def PCG2similarityarray(PCG,organism,aalength,omit_unigenes=True):
""" """
array_algsimilarity = zeros(aalength)
for orgS in PCG.organism_set():
if organism == orgS: continue
pacbporfs = order_pacbporf_list(PCG.get_pacbps_by_organisms(organism,orgS))
if pacbporfs and omit_unigenes and hasattr(pacbporfs[0].orfS,ORF_IS_UNIGENE_LABEL):
continue
orgSimArray = pacbporflist2similarityarray(pacbporfs,"query",aalength)
array_algsimilarity+=orgSimArray
# return similarity array
return array_algsimilarity
def PCG2codingarray(PCG, organism, aalength, omit_unigenes=True):
""" """
array_algpresence = zeros(aalength)
for orgS in PCG.organism_set():
if organism == orgS: continue
pacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(organism, orgS))
if pacbporfs and omit_unigenes and hasattr(pacbporfs[0].orfS,
ORF_IS_UNIGENE_LABEL):
continue
orgPresArray = pacbporflist2codingarray(pacbporfs, "query", aalength)
array_algpresence += orgPresArray
# return coding/presence array
return array_algpresence
def _convert_tinyexon_proteinmatches_to_pacbporfs(target,
protmatches,
tinyexondata,
PCG,
min_discovery_count=2):
""" """
target_tinyexon_pacbporf_data = {}
# fish these protein matches from the tinyexons and convert to PacbPORFs
for informant in tinyexondata.keys():
if informant == target: continue
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(target, informant))
for exonQ in tinyexondata[target]:
if exonQ.orf.id in [pf.orfQ.id for pf in thepacbporfs]: continue
if exonQ.proteinsequence() not in protmatches.keys(): continue
for exonS in tinyexondata[informant]:
if exonS.length > exonQ.length: break
if exonS.proteinsequence() not in protmatches.keys(): continue
# omit non-identical exons
if not _are_tinyexons_similar(exonQ, exonS): continue
# if here: similar exons. make PacbPORF
pacbporf = exononorfs2pacbporf(exonQ,
exonS,
matrix=TINYEXON_MATRIX)
if not pacbporf: continue
# check if placeable in PCG/pacbporflist
rejected = [
pf.is_postioned_compatibly(pacbporf) for pf in thepacbporfs
].count(False) > 0
# label pacbporf as found by tinyexon PP
pacbporf._tinyexon_label = "PP"
# store to target_tinyexon_pacbporf_data
key = (exonQ.proteinsequence(), exonQ.start)
_update_tinyexon_pacbporf_dict(target_tinyexon_pacbporf_data,
key, pacbporf, rejected,
informant)
# cleanup tinyexon protein matches that have been observed to litte
_remove_dict_elements_with_short_value_list(
target_tinyexon_pacbporf_data, min_value_list_size=min_discovery_count)
# return target_tinyexon_pacbporf_data
return target_tinyexon_pacbporf_data
def _has_pp_tinyexonpacbporf_perfect_introns(tinyexonPF,target,informant,PCG):
""" """
# check if a (perfect) introns can be mapped
is_confirmed_with_introns = False
if tinyexonPF._tinyexon_label != 'PP': return False
# get ordered PacbPORFS for this informant
thepacbporfs = order_pacbporf_list(PCG.get_pacbps_by_organisms(target,informant))
for pos in range(1,len(thepacbporfs)):
prevPF,nextPF = thepacbporfs[pos-1],thepacbporfs[pos]
if prevPF.distance_towards(tinyexonPF) > 0 and\
tinyexonPF.distance_towards(nextPF) > 0:
intronsPREV = merge_pacbporfs_with_introns(
prevPF,tinyexonPF,max_aa_offset=0,
max_intron_nt_length=None)
intronsNEXT = merge_pacbporfs_with_introns(
tinyexonPF,nextPF,max_aa_offset=0,
max_intron_nt_length=None)
if len(intronsPREV) >= 1 and len(intronsNEXT) >= 1:
perfect_prev_intron = False
