def __init__(self, ic=[], ignore_unambiguity=True, relativescore=True):
""" """
if not ic: ic = IC_ACCEPTOR
Pssm.__init__(self,
ic=ic,
ignore_unambiguity=ignore_unambiguity,
relativescore=relativescore)
def scan_pssm_splice_site(seq,splicetype="donor",
override_pattern_offset=(),min_pssm_score=None,
allow_non_canonical=False,non_canonical_min_pssm_score=0.0,
ignore_unambiguity=False,relativescore=False,):
"""
Find splice sites by a PSSM on input sequence
@type seq: string
@param seq: DNA sequence of EXACT length of the PSSM
@type splicetype: string
@param splicetype: 'donor' or 'acceptor'
@type min_pssm_score: float
@param min_pssm_score:
@type allow_non_canonical: boolean
@param allow_non_canonical: True of False
@type non_canonical_min_pssm_score: float
@param non_canonical_min_pssm_score:
@type override_pattern_offset: tuple
@param override_pattern_offset: tuple with 2 integers; use cautiously!!
@rtype: list
@return: list with SpliceDonors or SpliceAcceptors
"""
if splicetype == 'acceptor':
PSSM_MATRIX = IC_ACCEPTOR
pattern_offset = IC_ACCEPTOR_PATTERN_OFFSET
canonical = "AG"
# initialize Psmm (scoring) class
canssPssm = Pssm(ic=IC_ACCEPTOR,ignore_unambiguity=ignore_unambiguity,relativescore=relativescore)
# import output SpliceAcceptor object
from acceptor import SpliceAcceptor
elif splicetype == 'donor':
PSSM_MATRIX = IC_DONOR
pattern_offset = IC_DONOR_PATTERN_OFFSET
canonical = "GT"
# initialize Psmm (scoring) class
canssPssm = Pssm(ic=IC_DONOR,ignore_unambiguity=ignore_unambiguity,relativescore=relativescore)
# import output SpliceDonor object
from donor import SpliceDonor
else:
message = "'splicetype' (%s) not in [donor,acceptor]" % splicetype
raise InproperlyAppliedArgument, message
if allow_non_canonical:
# obtain PSSM_IC for non-canonical (GC) donors
IC_NCGC_DONOR = parse_ic_file(IC_DONOR_NCGC_DATA_FILE)
noncanonical = ["GC"]
# initialize Psmm (scoring) class
noncanssPssm = Pssm(ic=IC_NCGC_DONOR,ignore_unambiguity=ignore_unambiguity,relativescore=relativescore)
# hmm... somebody knows what he or she is doing ;-)
if override_pattern_offset:
pattern_offset = override_pattern_offset
pssmlength = len(PSSM_MATRIX)
sites = []
for offset in range(0, len(seq) - pssmlength + 1 ):
# get sequence slice of pattern and actual splice site
seqpart = seq[offset:offset+pssmlength].upper()
splicesite = seqpart[pattern_offset[0]:-pattern_offset[1]]
# continue if non-canonical sites if not requested for
if not allow_non_canonical and splicesite != canonical:
continue
elif splicesite == canonical:
# score this splicesite
#score = _score_splice_site(seqpart,splicetype=splicetype)
score = canssPssm.score(seqpart)
# check if site must be stored
if min_pssm_score or min_pssm_score == 0.0:
if score < min_pssm_score:
continue
elif splicesite != canonical and splicetype == 'donor' and splicesite in noncanonical:
# score non-canonical donor site
#score = pssmscore(seqpart,IC_NCGC_DONOR)
score = noncanssPssm.score(seqpart)
# check if site must be stored
if non_canonical_min_pssm_score or non_canonical_min_pssm_score == 0.0:
if score < non_canonical_min_pssm_score:
continue
####print seqpart, score, offset
else:
continue
if splicetype=='acceptor':
a = SpliceAcceptor(offset,seqpart,acceptor=splicesite,pssm_score=score)
sites.append(a)
else:
d = SpliceDonor(offset,seqpart,donor=splicesite,pssm_score=score)
sites.append(d)
# return sites for Donor
if splicetype == 'donor':
sites.reverse()
# and return
return sites
def __init__(self,ic=[],ignore_unambiguity=True,relativescore=True):
""" """
if not ic: ic=IC_ACCEPTOR
Pssm.__init__(self,ic=ic,ignore_unambiguity=ignore_unambiguity,relativescore=relativescore)
def scan_pssm_splice_site(
seq,
splicetype="donor",
override_pattern_offset=(),
min_pssm_score=None,
allow_non_canonical=False,
