def __init__(self, instream, taxon_id=None):
gt.extended.CustomStream.__init__(self)
self.iterator = FlatFileIterator(instream)
self.outqueue = collections.deque()
self.go_coll = GOCollection(taxon_id)
self.uniprots = {}
self.typemaps = {
'gene': 'pseudogene',
'mRNA': 'pseudogenic_transcript',
'rRNA': 'pseudogenic_transcript',
'tRNA': 'pseudogenic_transcript',
'exon': 'pseudogenic_exon'
}
def __init__(self, instream, taxon_id=None):
gt.extended.CustomStream.__init__(self)
self.iterator = FlatFileIterator(instream)
self.outqueue = collections.deque()
self.go_coll = GOCollection(taxon_id)
self.uniprots = {}
self.typemaps = {'gene': 'pseudogene',
'mRNA': 'pseudogenic_transcript',
'rRNA': 'pseudogenic_transcript',
'tRNA': 'pseudogenic_transcript',
'exon': 'pseudogenic_exon'}
class TableInStream(gt.extended.CustomStream):
def __init__(self, instream, taxon_id=None):
gt.extended.CustomStream.__init__(self)
self.iterator = FlatFileIterator(instream)
self.outqueue = collections.deque()
self.go_coll = GOCollection(taxon_id)
self.uniprots = {}
self.typemaps = {
'gene': 'pseudogene',
'mRNA': 'pseudogenic_transcript',
'rRNA': 'pseudogenic_transcript',
'tRNA': 'pseudogenic_transcript',
'exon': 'pseudogenic_exon'
}
def get_transcript_length(self, v):
transcript_length = 0
for f in v['Gene Model']:
if int(f['Start']) <= int(f['End']) and f['Type'] == 'exon':
transcript_length += (int(f['End']) - int(f['Start']) + 1)
return transcript_length
def finaltype(self, v, mtype):
if 'pseudo' in v and v['pseudo']:
if mtype in self.typemaps:
return self.typemaps[mtype]
return mtype
def make_noncoding(self, v, gtype, gene):
rna = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, gtype),
int(v['start']),
int(v['stop']),
v['strand'])
rna.add_attribute("ID", v['ID'] + ":" + gtype)
gene.add_child(rna)
if 'Gene Model' in v:
for f in v['Gene Model']:
if int(f['Start']) <= int(f['End']):
newfeat = gt.extended.FeatureNode.create_new(
v['seqid'], self.finaltype(v, f['Type']),
int(f['Start']), int(f['End']), v['strand'])
rna.add_child(newfeat)
else:
sys.stderr.write("invalid feature range, skipping " +
str(f['Type']) + " " + v['ID'] + "\n")
def make_coding(self, v, gene):
transcript = gt.extended.FeatureNode.create_new(
v['seqid'], self.finaltype(v, "mRNA"), int(v['start']),
int(v['stop']), v['strand'])
transcript.add_attribute("ID", v['ID'] + ".1")
gene.add_child(transcript)
if 'Gene Model' in v:
newend = 1
left_i = 0
left_type = 'five_prime_UTR'
left_key = 'utr_5'
right_type = 'three_prime_UTR'
right_key = 'utr_3'
left_c = None
right_c = None
if 'utr_5' in v or 'utr_3' in v:
if v['strand'] == '-':
if 'utr_5' in v:
right_c = int(v['utr_5'])
if 'utr_3' in v:
left_c = int(v['utr_3'])
left_type = 'three_prime_UTR'
right_type = 'five_prime_UTR'
left_key = 'utr_3'
right_key = 'utr_5'
else:
if 'utr_5' in v:
left_c = int(v['utr_5'])
if 'utr_3' in v:
right_c = int(v['utr_3'])
if v['strand'] == '-':
v['Gene Model'].reverse()
transcript_length = self.get_transcript_length(v)
for f in v['Gene Model']:
if int(f['Start']) <= int(f['End']):
newfeat = gt.extended.FeatureNode.create_new(
v['seqid'], self.finaltype(v, f['Type']),
int(f['Start']), int(f['End']), v['strand'])
transcript.add_child(newfeat)
# decide whether to make CDS
if f['Type'] == 'exon':
scoord = int(f['Start'])
ecoord = int(f['End'])
newend = left_i + (int(f['End']) -
int(f['Start']) + 1)
if left_key in v:
if newend > left_c:
if left_i < left_c:
newfeat = gt.extended.FeatureNode.create_new(
v['seqid'],
self.finaltype(v,
left_type), scoord,
int(f['Start']) +
(left_c - left_i) - 1, v['strand'])
transcript.add_child(newfeat)
scoord = int(
f['Start']) + (left_c - left_i)
else:
scoord = int(f['Start'])
else:
newfeat = gt.extended.FeatureNode.create_new(
v['seqid'],
self.finaltype(v, left_type), scoord,
ecoord, v['strand'])