perfect_next_intron = False
for intronQ,intronS in intronsPREV:
intronQ.assign_bp_and_ppts()
intronS.assign_bp_and_ppts()
if intronQ.branchpoint and intronS.branchpoint:
perfect_prev_intron = True
break
for intronQ,intronS in intronsNEXT:
intronQ.assign_bp_and_ppts()
intronS.assign_bp_and_ppts()
if intronQ.branchpoint and intronS.branchpoint:
perfect_next_intron = True
break
# check if both intron options have a perfect candidate
if perfect_prev_intron and perfect_next_intron:
is_confirmed_with_introns = True
# break out
break
# return is_confirmed_with_introns status
return is_confirmed_with_introns
def _has_pp_tinyexonpacbporf_perfect_introns(tinyexonPF, target, informant,
PCG):
""" """
# check if a (perfect) introns can be mapped
is_confirmed_with_introns = False
if tinyexonPF._tinyexon_label != 'PP': return False
# get ordered PacbPORFS for this informant
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(target, informant))
for pos in range(1, len(thepacbporfs)):
prevPF, nextPF = thepacbporfs[pos - 1], thepacbporfs[pos]
if prevPF.distance_towards(tinyexonPF) > 0 and\
tinyexonPF.distance_towards(nextPF) > 0:
intronsPREV = merge_pacbporfs_with_introns(
prevPF, tinyexonPF, max_aa_offset=0, max_intron_nt_length=None)
intronsNEXT = merge_pacbporfs_with_introns(
tinyexonPF, nextPF, max_aa_offset=0, max_intron_nt_length=None)
if len(intronsPREV) >= 1 and len(intronsNEXT) >= 1:
perfect_prev_intron = False
perfect_next_intron = False
for intronQ, intronS in intronsPREV:
intronQ.assign_bp_and_ppts()
intronS.assign_bp_and_ppts()
if intronQ.branchpoint and intronS.branchpoint:
perfect_prev_intron = True
break
for intronQ, intronS in intronsNEXT:
intronQ.assign_bp_and_ppts()
intronS.assign_bp_and_ppts()
if intronQ.branchpoint and intronS.branchpoint:
perfect_next_intron = True
break
# check if both intron options have a perfect candidate
if perfect_prev_intron and perfect_next_intron:
is_confirmed_with_introns = True
# break out
break
# return is_confirmed_with_introns status
return is_confirmed_with_introns
def _convert_tinyexon_proteinmatches_to_pacbporfs(target,protmatches,
tinyexondata,PCG,min_discovery_count=2):
""" """
target_tinyexon_pacbporf_data = {}
# fish these protein matches from the tinyexons and convert to PacbPORFs
for informant in tinyexondata.keys():
if informant == target: continue
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(target,informant))
for exonQ in tinyexondata[target]:
if exonQ.orf.id in [ pf.orfQ.id for pf in thepacbporfs ]: continue
if exonQ.proteinsequence() not in protmatches.keys(): continue
for exonS in tinyexondata[informant]:
if exonS.length > exonQ.length: break
if exonS.proteinsequence() not in protmatches.keys(): continue
# omit non-identical exons
if not _are_tinyexons_similar(exonQ,exonS): continue
# if here: similar exons. make PacbPORF
pacbporf = exononorfs2pacbporf(exonQ,exonS,matrix=TINYEXON_MATRIX)
if not pacbporf: continue
# check if placeable in PCG/pacbporflist
rejected = [ pf.is_postioned_compatibly(pacbporf) for pf in thepacbporfs ].count(False) > 0
# label pacbporf as found by tinyexon PP
pacbporf._tinyexon_label = "PP"
# store to target_tinyexon_pacbporf_data
key = (exonQ.proteinsequence(),exonQ.start)
_update_tinyexon_pacbporf_dict(target_tinyexon_pacbporf_data,
key,pacbporf,rejected,informant)