non_canonical_min_pssm_score=0.0,
ignore_unambiguity=False,
relativescore=False,
):
"""
Find splice sites by a PSSM on input sequence
@type seq: string
@param seq: DNA sequence of EXACT length of the PSSM
@type splicetype: string
@param splicetype: 'donor' or 'acceptor'
@type min_pssm_score: float
@param min_pssm_score:
@type allow_non_canonical: boolean
@param allow_non_canonical: True of False
@type non_canonical_min_pssm_score: float
@param non_canonical_min_pssm_score:
@type override_pattern_offset: tuple
@param override_pattern_offset: tuple with 2 integers; use cautiously!!
@rtype: list
@return: list with SpliceDonors or SpliceAcceptors
"""
if splicetype == 'acceptor':
PSSM_MATRIX = IC_ACCEPTOR
pattern_offset = IC_ACCEPTOR_PATTERN_OFFSET
canonical = "AG"
# initialize Psmm (scoring) class
canssPssm = Pssm(ic=IC_ACCEPTOR,
ignore_unambiguity=ignore_unambiguity,
relativescore=relativescore)
# import output SpliceAcceptor object
from acceptor import SpliceAcceptor
elif splicetype == 'donor':
PSSM_MATRIX = IC_DONOR
pattern_offset = IC_DONOR_PATTERN_OFFSET
canonical = "GT"
# initialize Psmm (scoring) class
canssPssm = Pssm(ic=IC_DONOR,
ignore_unambiguity=ignore_unambiguity,
relativescore=relativescore)
# import output SpliceDonor object
from donor import SpliceDonor
else:
message = "'splicetype' (%s) not in [donor,acceptor]" % splicetype
raise InproperlyAppliedArgument, message
if allow_non_canonical:
# obtain PSSM_IC for non-canonical (GC) donors
IC_NCGC_DONOR = parse_ic_file(IC_DONOR_NCGC_DATA_FILE)
noncanonical = ["GC"]
# initialize Psmm (scoring) class
noncanssPssm = Pssm(ic=IC_NCGC_DONOR,
ignore_unambiguity=ignore_unambiguity,
relativescore=relativescore)
# hmm... somebody knows what he or she is doing ;-)
if override_pattern_offset:
pattern_offset = override_pattern_offset
pssmlength = len(PSSM_MATRIX)
sites = []
for offset in range(0, len(seq) - pssmlength + 1):
# get sequence slice of pattern and actual splice site
seqpart = seq[offset:offset + pssmlength].upper()
splicesite = seqpart[pattern_offset[0]:-pattern_offset[1]]
# continue if non-canonical sites if not requested for
if not allow_non_canonical and splicesite != canonical:
continue
elif splicesite == canonical:
# score this splicesite
#score = _score_splice_site(seqpart,splicetype=splicetype)
score = canssPssm.score(seqpart)
# check if site must be stored
if min_pssm_score or min_pssm_score == 0.0:
if score < min_pssm_score:
continue
elif splicesite != canonical and splicetype == 'donor' and splicesite in noncanonical:
# score non-canonical donor site
#score = pssmscore(seqpart,IC_NCGC_DONOR)
score = noncanssPssm.score(seqpart)
# check if site must be stored
if non_canonical_min_pssm_score or non_canonical_min_pssm_score == 0.0:
if score < non_canonical_min_pssm_score:
continue
####print seqpart, score, offset
else:
continue
if splicetype == 'acceptor':
a = SpliceAcceptor(offset,
seqpart,
acceptor=splicesite,
pssm_score=score)
sites.append(a)
else:
d = SpliceDonor(offset,
seqpart,
donor=splicesite,
pssm_score=score)
sites.append(d)
# return sites for Donor
if splicetype == 'donor':
sites.reverse()
# and return
return sites
def scan_pssm_tss(seq,override_pattern_offset=(),min_pssm_score=None,
allow_non_canonical=False,non_canonical_min_pssm_score=0.0,
ignore_unambiguity=False,relativescore=False,):
"""
Find TSS's by a PSSM on input sequence
@type seq: string
@param seq: DNA sequence of EXACT length of the PSSM
@type min_pssm_score: float
@param min_pssm_score:
@type allow_non_canonical: boolean
@param allow_non_canonical: True of False
@type non_canonical_min_pssm_score: float
@param non_canonical_min_pssm_score:
@type override_pattern_offset: tuple
@param override_pattern_offset: tuple with 2 integers; use cautiously!!
