def _create_failed_intron_gff(introndata, assessed_interfaces, intron2label):
""" """
failed_introns = {}
for kk in introndata.keys():
start, end = kk
introns = introndata[kk]
label = intron2label[kk]
value = assessed_interfaces[label]
if start in [s for s, e in failed_introns.keys()]:
if end < min([e for s, e in failed_introns.keys()]):
# replace this key
for s, e in failed_introns.keys():
if start == s and e == min([_e for _s, _e in failed_introns.keys()]):
del (failed_introns[(s, e)])
failed_introns[kk] = value
break
elif end in [e for s, e in failed_introns.keys()]:
if start > max([s for s, e in failed_introns.keys()]):
# replace this key
for s, e in failed_introns.keys():
if end == e and s == max([_s for _s, _e in failed_introns.keys()]):
del (failed_introns[(s, e)])
failed_introns[kk] = value
break
else:
# new key -> store
failed_introns[kk] = value
# create `failed` intron gff lines
gfflines = []
for (start, end) in failed_introns.keys():
assesed_org_list = failed_introns[(start, end)]
for orgid in assesed_org_list:
if orgid in [intron._reference for intron in introndata[(start, end)]]:
continue
# if here, a failed organism
gclass = "failedintron" # "%s_%s" % (start, end)
gname = "%s_%s_%s" % (orgid, start, end)
orgSfullname = _get_organism_full_name(orgid)
# if ABGP_ORGANISM2FULLSPECIESNAME_MAPPING.has_key(orgid):
# orgSfullname = ABGP_ORGANISM2FULLSPECIESNAME_MAPPING[orgid]
# elif ABGP_ORGANISM2FULLSPECIESNAME_MAPPING.has_key(orgid[0:-1]):
# orgSfullname = ABGP_ORGANISM2FULLSPECIESNAME_MAPPING[orgid[0:-1]]
# else:
# orgSfullname = orgid
# create gclass/gname and GFF decoration string
lastcol = """%s %s; Reference %s; Note '%s'""" % (gclass, gname, orgid, orgSfullname)
# get fmethod/class from 1 of the intron(s)
example = introndata[(start, end)][0]
fmethod = example.__class__.__name__
gffline = (None, "ABGPfailed", fmethod, start + 1, end, ".", "+", ".", lastcol)
gfflines.append(gffline)
# return gfflines list
return gfflines
def create_projected_signalp_for_informants(organism,
PCG,
input,
minimal_aa_overlap=None,
verbose=False):
""" """
# return list with ProjectedSignalPeptides
return_projsignalpeptides = []
for orgS in PCG.organism_set():
if organism == orgS: continue
if verbose: print orgS, organism
# dict in order to create unique projections
projections = {}
for pacbporf in PCG.get_pacbps_by_organisms(organism, orgS):
if not (pacbporf.orfS._has_signalp_sites_predicted and\
pacbporf.orfS._signalp_sites):
# no SignalPeptides on this Orf of the informant
continue
if minimal_aa_overlap and not pacbporf.orfQ._signalp_sites:
# overlap required AND no SignalPeptides on the target Orf
continue
else:
# calculate DNA range of target gene's SignalPeptides
qdnarange = []
for spQ in pacbporf.orfQ._signalp_sites:
qdnarange.extend(range(spQ.start, spQ.end))
# get full name tag of informant
orgSfullname = "%s [%s]" % (_get_organism_full_name(
orgS, truncate=True), input[orgS]['proteinfref'])
# loop over the informant's SignalPeptides
for spS in pacbporf.orfS._signalp_sites:
prjQSPstart = pacbporf.dnapos_sbjct2query(spS.start)
prjQSPend = pacbporf.dnapos_sbjct2query(spS.end)
prjQTSSstart = pacbporf.dnapos_sbjct2query(spS.tss.start)
prjQTSSend = pacbporf.dnapos_sbjct2query(spS.tss.end)
coords = [prjQSPstart, prjQSPend, prjQTSSstart, prjQTSSend]
if CoordinateOutOfRange in coords:
############################################################
if verbose: print "OUT-OF-RANGE:", spS
