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
import sys
from pssm import parse_ic_data, parse_ic_file
if len(sys.argv) == 2:
IC = parse_ic_file(sys.argv[1])
else:
ic_data = []
for line in sys.stdin.readlines():
if line[0] == '#': continue
ic_data.append(line.strip())
IC = parse_ic_data("\n".join(ic_data))
buffer = []
num_seqs = 1000
for col in range(len(IC)):
buffer.append([])
vdict = IC[col]
for base,value in vdict.iteritems():
freq = pow(2,value-2.0)
cnt = int(round(freq*num_seqs))
buffer[-1].extend( [ base ]*cnt )
while len(buffer[-1]) < num_seqs:
buffer[-1].append("n")
seqs = {}
cnt=1
while buffer[0]:
seq = "".join( [ buffer[col].pop() for col in range(len(IC)) ] )
seqs[cnt] = seq
IncompatibleSpliceSitePhases,
)
from pssm import parse_ic_file, Pssm, pssmscore
# Import Global variables
from settings.splicesites import (
IC_DONOR_PATTERN_OFFSET,
IC_DONOR_DATA_FILE,
IC_DONOR_NCGC_DATA_FILE,
IC_ACCEPTOR_PATTERN_OFFSET,
IC_ACCEPTOR_DATA_FILE,
)
# parse IC PSSM files of cannonical sites
IC_ACCEPTOR = parse_ic_file(IC_ACCEPTOR_DATA_FILE)
IC_DONOR = parse_ic_file(IC_DONOR_DATA_FILE)
IC_NC_GC_DONOR = parse_ic_file(IC_DONOR_NCGC_DATA_FILE)
class SpliceSiteBase(BasicGFF):
""" """
def __init__(self,start,phase=None,strand='+',pattern=None,
pattern_offset=(0,0),pssm_score=None,gff={}):
"""
Initialization function of Basal SpliceSite logic
Recommended is to use only one of the inheriting classes
@type start: number
@param start: start coord of site (e.g GT) or pattern (e.g. tgtGTcgat)
@type phase: number
import sys
from pssm import parse_ic_data, parse_ic_file
if len(sys.argv) == 2:
IC = parse_ic_file(sys.argv[1])
else:
ic_data = []
for line in sys.stdin.readlines():
if line[0] == '#': continue
ic_data.append(line.strip())
IC = parse_ic_data("\n".join(ic_data))
buffer = []
num_seqs = 1000
for col in range(len(IC)):
buffer.append([])
vdict = IC[col]
for base, value in vdict.iteritems():
freq = pow(2, value - 2.0)
cnt = int(round(freq * num_seqs))
buffer[-1].extend([base] * cnt)
while len(buffer[-1]) < num_seqs:
buffer[-1].append("n")
seqs = {}
cnt = 1
while buffer[0]:
seq = "".join([buffer[col].pop() for col in range(len(IC))])
seqs[cnt] = seq
cnt += 1
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
UnexpectedSpliceSitePhase,
IncompatibleSpliceSitePhases,
)
from pssm import parse_ic_file, Pssm, pssmscore
# Import Global variables
from settings.splicesites import (
IC_DONOR_PATTERN_OFFSET,
IC_DONOR_DATA_FILE,
IC_DONOR_NCGC_DATA_FILE,
IC_ACCEPTOR_PATTERN_OFFSET,
IC_ACCEPTOR_DATA_FILE,
)
# parse IC PSSM files of cannonical sites
IC_ACCEPTOR = parse_ic_file(IC_ACCEPTOR_DATA_FILE)
IC_DONOR = parse_ic_file(IC_DONOR_DATA_FILE)
IC_NC_GC_DONOR = parse_ic_file(IC_DONOR_NCGC_DATA_FILE)
class SpliceSiteBase(BasicGFF):
""" """
def __init__(self,
start,
phase=None,
strand='+',
pattern=None,
pattern_offset=(0, 0),
pssm_score=None,
gff={}):
"""
# Python Imports
from re import finditer, compile
from copy import deepcopy
# Import Global variables
from settings.translationalstartsites import (
TSS_MIN_PSSM_SCORE,
TSS_ALLOW_NON_CANONICAL,
TSS_NON_CANONICAL_MIN_PSSM_SCORE,
IC_TSS_DATA_FILE,
IC_TSS_PATTERN_OFFSET,
)
# parse IC PSSM file of TSS
IC_TSS = parse_ic_file(IC_TSS_DATA_FILE)
class StartCodon(BasicGFF):
def __init__(self,pos,gff={}):
""" """
BasicGFF.__init__(self)
self._gff.update(gff)
self.pos = pos
self.start = self.pos
self.end = self.start+3
self.pssm_score = 1.0 # default, dummy value
self.phase = 0
# end of function __init__