def update_PCG_with_signalpexons(signalpexonseqs,
PCG,
OPTIONS,
min_pacbporf_identityscore=0.20,
verbose=True):
""" """
if not signalpexonseqs.has_key(OPTIONS.target): return False
is_any_pacbporf_added = False
for targetSPexon in signalpexonseqs[OPTIONS.target]:
target = OPTIONS.target
for informant, infSPlist in signalpexonseqs.iteritems():
if informant == OPTIONS.target: continue
# check if informant has been deleted in the meanwhile
if informant not in PCG.organism_set(): continue
# list to store signalp exons into
signalpexon_pacbp_list = []
# get ordered pacbporfs fromt he PCG
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(OPTIONS.target, informant))
if not thepacbporfs:
# no alignments present for this organism (can happen!)
continue
for informantSPexon in infSPlist:
coords = [
targetSPexon.protein_start(),
targetSPexon.protein_end(),
informantSPexon.protein_start(),
informantSPexon.protein_end(),
]
# prior to making ClustalW-PacbP, check PacbPCOORD placeability
# into the list of pacbporfs
pacbpCoordsObj = PacbPCOORDS(input=(
targetSPexon.proteinsequence(),
informantSPexon.proteinsequence(),
targetSPexon.protein_start(),
informantSPexon.protein_start(),
))
if False in [
pacbpCoordsObj.is_positioned_compatibly(pacbporf)
for pacbporf in thepacbporfs
]:
# *NOT* placable in current ordered list of PacbPORFS
continue
dist = pacbpCoordsObj.distance_towards(thepacbporfs[0])
if dist > SIGNALP_FIRSTEXON_MAX_INTRON_NT_LENGTH / 3:
# WAY TO FAR in front of current gene structure parts.
# Do not allow (pooras a *NOT* placable in current ordered list of PacbPORFS
continue
elif dist == 0:
# NOT placeable in front of the rest of the PacbPORFS.
continue
else:
pass
# perform ClustalW alignment on the SP exons
(alignedseqs,alignment) =\
clustalw( seqs= {
OPTIONS.target: targetSPexon.proteinsequence(),
informant: informantSPexon.proteinsequence() } )
# make pacbp from clustalw alignment
pacbp = pacbp_from_clustalw(
alignment=(alignedseqs[OPTIONS.target], alignment,
alignedseqs[informant]),
coords=coords)
# is there any alignment constructed?
if not pacbp: continue
# ignore (very) poor identyscore alignments
if pacbp.identityscore < min_pacbporf_identityscore: continue
# if here make extended pacbpORF
signalpexonPacbpORF = pacbp2pacbporf(pacbp, targetSPexon.orf,
informantSPexon.orf)
signalpexonPacbpORF.extend_pacbporf_after_stops()
# and store in signalpexon_pacbp_list
signalpexon_pacbp_list.append(signalpexonPacbpORF)
################################################################
if verbose:
print alignedseqs[OPTIONS.target], OPTIONS.target
print alignment
print alignedseqs[informant], informant
if pacbp:
print pacbp, (OPTIONS.target, targetSPexon.orf.id),
print(informant, informantSPexon.orf.id),
print "DISTANCE::", dist
pacbp.print_protein()
print ""
################################################################
# If there are signalpexon-guided pacbporfs found, store the one
# with the highest bitscore
if signalpexon_pacbp_list:
signalpexon_pacbp_list = order_list_by_attribute(
signalpexon_pacbp_list, order_by='bits', reversed=True)
# store best bitscoring pacbporf to PCG
signalp_pacbporf = signalpexon_pacbp_list[0]
pacbporf2PCG(signalp_pacbporf,
OPTIONS.target,
informant,
PCG,
source='SignalP-ClustalW')
is_any_pacbporf_added = True
####################################################################
if verbose:
print "SignalP Exon added to PCG:", signalp_pacbporf, informant
####################################################################
else:
pass
# return pointer is_any_pacbporf_added
return is_any_pacbporf_added
def update_PCG_with_signalpexons(signalpexonseqs,PCG,OPTIONS,
min_pacbporf_identityscore=0.20,verbose=True):
""" """
if not signalpexonseqs.has_key(OPTIONS.target): return False
is_any_pacbporf_added = False
for targetSPexon in signalpexonseqs[OPTIONS.target]:
target = OPTIONS.target
for informant,infSPlist in signalpexonseqs.iteritems():
if informant == OPTIONS.target: continue
# check if informant has been deleted in the meanwhile
if informant not in PCG.organism_set(): continue
# list to store signalp exons into
signalpexon_pacbp_list = []
# get ordered pacbporfs fromt he PCG
thepacbporfs = order_pacbporf_list(PCG.get_pacbps_by_organisms(OPTIONS.target,informant))
if not thepacbporfs:
