def _find_qp_or_pq_match_on_orfobj(exon,
orfObj,
min_identity_count=1,
max_dissimilar_count=0):
""" """
tinyexonmatches = []
protseq = exon.proteinsequence()
protlen = len(protseq)
for offset in range(0, orfObj.protein_length - protlen):
seqpart = orfObj.protein_sequence[offset:offset + protlen]
match = make_alignment_match(protseq,
seqpart,
matrix=TINYEXON_MATRIX.matrix)
if match.count(" ") <= max_dissimilar_count and\
match.count("*") >= min_identity_count:
aapos = orfObj.protein_startPY + offset
dnapos = orfObj.aapos2dnapos(aapos)
if exon.acceptor.phase == 2: dnapos -= 1
if exon.acceptor.phase == 1: dnapos -= 2
dnaseq = orfObj.inputgenomicsequence[dnapos - 2:dnapos +
exon.length + 2].upper()
if dnaseq[0:2] == 'AG' or dnaseq[-2:] in ['GT', 'GC']:
tinyexonmatches.append((seqpart, aapos))
# return list of tinyexon match tuples
return tinyexonmatches
def _are_tinyexons_similar(exonQ,exonS,min_identity_count=1,max_dissimilar_count=0):
""" """
if exonS.length != exonQ.length: return False
if exonQ.donor.phase != exonS.donor.phase: return False
if exonQ.proteinsequence() == exonS.proteinsequence(): return True
# if here: test similarity
match = make_alignment_match(
exonQ.proteinsequence(),
exonS.proteinsequence(),
matrix=TINYEXON_MATRIX.matrix )
if match.count(" ") <= max_dissimilar_count and\
match.count("*") >= min_identity_count:
return True
else:
return False
def _find_match_on_orfobj(exon,
orfObj,
min_identity_count=1,
max_dissimilar_count=0):
""" """
tinyexonmatches = []
protseq = exon.proteinsequence()
protlen = len(protseq)
for offset in range(0, orfObj.protein_length - protlen):
seqpart = orfObj.protein_sequence[offset:offset + protlen]
match = make_alignment_match(protseq,
seqpart,
matrix=TINYEXON_MATRIX.matrix)
if protseq == "SGWNAA" and seqpart in [
'SGFNSA', 'SGWNAA', 'GLFNSV', 'SGFTSA', 'GGFTSA', 'GDFNAV',
'GKFNTI', 'SGFNSA', 'GNFTTI', 'GGGSTN', 'GDFSAV', 'GKFNTI',
'GAFTSA'
]:
maxQ = TINYEXON_MATRIX.scorealignment(protseq, protseq)
maxS = TINYEXON_MATRIX.scorealignment(seqpart, seqpart)
print True, protseq, "'%s'" % match, seqpart, (
TINYEXON_MATRIX.scorealignment(protseq,
seqpart), maxQ, maxS), orfObj
#if protseq == "LSPSM":
# maxQ = TINYEXON_MATRIX.scorealignment(protseq,protseq)
# maxS = TINYEXON_MATRIX.scorealignment(seqpart,seqpart)
# print False, protseq, "'%s'" % match, seqpart, (TINYEXON_MATRIX.scorealignment(protseq,seqpart),maxQ,maxS), orfObj
if match.count(" ") <= max_dissimilar_count and\
match.count("*") >= min_identity_count:
aapos = orfObj.protein_startPY + offset
tinyexonmatches.append((seqpart, aapos))
elif len(seqpart) >= 5 and match.count("*") >= min_identity_count and\
TINYEXON_MATRIX.scorealignment(protseq,seqpart) > 0 and\
match.count(" ") <= max_dissimilar_count+1:
# escape for longer tinyexons; relax constrain a little bit
aapos = orfObj.protein_startPY + offset
tinyexonmatches.append((seqpart, aapos))
else:
pass
# return list of tinyexon match tuples
return tinyexonmatches
def _are_tinyexons_similar(exonQ,
exonS,
min_identity_count=1,
max_dissimilar_count=0):
""" """
if exonS.length != exonQ.length: return False
if exonQ.donor.phase != exonS.donor.phase: return False
if exonQ.proteinsequence() == exonS.proteinsequence(): return True
# if here: test similarity
match = make_alignment_match(exonQ.proteinsequence(),
exonS.proteinsequence(),
matrix=TINYEXON_MATRIX.matrix)
if match.count(" ") <= max_dissimilar_count and\
match.count("*") >= min_identity_count:
return True
else:
return False
def _find_qp_or_pq_match_on_orfobj(exon,orfObj,min_identity_count=1,max_dissimilar_count=0):
""" """
tinyexonmatches = []
protseq = exon.proteinsequence()
protlen = len(protseq)
for offset in range(0,orfObj.protein_length-protlen):
seqpart = orfObj.protein_sequence[offset:offset+protlen]
match = make_alignment_match(protseq,seqpart,matrix=TINYEXON_MATRIX.matrix)
if match.count(" ") <= max_dissimilar_count and\
match.count("*") >= min_identity_count:
aapos = orfObj.protein_startPY + offset
dnapos = orfObj.aapos2dnapos(aapos)
if exon.acceptor.phase == 2: dnapos-=1
if exon.acceptor.phase == 1: dnapos-=2
dnaseq = orfObj.inputgenomicsequence[dnapos-2:dnapos+exon.length+2].upper()
if dnaseq[0:2] == 'AG' or dnaseq[-2:] in ['GT','GC']:
tinyexonmatches.append( (seqpart,aapos) )
# return list of tinyexon match tuples
return tinyexonmatches
def _find_match_on_orfobj(exon,orfObj,min_identity_count=1,max_dissimilar_count=0):
""" """
tinyexonmatches = []
protseq = exon.proteinsequence()
protlen = len(protseq)
for offset in range(0,orfObj.protein_length-protlen):
seqpart = orfObj.protein_sequence[offset:offset+protlen]
match = make_alignment_match(protseq,seqpart,matrix=TINYEXON_MATRIX.matrix)
if protseq == "SGWNAA" and seqpart in ['SGFNSA','SGWNAA','GLFNSV','SGFTSA','GGFTSA','GDFNAV','GKFNTI','SGFNSA','GNFTTI','GGGSTN','GDFSAV','GKFNTI','GAFTSA']:
maxQ = TINYEXON_MATRIX.scorealignment(protseq,protseq)
maxS = TINYEXON_MATRIX.scorealignment(seqpart,seqpart)
print True, protseq, "'%s'" % match, seqpart, (TINYEXON_MATRIX.scorealignment(protseq,seqpart),maxQ,maxS), orfObj
#if protseq == "LSPSM":
# maxQ = TINYEXON_MATRIX.scorealignment(protseq,protseq)
# maxS = TINYEXON_MATRIX.scorealignment(seqpart,seqpart)
# print False, protseq, "'%s'" % match, seqpart, (TINYEXON_MATRIX.scorealignment(protseq,seqpart),maxQ,maxS), orfObj
if match.count(" ") <= max_dissimilar_count and\
match.count("*") >= min_identity_count:
aapos = orfObj.protein_startPY + offset
tinyexonmatches.append( (seqpart,aapos) )
elif len(seqpart) >= 5 and match.count("*") >= min_identity_count and\
TINYEXON_MATRIX.scorealignment(protseq,seqpart) > 0 and\
match.count(" ") <= max_dissimilar_count+1:
# escape for longer tinyexons; relax constrain a little bit
aapos = orfObj.protein_startPY + offset
tinyexonmatches.append( (seqpart,aapos) )
else:
pass
# return list of tinyexon match tuples
return tinyexonmatches