Bio.UniGene

1 # Copyright 2006 by Sean Davis. All rights reserved. 2 # This code is part of the Biopython distribution and governed by its 3 # license. Please see the LICENSE file that should have been included 4 # as part of this package. 5 # 6 # $Id: __init__.py,v 1.12 2009-04-24 12:03:45 mdehoon Exp $ 7 # Sean Davis <sdavis2 at mail dot nih dot gov> 8 # National Cancer Institute 9 # National Institutes of Health 10 # Bethesda, MD, USA 11 # 12 13 """Parse Unigene flat file format files such as the Hs.data file. 14 15 Here is an overview of the flat file format that this parser deals with: 16 Line types/qualifiers: 17 18 ID UniGene cluster ID 19 TITLE Title for the cluster 20 GENE Gene symbol 21 CYTOBAND Cytological band 22 EXPRESS Tissues of origin for ESTs in cluster 23 RESTR_EXPR Single tissue or development stage contributes 24 more than half the total EST frequency for this gene. 25 GNM_TERMINUS genomic confirmation of presence of a 3' terminus; 26 T if a non-templated polyA tail is found among 27 a cluster's sequences; else 28 I if templated As are found in genomic sequence or 29 S if a canonical polyA signal is found on 30 the genomic sequence 31 GENE_ID Entrez gene identifier associated with at least one 32 sequence in this cluster; 33 to be used instead of LocusLink. 34 LOCUSLINK LocusLink identifier associated with at least one 35 sequence in this cluster; 36 deprecated in favor of GENE_ID 37 HOMOL Homology; 38 CHROMOSOME Chromosome. For plants, CHROMOSOME refers to mapping 39 on the arabidopsis genome. 40 STS STS 41 ACC= GenBank/EMBL/DDBJ accession number of STS 42 [optional field] 43 UNISTS= identifier in NCBI's UNISTS database 44 TXMAP Transcript map interval 45 MARKER= Marker found on at least one sequence in this 46 cluster 47 RHPANEL= Radiation Hybrid panel used to place marker 48 PROTSIM Protein Similarity data for the sequence with 49 highest-scoring protein similarity in this cluster 50 ORG= Organism 51 PROTGI= Sequence GI of protein 52 PROTID= Sequence ID of protein 53 PCT= Percent alignment 54 ALN= length of aligned region (aa) 55 SCOUNT Number of sequences in the cluster 56 SEQUENCE Sequence 57 ACC= GenBank/EMBL/DDBJ accession number of sequence 58 NID= Unique nucleotide sequence identifier (gi) 59 PID= Unique protein sequence identifier (used for 60 non-ESTs) 61 CLONE= Clone identifier (used for ESTs only) 62 END= End (5'/3') of clone insert read (used for 63 ESTs only) 64 LID= Library ID; see Hs.lib.info for library name 65 and tissue 66 MGC= 5' CDS-completeness indicator; if present, the 67 clone associated with this sequence is believed 68 CDS-complete. A value greater than 511 is the gi 69 of the CDS-complete mRNA matched by the EST, 70 otherwise the value is an indicator of the 71 reliability of the test indicating CDS 72 completeness; higher values indicate more 73 reliable CDS-completeness predictions. 74 SEQTYPE= Description of the nucleotide sequence. 75 Possible values are mRNA, EST and HTC. 76 TRACE= The Trace ID of the EST sequence, as provided by 77 NCBI Trace Archive 78 """ 79 80

81 -class SequenceLine(object):

