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  """Parse the header of a PDB file.""" 
 25   
 26  from __future__ import print_function 
 27   
 28  import re 
 29   
 30  from Bio import File 
 31   
 32   
33 -def _get_journal(inl):
34 # JRNL AUTH L.CHEN,M.DOI,F.S.MATHEWS,A.Y.CHISTOSERDOV, 2BBK 7 35 journal = "" 36 for l in inl: 37 if re.search("\AJRNL", l): 38 journal += l[19:72].lower() 39 journal = re.sub("\s\s+", " ", journal) 40 return journal
41 42
43 -def _get_references(inl):
44 # REMARK 1 REFERENCE 1 1CSE 11 45 # REMARK 1 AUTH W.BODE,E.PAPAMOKOS,D.MUSIL 1CSE 12 46 references = [] 47 actref = "" 48 for l in inl: 49 if re.search("\AREMARK 1", l): 50 if re.search("\AREMARK 1 REFERENCE", l): 51 if actref != "": 52 actref = re.sub("\s\s+", " ", actref) 53 if actref != " ": 54 references.append(actref) 55 actref = "" 56 else: 57 actref += l[19:72].lower() 58 59 if actref != "": 60 actref = re.sub("\s\s+", " ", actref) 61 if actref != " ": 62 references.append(actref) 63 return references
64 65 66 # bring dates to format: 1909-01-08
67 -def _format_date(pdb_date):
68 """Converts dates from DD-Mon-YY to YYYY-MM-DD format.""" 69 date = "" 70 year = int(pdb_date[7:]) 71 if year < 50: 72 century = 2000 73 else: 74 century = 1900 75 date = str(century + year) + "-" 76 all_months = ['xxx', 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 77 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'] 78 month = str(all_months.index(pdb_date[3:6])) 79 if len(month) == 1: 80 month = '0' + month 81 date = date + month + '-' + pdb_date[:2] 82 return date
83 84
85 -def _chop_end_codes(line):
86 """Chops lines ending with ' 1CSA 14' and the like.""" 87 return re.sub("\s\s\s\s+[\w]{4}.\s+\d*\Z", "", line)
88 89
90 -def _chop_end_misc(line):
91 """Chops lines ending with ' 14-JUL-97 1CSA' and the like.""" 92 return re.sub("\s\s\s\s+.*\Z", "", line)
93 94
95 -def _nice_case(line):
96 """Makes A Lowercase String With Capitals.""" 97 l = line.lower() 98 s = "" 99 i = 0 100 nextCap = 1 101 while i < len(l): 102 c = l[i] 103 if c >= 'a' and c <= 'z' and nextCap: 104 c = c.upper() 105 nextCap = 0 106 elif c == ' ' or c == '.' or c == ',' or c == ';' or c == ':' or c == '\t' or\ 107 c == '-' or c == '_': 108 nextCap = 1 109 s += c 110 i += 1 111 return s
112 113
114 -def parse_pdb_header(infile):
115 """ 116 Returns the header lines of a pdb file as a dictionary. 117 118 Dictionary keys are: head, deposition_date, release_date, structure_method, 119 resolution, structure_reference, journal_reference, author and 120 compound. 121 """ 122 header = [] 123 with File.as_handle(infile, 'r') as f: 124 for l in f: 125 record_type = l[0:6] 126 if (record_type == 'ATOM ' or record_type == 'HETATM' or 127 record_type == 'MODEL '): 128 break 129 else: 130 header.append(l) 131 return _parse_pdb_header_list(header)
132 133
134 -def _parse_pdb_header_list(header):
135 # database fields 136 dict = {'name': "", 137 'head': '', 138 'deposition_date': "1909-01-08", 139 'release_date': "1909-01-08", 140 'structure_method': "unknown", 141 'resolution': 0.0, 142 'structure_reference': "unknown", 143 'journal_reference': "unknown", 144 'author': "", 145 'compound': {'1': {'misc': ''}}, 'source': {'1': {'misc': ''}}} 146 147 dict['structure_reference'] = _get_references(header) 148 dict['journal_reference'] = _get_journal(header) 149 comp_molid = "1" 150 src_molid = "1" 151 last_comp_key = "misc" 152 last_src_key = "misc" 153 154 for hh in header: 155 h = re.sub("[\s\n\r]*\Z", "", hh) # chop linebreaks off 156 # key=re.sub("\s.