@rtype: list
@return: list with TranslationalStartSites
"""
pattern_offset = IC_TSS_PATTERN_OFFSET
canonical = "ATG"
regex_pattern = "ATG"
# obtain Psmm (scoring) class
tssPssm = Pssm(ic=IC_TSS,ignore_unambiguity=ignore_unambiguity,relativescore=relativescore)
# hmm... somebody knows what he or she is doing ;-)
if override_pattern_offset:
pattern_offset = override_pattern_offset
if allow_non_canonical:
# http://en.wikipedia.org/wiki/Start_codon
# Blattner, F. R.; Plunkett, G.; Bloch, C. A.; Perna, N. T.; Burland, V.; Riley, M.; Collado-vides, J.; Glasner, J. D. et al (1997). "The Complete Genome Sequence of Escherichia coli K-12". Science 277 (5331): 1453-62
noncanonicalGTG = "GTG"
noncanonicalTTG = "TTG"
regex_pattern = "[ATG]TG"
tssPssmGTG = deepcopy(tssPssm)
tssPssmGTG.ic[pattern_offset[0]]['A'],tssPssmGTG.ic[pattern_offset[0]]['G'] = tssPssmGTG.ic[pattern_offset[0]]['G'],tssPssmGTG.ic[pattern_offset[0]]['A']
tssPssmTTG = deepcopy(tssPssm)
tssPssmTTG.ic[pattern_offset[0]]['A'],tssPssmGTG.ic[pattern_offset[0]]['T'] = tssPssmTTG.ic[pattern_offset[0]]['T'],tssPssmTTG.ic[pattern_offset[0]]['A']
pssmlength = len(IC_TSS)
sites = []
for atg in finditer(compile(regex_pattern),seq.upper()[0:-pattern_offset[1]]):
if atg.start() < pattern_offset[0]: continue
seqpart = seq[atg.start()-pattern_offset[0]:atg.end()+pattern_offset[1]]
tss = atg.group()
offset = atg.start()-pattern_offset[0]
# continue if non-canonical sites if not requested for
if tss == canonical:
# score this TSS
score = tssPssm.score(seqpart)
# continue if score is to low
if min_pssm_score or min_pssm_score == 0.0:
if score < min_pssm_score:
continue
else:
if atg.group() == noncanonicalGTG:
score = tssPssmGTG.score(seqpart)
elif atg.group() == noncanonicalTTG:
score = tssPssmTTG.score(seqpart)
else:
continue
# continue if score is to low
if non_canonical_min_pssm_score or non_canonical_min_pssm_score == 0.0:
if score < non_canonical_min_pssm_score:
continue
# phase of this tss
phase = (offset+pattern_offset[0]) % 3
# instantiate tss object
t = TranslationalStartSite(offset,seqpart,pssm_score=score,phase=phase)
sites.append(t)
# and return
return sites