transcript.add_child(newfeat)
scoord = None
if right_key in v:
if newend > (transcript_length - right_c):
if left_i < (transcript_length - right_c):
newfeat = gt.extended.FeatureNode.create_new(
v['seqid'],
self.finaltype(v, right_type),
int(f['Start']) +
((transcript_length - right_c) -
left_i), ecoord, v['strand'])
transcript.add_child(newfeat)
ecoord = int(f['Start']) + (
(transcript_length - right_c) -
left_i - 1)
else:
newfeat = gt.extended.FeatureNode.create_new(
v['seqid'],
self.finaltype(v, right_type),
scoord, ecoord, v['strand'])
transcript.add_child(newfeat)
ecoord = None
if ecoord and scoord:
newfeat = gt.extended.FeatureNode.create_new(
v['seqid'], self.finaltype(v, 'CDS'),
scoord, ecoord, v['strand'])
transcript.add_child(newfeat)
left_i = newend
else:
sys.stderr.write("invalid feature range, skipping " +
str(f['Type']) + " " + v['ID'] + "\n")
# make polypeptide
polypeptide = gt.extended.FeatureNode.create_new(
v['seqid'], "polypeptide", int(v['start']), int(v['stop']),
v['strand'])
polypeptide.add_attribute("ID",
transcript.get_attribute("ID") + ":pep")
polypeptide.add_attribute("Derives_from",
transcript.get_attribute("ID"))
if 'product' in v and v['product'] != "null":
polypeptide.add_attribute(
"product",
'term%3D' + urllib.quote(v['product'], safe=' '))
self.outqueue.append(polypeptide)
# register GO terms
if self.go_coll:
self.go_coll.add_item(v)
def next(self):
while True:
#try:
if len(self.outqueue) > 0:
return self.outqueue.popleft()
try:
v = self.iterator.next()
except exceptions.StopIteration:
return None
if not v:
return None
gene = gt.extended.FeatureNode.create_new(
v['seqid'], self.finaltype(v, "gene"), int(v['start']),
int(v['stop']), v['strand'])
gene.add_attribute("ID", v['ID'])
if 'name' in v:
gene.add_attribute("Name", v['name'])
# track UniProtID -> gene mappings
# XXX handle multiple transcripts per product once supported by
# EuPathDB!
if 'uniprot_id' in v:
self.uniprots[v['uniprot_id']] = v['ID']
# non-coding RNA
m = re.match(r"(.RNA) encoding", v['type'])
if m:
self.make_noncoding(v, m.group(1), gene)
# protein coding gene
if v['type'] == 'protein coding':
self.make_coding(v, gene)
break
#except:
# if v:
# sys.stderr.write("error creating feature for %s\n" % v['ID'])
# else:
# sys.stderr.write("error creating feature , no ID yet\n")
# continue
return gene
class TableInStream(gt.extended.CustomStream):
def __init__(self, instream, taxon_id=None):
gt.extended.CustomStream.__init__(self)
self.iterator = FlatFileIterator(instream)
self.outqueue = collections.deque()
self.go_coll = GOCollection(taxon_id)
self.uniprots = {}
self.typemaps = {'gene': 'pseudogene',
'mRNA': 'pseudogenic_transcript',
'rRNA': 'pseudogenic_transcript',
'tRNA': 'pseudogenic_transcript',
'exon': 'pseudogenic_exon'}
def get_transcript_length(self, v):
transcript_length = 0
for f in v['Gene Model']:
if int(f['Start']) <= int(f['End']) and f['Type'] == 'exon':
transcript_length += (int(f['End']) - int(f['Start']) + 1)
return transcript_length
def finaltype(self, v, mtype):
if 'pseudo' in v and v['pseudo']:
if mtype in self.typemaps:
return self.typemaps[mtype]
return mtype
def make_noncoding(self, v, gtype, gene):
rna = gt.extended.FeatureNode.create_new(v['seqid'], self.finaltype(v, gtype),
int(v['start']), int(v['stop']),
v['strand'])
rna.add_attribute("ID", v['ID'] + ":" + gtype)
gene.add_child(rna)
if 'Gene Model' in v:
for f in v['Gene Model']:
if int(f['Start']) <= int(f['End']):
newfeat = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, f['Type']),
int(f['Start']),
int(f['End']),
v['strand'])
rna.add_child(newfeat)
else:
sys.stderr.write("invalid feature range, skipping "
+ str(f['Type']) + " " + v['ID']
+ "\n")
def make_coding(self, v, gene):
transcript = gt.extended.FeatureNode.create_new(v['seqid'], self.finaltype(v, "mRNA"),