+\s*","",h) 157 key = h[:6].strip() 158 # tail=re.sub("\A\w+\s+\d*\s*","",h) 159 tail = h[10:].strip() 160 # print("%s:%s" % (key, tail) 161 162 # From here, all the keys from the header are being parsed 163 if key == "TITLE": 164 name = _chop_end_codes(tail).lower() 165 if 'name' in dict: 166 dict['name'] += " " + name 167 else: 168 dict['name'] = name 169 elif key == "HEADER": 170 rr = re.search("\d\d-\w\w\w-\d\d", tail) 171 if rr is not None: 172 dict['deposition_date'] = _format_date(_nice_case(rr.group())) 173 head = _chop_end_misc(tail).lower() 174 dict['head'] = head 175 elif key == "COMPND": 176 tt = re.sub("\;\s*\Z", "", _chop_end_codes(tail)).lower() 177 # look for E.C. numbers in COMPND lines 178 rec = re.search('\d+\.\d+\.\d+\.\d+', tt) 179 if rec: 180 dict['compound'][comp_molid]['ec_number'] = rec.group() 181 tt = re.sub("\((e\.c\.)*\d+\.\d+\.\d+\.\d+\)", "", tt) 182 tok = tt.split(":") 183 if len(tok) >= 2: 184 ckey = tok[0] 185 cval = re.sub("\A\s*", "", tok[1]) 186 if ckey == 'mol_id': 187 dict['compound'][cval] = {'misc': ''} 188 comp_molid = cval 189 last_comp_key = "misc" 190 else: 191 dict['compound'][comp_molid][ckey] = cval 192 last_comp_key = ckey 193 else: 194 dict['compound'][comp_molid][last_comp_key] += tok[0] + " " 195 elif key == "SOURCE": 196 tt = re.sub("\;\s*\Z", "", _chop_end_codes(tail)).lower() 197 tok = tt.split(":") 198 # print(tok) 199 if len(tok) >= 2: 200 ckey = tok[0] 201 cval = re.sub("\A\s*", "", tok[1]) 202 if ckey == 'mol_id': 203 dict['source'][cval] = {'misc': ''} 204 comp_molid = cval 205 last_src_key = "misc" 206 else: 207 dict['source'][comp_molid][ckey] = cval 208 last_src_key = ckey 209 else: 210 dict['source'][comp_molid][last_src_key] += tok[0] + " " 211 elif key == "KEYWDS": 212 kwd = _chop_end_codes(tail).lower() 213 if 'keywords' in dict: 214 dict['keywords'] += " " + kwd 215 else: 216 dict['keywords'] = kwd 217 elif key == "EXPDTA": 218 expd = _chop_end_codes(tail) 219 # chop junk at end of lines for some structures 220 expd = re.sub('\s\s\s\s\s\s\s.*\Z', '', expd) 221 # if re.search('\Anmr',expd,re.IGNORECASE): expd='nmr' 222 # if re.search('x-ray diffraction',expd,re.IGNORECASE): expd='x-ray diffraction' 223 dict['structure_method'] = expd.lower() 224 elif key == "CAVEAT": 225 # make Annotation entries out of these!!! 226 pass 227 elif key == "REVDAT": 228 rr = re.search("\d\d-\w\w\w-\d\d", tail) 229 if rr is not None: 230 dict['release_date'] = _format_date(_nice_case(rr.group())) 231 elif key == "JRNL": 232 # print("%s:%s" % (key, tail)) 233 if 'journal' in dict: 234 dict['journal'] += tail 235 else: 236 dict['journal'] = tail 237 elif key == "AUTHOR": 238 auth = _nice_case(_chop_end_codes(tail)) 239 if 'author' in dict: 240 dict['author'] += auth 241 else: 242 dict['author'] = auth 243 elif key == "REMARK": 244 if re.search("REMARK 2 RESOLUTION.", hh): 245 r = _chop_end_codes(re.sub("REMARK 2 RESOLUTION.", '', hh)) 246 r = re.sub("\s+ANGSTROM.*", "", r) 247 try: 248 dict['resolution'] = float(r) 249 except: 250 # print('nonstandard resolution %r' % r) 251 dict['resolution'] = None 252 else: 253 # print(key) 254 pass 255 if dict['structure_method'] == 'unknown': 256 if dict['resolution'] > 0.0: 257 dict['structure_method'] = 'x-ray diffraction' 258 return dict
259 260 if __name__ == '__main__': 261 # Reads a PDB file passed as argument, parses its header, extracts 262 # some data and returns it as a dictionary. 263 import sys 264 filename = sys.argv[1] 265 with open(filename, 'r') as handle: 266 data_dict = parse_pdb_header(handle) 267 268 # print the dictionary 269 for k, y in data_dict.items(): 270 print("-" * 40) 271 print(k) 272 print(y) 273