82 """Store the information for one SEQUENCE line from a Unigene file 83 84 Initialize with the text part of the SEQUENCE line, or nothing. 85 86 Attributes and descriptions (access as LOWER CASE) 87 ACC= GenBank/EMBL/DDBJ accession number of sequence 88 NID= Unique nucleotide sequence identifier (gi) 89 PID= Unique protein sequence identifier (used for non-ESTs) 90 CLONE= Clone identifier (used for ESTs only) 91 END= End (5'/3') of clone insert read (used for ESTs only) 92 LID= Library ID; see Hs.lib.info for library name and tissue 93 MGC= 5' CDS-completeness indicator; if present, 94 the clone associated with this sequence 95 is believed CDS-complete. A value greater than 511 96 is the gi of the CDS-complete mRNA matched by the EST, 97 otherwise the value is an indicator of the reliability 98 of the test indicating CDS completeness; 99 higher values indicate more reliable CDS-completeness 100 predictions. 101 SEQTYPE= Description of the nucleotide sequence. Possible values 102 are mRNA, EST and HTC. 103 TRACE= The Trace ID of the EST sequence, as provided by NCBI 104 Trace Archive 105 """ 106

107 - def __init__(self,text=None):

108 self.acc = '' 109 self.nid = '' 110 self.lid = '' 111 self.pid = '' 112 self.clone = '' 113 self.image = '' 114 self.is_image = False 115 self.end = '' 116 self.mgc = '' 117 self.seqtype = '' 118 self.trace = '' 119 if not text==None: 120 self.text=text 121 self._init_from_text(text)

122

123 - def _init_from_text(self,text):

124 parts = text.split('; '); 125 for part in parts: 126 key, val = part.split("=") 127 if key=='CLONE': 128 if val[:5]=='IMAGE': 129 self.is_image=True 130 self.image = val[6:] 131 setattr(self,key.lower(),val)

132

133 - def __repr__(self):

134 return self.text

135 136

137 -class ProtsimLine(object):

138 """Store the information for one PROTSIM line from a Unigene file 139 140 Initialize with the text part of the PROTSIM line, or nothing. 141 142 Attributes and descriptions (access as LOWER CASE) 143 ORG= Organism 144 PROTGI= Sequence GI of protein 145 PROTID= Sequence ID of protein 146 PCT= Percent alignment 147 ALN= length of aligned region (aa) 148 """ 149

150 - def __init__(self,text=None):

151 self.org = '' 152 self.protgi = '' 153 self.protid = '' 154 self.pct = '' 155 self.aln = '' 156 if not text==None: 157 self.text=text 158 self._init_from_text(text)

159

160 - def _init_from_text(self,text):

161 parts = text.split('; '); 162 163 for part in parts: 164 key, val = part.split("=") 165 setattr(self,key.lower(),val)

166

167 - def __repr__(self):

168 return self.text

169 170

171 -class STSLine(object):

172 """Store the information for one STS line from a Unigene file 173 174 Initialize with the text part of the STS line, or nothing. 175 176 Attributes and descriptions (access as LOWER CASE) 177 178 ACC= GenBank/EMBL/DDBJ accession number of STS [optional field] 179 UNISTS= identifier in NCBI's UNISTS database 180 """ 181

182 - def __init__(self,text=None):

183 self.acc = '' 184 self.unists = '' 185 if not text==None: 186 self.text=text 187 self._init_from_text(text)

188

189 - def _init_from_text(self,text):

190 parts = text.split(' '); 191 192 for part in parts: 193 key, val = part.split("=") 194 setattr(self,key.lower(),val)

195

196 - def __repr__(self):

197 return self.text

198 199

200 -class Record(object):

201 """Store a Unigene record 202 203 Here is what is stored: 204 205 self.ID = '' # ID line 206 self.species = '' # Hs, Bt, etc. 207 self.title = '' # TITLE line 208 self.symbol = '' # GENE line 209 self.cytoband = '' # CYTOBAND line 210 self.express = [] # EXPRESS line, parsed on ';' 211 # Will be an array of strings 212 self.restr_expr = '' # RESTR_EXPR line 213 self.gnm_terminus = '' # GNM_TERMINUS line 214 self.gene_id = '' # GENE_ID line 215 self.locuslink = '' # LOCUSLINK line 216 self.homol = '' # HOMOL line 217 self.chromosome = '' # CHROMOSOME line 218 self.protsim = [] # PROTSIM entries, array of Protsims 219 # Type ProtsimLine 220 self.sequence = [] # SEQUENCE entries, array of Sequence entries 221 # Type SequenceLine 222 self.sts = [] # STS entries, array of STS entries 223 # Type STSLine 224 self.txmap = [] # TXMAP entries, array of TXMap entries 225 """ 226