int(v['start']),
int(v['stop']),
v['strand'])
transcript.add_attribute("ID", v['ID'] + ".1")
gene.add_child(transcript)
if 'Gene Model' in v:
newend = 1
left_i = 0
left_type = 'five_prime_UTR'
left_key = 'utr_5'
right_type = 'three_prime_UTR'
right_key = 'utr_3'
left_c = None
right_c = None
if 'utr_5' in v or 'utr_3' in v:
if v['strand'] == '-':
if 'utr_5' in v:
right_c = int(v['utr_5'])
if 'utr_3' in v:
left_c = int(v['utr_3'])
left_type = 'three_prime_UTR'
right_type = 'five_prime_UTR'
left_key = 'utr_3'
right_key = 'utr_5'
else:
if 'utr_5' in v:
left_c = int(v['utr_5'])
if 'utr_3' in v:
right_c = int(v['utr_3'])
if v['strand'] == '-':
v['Gene Model'].reverse()
transcript_length = self.get_transcript_length(v)
for f in v['Gene Model']:
if int(f['Start']) <= int(f['End']):
newfeat = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, f['Type']),
int(f['Start']),
int(f['End']),
v['strand'])
transcript.add_child(newfeat)
# decide whether to make CDS
if f['Type'] == 'exon':
scoord = int(f['Start'])
ecoord = int(f['End'])
newend = left_i + (int(f['End']) - int(f['Start']) + 1)
if left_key in v:
if newend > left_c:
if left_i < left_c:
newfeat = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, left_type),
scoord,
int(f['Start']) + (left_c - left_i) - 1,
v['strand'])
transcript.add_child(newfeat)
scoord = int(f['Start']) + (left_c - left_i)
else:
scoord = int(f['Start'])
else:
newfeat = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, left_type),
scoord, ecoord,
v['strand'])
transcript.add_child(newfeat)
scoord = None
if right_key in v:
if newend > (transcript_length - right_c):
if left_i < (transcript_length - right_c):
newfeat = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, right_type),
int(f['Start']) + ((transcript_length - right_c) - left_i),
ecoord,
v['strand'])
transcript.add_child(newfeat)
ecoord = int(f['Start']) + ((transcript_length - right_c) - left_i - 1)
else:
newfeat = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, right_type),
scoord, ecoord,
v['strand'])
transcript.add_child(newfeat)
ecoord = None
if ecoord and scoord:
newfeat = gt.extended.FeatureNode.create_new(v['seqid'],
self.finaltype(v, 'CDS'),
scoord, ecoord,
v['strand'])
transcript.add_child(newfeat)
left_i = newend
else:
sys.stderr.write("invalid feature range, skipping "
+ str(f['Type']) + " " + v['ID']
+ "\n")
# make polypeptide
polypeptide = gt.extended.FeatureNode.create_new(v['seqid'],
"polypeptide",
int(v['start']),
int(v['stop']),
v['strand'])
polypeptide.add_attribute("ID",
transcript.get_attribute("ID") + ":pep")
polypeptide.add_attribute("Derives_from",
transcript.get_attribute("ID"))
if 'product' in v and v['product'] != "null":
polypeptide.add_attribute("product",
'term%3D' +
urllib.quote(v['product'], safe=' '))
self.outqueue.append(polypeptide)
# register GO terms
if self.go_coll:
self.go_coll.add_item(v)
def next(self):
while True:
#try:
if len(self.outqueue) > 0:
return self.outqueue.popleft()
try:
v = self.iterator.next()
except exceptions.StopIteration:
return None
if not v:
return None
gene = gt.extended.FeatureNode.create_new(v['seqid'], self.finaltype(v, "gene"),
int(v['start']),
int(v['stop']), v['strand'])
gene.add_attribute("ID", v['ID'])
if 'name' in v:
gene.add_attribute("Name", v['name'])
# track UniProtID -> gene mappings
# XXX handle multiple transcripts per product once supported by
# EuPathDB!
if 'uniprot_id' in v:
self.uniprots[v['uniprot_id']] = v['ID']
# non-coding RNA
m = re.match(r"(.RNA) encoding", v['type'])
if m:
self.make_noncoding(v, m.group(1), gene)
# protein coding gene
if v['type'] == 'protein coding':
self.make_coding(v, gene)
break
#except:
# if v:
# sys.stderr.write("error creating feature for %s\n" % v['ID'])
# else:
# sys.stderr.write("error creating feature , no ID yet\n")
# continue
return gene