############################################################
# projection on query NOT possible (out of range of Orf)
continue
################################################################
if verbose:
print orgS, pacbporf, len(pacbporf.orfQ._signalp_sites),
print len(pacbporf.orfS._signalp_sites)
################################################################
# calculate AA overlap with target organism SignalPeptides
sdnarange = Set(range(prjQSPstart, prjQSPend))
dna_overlap = sdnarange.intersection(qdnarange)
aa_overlap = len(dna_overlap) / 3
if minimal_aa_overlap and aa_overlap < minimal_aa_overlap:
# minimal overlap required and not achieved
continue
# create a ProjectedSignalPSignalPeptide
prjTSS = TranslationalStartSite(prjQTSSstart,
"n" * 19,
pssm_score=spS.tss.pssm_score)
prjTSS._gff['fmethod'] = 'projectedTSSpssm'
prjTSS._gff['column9data'] = {'Informant': orgSfullname}
prjSignalP_gff = {
'fmethod': 'projectedSignalPeptide',
'column9data': {
'Informant': orgSfullname
},
'gname': "%s_%s_%s" % (orgS, prjQSPstart, prjQSPend)
}
prjSignalP = ProjectedSignalPSignalPeptide(prjQSPstart,
prjQSPend,
spS.pssm_score,
tss=prjTSS,
gff=prjSignalP_gff)
prjSignalP._gff['fmethod'] = 'projectedSignalPeptide'
prjSignalP._gff['column9data'] = {'Informant': orgSfullname}
# store to prjections dict
projections[(prjSignalP.start, prjSignalP.end)] = prjSignalP
################################################################
if verbose: print prjSignalP, aa_overlap
################################################################
# store unique projections to return_projsignalpeptides
for sgp in projections.values():
return_projsignalpeptides.append(sgp)
# return list of ProjectedSignalPeptides
return return_projsignalpeptides
def _create_failed_intron_gff(introndata, assessed_interfaces, intron2label):
""" """
failed_introns = {}
for kk in introndata.keys():
start, end = kk
introns = introndata[kk]
label = intron2label[kk]
value = assessed_interfaces[label]
if start in [s for s, e in failed_introns.keys()]:
if end < min([e for s, e in failed_introns.keys()]):
# replace this key
for s, e in failed_introns.keys():
if start == s and e == min(
[_e for _s, _e in failed_introns.keys()]):
del (failed_introns[(s, e)])
failed_introns[kk] = value
break
elif end in [e for s, e in failed_introns.keys()]:
if start > max([s for s, e in failed_introns.keys()]):
# replace this key
for s, e in failed_introns.keys():
if end == e and s == max(
[_s for _s, _e in failed_introns.keys()]):
del (failed_introns[(s, e)])
failed_introns[kk] = value
break
else:
# new key -> store
failed_introns[kk] = value
# create `failed` intron gff lines
gfflines = []
for (start, end) in failed_introns.keys():
assesed_org_list = failed_introns[(start, end)]
for orgid in assesed_org_list:
if orgid in [
intron._reference for intron in introndata[(start, end)]
]:
continue
# if here, a failed organism
gclass = "failedintron" # "%s_%s" % (start, end)
gname = "%s_%s_%s" % (orgid, start, end)
orgSfullname = _get_organism_full_name(orgid)
#if ABGP_ORGANISM2FULLSPECIESNAME_MAPPING.has_key(orgid):
# orgSfullname = ABGP_ORGANISM2FULLSPECIESNAME_MAPPING[orgid]
#elif ABGP_ORGANISM2FULLSPECIESNAME_MAPPING.has_key(orgid[0:-1]):
# orgSfullname = ABGP_ORGANISM2FULLSPECIESNAME_MAPPING[orgid[0:-1]]
#else:
# orgSfullname = orgid
# create gclass/gname and GFF decoration string
lastcol = """%s %s; Reference %s; Note '%s'""" % (
gclass, gname, orgid, orgSfullname)