# cleanup tinyexon protein matches that have been observed to litte
_remove_dict_elements_with_short_value_list(
target_tinyexon_pacbporf_data,
min_value_list_size=min_discovery_count)
# return target_tinyexon_pacbporf_data
return target_tinyexon_pacbporf_data
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 _find_qq_tinyexons_as_pacbporfs(target,tinyexondata,PCG,min_discovery_count=2):
""" """
target_tinyexon_pacbporf_data = {}
for informant in tinyexondata.keys():
if informant == target: continue
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(target,informant))
for exonQ in tinyexondata[target]:
if exonQ.orf.id in [ pf.orfQ.id for pf in thepacbporfs ]: continue
for (prevpos,nextpos) in [ (pos-1,pos) for pos in range(1,len(thepacbporfs)) ]:
prevPF = thepacbporfs[prevpos]
nextPF = thepacbporfs[nextpos]
if prevPF.orfS.id == nextPF.orfS.id:
# check if PacbPORFs are positioned more or less okay
if prevPF.distance_towards(nextPF) > 20: continue
# check if exonQ is positioned ~between these PacbPORFs
if exonQ.orf.dnapos2aapos(exonQ.end) < max(prevPF.alignment_protein_range_query())-12:
continue
if exonQ.orf.dnapos2aapos(exonQ.start) > min(nextPF.alignment_protein_range_query())+12:
continue
# check if gap can be projected already by a perfect intron
introns = merge_pacbporfs_by_intron_in_query(
prevPF,nextPF,max_aa_offset=1)
# if introns found => continue
if introns: continue
# orfObj is the orfS of prevPF or nextPF (just take any)
orfObj = prevPF.orfS
# assign elegiable range of tinyexon match on SBJCT
aapos_sbjct_range = range(
max(prevPF.alignment_protein_range_sbjct())-12,
min(nextPF.alignment_protein_range_sbjct())+12
)
tinyexonmatches = _find_match_on_orfobj(exonQ,orfObj)
for (aaseq,aapos) in tinyexonmatches:
# check if the match is obtained in the expected
# sbjct AA range; if not, ignore the match
if aapos not in aapos_sbjct_range: continue
# make pacbporf object
pacbpobj = PacbP(input=(
exonQ.proteinsequence(), aaseq,
exonQ.orf.dnapos2aapos(exonQ.start), aapos ) )
pacbporfobj = pacbp2pacbporf(pacbpobj,exonQ.orf,orfObj)
pacbporfobj.extend_pacbporf_after_stops()
# remove included pacbporfs
is_suborsuperset = False
for accepted_pacbporf in thepacbporfs:
if pacbporfobj.issubsetorsuperset(accepted_pacbporf):
is_suborsuperset = True
break
if is_suborsuperset:
continue
# check if 2 (perfect) introns can be projected
introns5p = merge_pacbporfs_by_intron_in_query(
prevPF,pacbporfobj,
max_aa_offset=1,
max_intron_nt_length=None)
#max_intron_nt_length=140)
introns3p = merge_pacbporfs_by_intron_in_query(
pacbporfobj,nextPF,
max_aa_offset=1,
max_intron_nt_length=None)
#max_intron_nt_length=140)
# continue if not is_confirmed_by_intron_projection
if not introns5p or not introns3p: continue
# check if placeable in PCG/pacbporflist
distPrev = prevPF.distance_towards(pacbporfobj)
distNext = pacbporfobj.distance_towards(nextPF)
ovrlPrev = pacbporfobj.overlap(prevPF)
ovrlNext = pacbporfobj.overlap(nextPF)
if distPrev and distNext:
rejected = False
elif not distPrev and ovrlPrev:
rejected = False
elif not distNext and ovrlNext:
rejected = False
elif ovrlPrev and ovrlNext:
rejected = False
else:
rejected = True
print "OKAY", exonQ.proteinsequence(), aaseq, rejected, informant, (distPrev,distNext,ovrlPrev,ovrlNext)
# label pacbporf as found by tinyexon QQ
pacbporfobj._tinyexon_label = "QQ"
# store to target_tinyexon_pacbporf_data
key = (exonQ.proteinsequence(),exonQ.start)