227 - def __init__(self):

228 self.ID = '' # ID line 229 self.species = '' # Hs, Bt, etc. 230 self.title = '' # TITLE line 231 self.symbol = '' # GENE line 232 self.cytoband = '' # CYTOBAND line 233 self.express = [] # EXPRESS line, parsed on ';' 234 self.restr_expr = '' # RESTR_EXPR line 235 self.gnm_terminus = '' # GNM_TERMINUS line 236 self.gene_id = '' # GENE_ID line 237 self.locuslink = '' # LOCUSLINK line 238 self.homol = '' # HOMOL line 239 self.chromosome = '' # CHROMOSOME line 240 self.protsim = [] # PROTSIM entries, array of Protsims 241 self.sequence = [] # SEQUENCE entries, array of Sequence entries 242 self.sts = [] # STS entries, array of STS entries 243 self.txmap = [] # TXMAP entries, array of TXMap entries

244

245 - def __repr__(self):

246 return "<%s> %s %s\n%s" % (self.__class__.__name__, 247 self.ID, self.symbol, self.title)

248

249 -def parse(handle):

250 while True: 251 record = _read(handle) 252 if not record: 253 return 254 yield record

255 256

257 -def read(handle):

258 record = _read(handle) 259 if not record: 260 raise ValueError("No SwissProt record found") 261 # We should have reached the end of the record by now 262 remainder = handle.read() 263 if remainder: 264 raise ValueError("More than one SwissProt record found") 265 return record

266 267 268 # Everything below is private 269 270

271 -def _read(handle):

272 UG_INDENT = 12 273 record = None 274 for line in handle: 275 tag, value = line[:UG_INDENT].rstrip(), line[UG_INDENT:].rstrip() 276 line = line.rstrip() 277 if tag=="ID": 278 record = Record() 279 record.ID = value 280 record.species = record.ID.split('.')[0] 281 elif tag=="TITLE": 282 record.title = value 283 elif tag=="GENE": 284 record.symbol = value 285 elif tag=="GENE_ID": 286 record.gene_id = value 287 elif tag=="LOCUSLINK": 288 record.locuslink = value 289 elif tag=="HOMOL": 290 if value=="YES": 291 record.homol = True 292 elif value=="NO": 293 record.homol = True 294 else: 295 raise ValueError, "Cannot parse HOMOL line %s" % line 296 elif tag=="EXPRESS": 297 record.express = [word.strip() for word in value.split("|")] 298 elif tag=="RESTR_EXPR": 299 record.restr_expr = [word.strip() for word in value.split("|")] 300 elif tag=="CHROMOSOME": 301 record.chromosome = value 302 elif tag=="CYTOBAND": 303 record.cytoband = value 304 elif tag=="PROTSIM": 305 protsim = ProtsimLine(value) 306 record.protsim.append(protsim) 307 elif tag=="SCOUNT": 308 scount = int(value) 309 elif tag=="SEQUENCE": 310 sequence = SequenceLine(value) 311 record.sequence.append(sequence) 312 elif tag=="STS": 313 sts = STSLine(value) 314 record.sts.append(sts) 315 elif tag=='//': 316 if len(record.sequence)!=scount: 317 raise ValueError, "The number of sequences specified in the record (%d) does not agree with the number of sequences found (%d)" % (scount, len(record.sequence)) 318 return record 319 else: 320 raise ValueError, "Unknown tag %s" % tag 321 if record: 322 raise ValueError("Unexpected end of stream.")

323

Source Code for Package Bio.UniGene