# no alignments present for this organism (can happen!)
continue
for informantSPexon in infSPlist:
coords = [ targetSPexon.protein_start(),
targetSPexon.protein_end(),
informantSPexon.protein_start(),
informantSPexon.protein_end(), ]
# prior to making ClustalW-PacbP, check PacbPCOORD placeability
# into the list of pacbporfs
pacbpCoordsObj = PacbPCOORDS(input=(
targetSPexon.proteinsequence(),
informantSPexon.proteinsequence(),
targetSPexon.protein_start(),
informantSPexon.protein_start(),
) )
if False in [ pacbpCoordsObj.is_positioned_compatibly(pacbporf) for pacbporf in thepacbporfs ]:
# *NOT* placable in current ordered list of PacbPORFS
continue
dist = pacbpCoordsObj.distance_towards(thepacbporfs[0])
if dist > SIGNALP_FIRSTEXON_MAX_INTRON_NT_LENGTH/3:
# WAY TO FAR in front of current gene structure parts.
# Do not allow (pooras a *NOT* placable in current ordered list of PacbPORFS
continue
elif dist == 0:
# NOT placeable in front of the rest of the PacbPORFS.
continue
else:
pass
# perform ClustalW alignment on the SP exons
(alignedseqs,alignment) =\
clustalw( seqs= {
OPTIONS.target: targetSPexon.proteinsequence(),
informant: informantSPexon.proteinsequence() } )
# make pacbp from clustalw alignment
pacbp = pacbp_from_clustalw(
alignment=(
alignedseqs[OPTIONS.target],
alignment,
alignedseqs[informant]
),
coords=coords
)
# is there any alignment constructed?
if not pacbp: continue
# ignore (very) poor identyscore alignments
if pacbp.identityscore < min_pacbporf_identityscore: continue
# if here make extended pacbpORF
signalpexonPacbpORF = pacbp2pacbporf(pacbp,
targetSPexon.orf,informantSPexon.orf)
signalpexonPacbpORF.extend_pacbporf_after_stops()
# and store in signalpexon_pacbp_list
signalpexon_pacbp_list.append( signalpexonPacbpORF )
################################################################
if verbose:
print alignedseqs[OPTIONS.target], OPTIONS.target
print alignment
print alignedseqs[informant], informant
if pacbp:
print pacbp, (OPTIONS.target, targetSPexon.orf.id),
print (informant, informantSPexon.orf.id),
print "DISTANCE::", dist
pacbp.print_protein()
print ""
################################################################
# If there are signalpexon-guided pacbporfs found, store the one
# with the highest bitscore
if signalpexon_pacbp_list:
signalpexon_pacbp_list = order_list_by_attribute(
signalpexon_pacbp_list,order_by='bits',reversed=True)
# store best bitscoring pacbporf to PCG
signalp_pacbporf = signalpexon_pacbp_list[0]
pacbporf2PCG(signalp_pacbporf,OPTIONS.target,informant,PCG,source='SignalP-ClustalW')
is_any_pacbporf_added = True
####################################################################
if verbose:
print "SignalP Exon added to PCG:", signalp_pacbporf, informant
####################################################################
else:
pass
# return pointer is_any_pacbporf_added
return is_any_pacbporf_added
def hmmpacbporf2PCG(hmmpacbporf, target, informant, PCG, source=''):
""" """
hmmpacbporf.source = 'HMM'
hmmpacbporf._gff['fsource'] = 'HMM'
pacbporf2PCG(hmmpacbporf, target, informant, PCG, source=source)
def hmmpacbporf2PCG(hmmpacbporf,target,informant,PCG,source=''):
""" """
hmmpacbporf.source = 'HMM'
hmmpacbporf._gff['fsource'] = 'HMM'
pacbporf2PCG(hmmpacbporf,target,informant,PCG,source=source)