# get fmethod/class from 1 of the intron(s)
example = introndata[(start, end)][0]
fmethod = example.__class__.__name__
gffline = (None, 'ABGPfailed', fmethod, start + 1, end, ".", "+",
".", lastcol)
gfflines.append(gffline)
# return gfflines list
return gfflines
def create_projected_signalp_for_informants(organism,PCG,input,
minimal_aa_overlap=None,verbose=False):
""" """
# return list with ProjectedSignalPeptides
return_projsignalpeptides = []
for orgS in PCG.organism_set():
if organism == orgS: continue
if verbose: print orgS, organism
# dict in order to create unique projections
projections = {}
for pacbporf in PCG.get_pacbps_by_organisms(organism,orgS):
if not (pacbporf.orfS._has_signalp_sites_predicted and\
pacbporf.orfS._signalp_sites):
# no SignalPeptides on this Orf of the informant
continue
if minimal_aa_overlap and not pacbporf.orfQ._signalp_sites:
# overlap required AND no SignalPeptides on the target Orf
continue
else:
# calculate DNA range of target gene's SignalPeptides
qdnarange = []
for spQ in pacbporf.orfQ._signalp_sites:
qdnarange.extend( range(spQ.start,spQ.end) )
# get full name tag of informant
orgSfullname = "%s [%s]" % (
_get_organism_full_name(orgS,truncate=True),
input[orgS]['proteinfref'] )
# loop over the informant's SignalPeptides
for spS in pacbporf.orfS._signalp_sites:
prjQSPstart = pacbporf.dnapos_sbjct2query(spS.start)
prjQSPend = pacbporf.dnapos_sbjct2query(spS.end)
prjQTSSstart= pacbporf.dnapos_sbjct2query(spS.tss.start)
prjQTSSend = pacbporf.dnapos_sbjct2query(spS.tss.end)
coords = [ prjQSPstart, prjQSPend, prjQTSSstart,prjQTSSend ]
if CoordinateOutOfRange in coords:
############################################################
if verbose: print "OUT-OF-RANGE:", spS
############################################################
# projection on query NOT possible (out of range of Orf)
continue
################################################################
if verbose:
print orgS, pacbporf, len(pacbporf.orfQ._signalp_sites),
print len(pacbporf.orfS._signalp_sites)
################################################################
# calculate AA overlap with target organism SignalPeptides
sdnarange = Set(range(prjQSPstart,prjQSPend))
dna_overlap = sdnarange.intersection(qdnarange)
aa_overlap = len(dna_overlap)/3
if minimal_aa_overlap and aa_overlap < minimal_aa_overlap:
# minimal overlap required and not achieved
continue
# create a ProjectedSignalPSignalPeptide
prjTSS = TranslationalStartSite(
prjQTSSstart,"n"*19,
pssm_score=spS.tss.pssm_score )
prjTSS._gff['fmethod'] = 'projectedTSSpssm'
prjTSS._gff['column9data'] = {'Informant': orgSfullname}
prjSignalP_gff = {
'fmethod' : 'projectedSignalPeptide',
'column9data': {'Informant': orgSfullname},
'gname' : "%s_%s_%s" % (orgS,prjQSPstart,prjQSPend) }
prjSignalP = ProjectedSignalPSignalPeptide(
prjQSPstart,prjQSPend,
spS.pssm_score,tss=prjTSS,
gff=prjSignalP_gff )
prjSignalP._gff['fmethod'] = 'projectedSignalPeptide'
prjSignalP._gff['column9data'] = {'Informant': orgSfullname}
# store to prjections dict
projections[(prjSignalP.start,prjSignalP.end)] = prjSignalP
################################################################
if verbose: print prjSignalP, aa_overlap
################################################################
# store unique projections to return_projsignalpeptides
for sgp in projections.values():
return_projsignalpeptides.append(sgp)
# return list of ProjectedSignalPeptides
return return_projsignalpeptides
