = parse_pdb_header(file) 264 265 # print the dictionary 266 for d in dict.keys(): 267 print "-"*40 268 print d 269 print dict[d] 270