_update_tinyexon_pacbporf_dict(
target_tinyexon_pacbporf_data,
key,pacbporfobj,rejected,informant)
# cleanup tinyexon protein matches that have been observed to litte
_remove_dict_elements_with_short_value_list(
target_tinyexon_pacbporf_data,
min_value_list_size=min_discovery_count)
# return target_tinyexon_pacbporf_data
return target_tinyexon_pacbporf_data
def _find_qp_and_pq_tinyexons_as_pacbporfs(target,tinyexondata,PCG,min_discovery_count=2):
""" """
target_tinyexon_pacbporf_data = {}
for informant in tinyexondata.keys():
if informant == target: continue
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(target,informant))
for exonQ in tinyexondata[target]:
if exonQ.orf.id in [ pf.orfQ.id for pf in thepacbporfs ]: continue
for orfObj in PCG.get_orfs_of_graph(organism=informant):
tinyexonmatches = _find_qp_or_pq_match_on_orfobj(exonQ,orfObj)
for (aaseq,aapos) in tinyexonmatches:
# make pacbporf object
pacbpobj = PacbP(input=(
exonQ.proteinsequence(), aaseq,
exonQ.orf.dnapos2aapos(exonQ.start), aapos ) )
pacbporfobj = pacbp2pacbporf(pacbpobj,exonQ.orf,orfObj)
pacbporfobj.extend_pacbporf_after_stops()
# remove included pacbporfs
is_suborsuperset = False
for accepted_pacbporf in thepacbporfs:
if pacbporfobj.issubsetorsuperset(accepted_pacbporf):
is_suborsuperset = True
break
if is_suborsuperset:
continue
# check if a (perfect) intron can be projected
is_confirmed_by_intron_projection = False
for accepted_pacbporf in thepacbporfs:
if accepted_pacbporf.orfS.id == pacbporfobj.orfS.id:
if min(accepted_pacbporf.alignment_dna_range_query()) > min(pacbporfobj.alignment_dna_range_query()):
try:
introns = merge_pacbporfs_by_intron_in_query(
pacbporfobj,accepted_pacbporf,
max_aa_offset=0,
max_intron_nt_length=None)
#max_intron_nt_length=140)
except IndexError:
# unexpected event: TODO: solve in merge_pacbporfs_by_intron_in_query
introns = []
else:
try:
introns = merge_pacbporfs_by_intron_in_query(
accepted_pacbporf,pacbporfobj,
max_aa_offset=0,
max_intron_nt_length=None)
#max_intron_nt_length=140)
except IndexError:
# unexpected event: TODO: solve in merge_pacbporfs_by_intron_in_query
introns = []
if len(introns) >= 1:
is_confirmed_by_intron_projection = True
break
# continue if not is_confirmed_by_intron_projection
if not is_confirmed_by_intron_projection: continue
# check if placeable in PCG/pacbporflist
rejected = [ pf.is_postioned_compatibly(pacbporfobj) for pf in thepacbporfs ].count(False) > 0
# label pacbporf as found by tinyexon QP
pacbporfobj._tinyexon_label = "QP"
# store to target_tinyexon_pacbporf_data
key = (exonQ.proteinsequence(),exonQ.start)
_update_tinyexon_pacbporf_dict(
target_tinyexon_pacbporf_data,
key,pacbporfobj,rejected,informant)
# cleanup tinyexon protein matches that have been observed to litte
_remove_dict_elements_with_short_value_list(
target_tinyexon_pacbporf_data,
min_value_list_size=min_discovery_count)
# return target_tinyexon_pacbporf_data
return target_tinyexon_pacbporf_data
def _find_qq_tinyexons_as_pacbporfs(target,
tinyexondata,
PCG,
min_discovery_count=2):
""" """
target_tinyexon_pacbporf_data = {}
for informant in tinyexondata.keys():
if informant == target: continue
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(target, informant))
for exonQ in tinyexondata[target]:
if exonQ.orf.id in [pf.orfQ.id for pf in thepacbporfs]: continue
for (prevpos, nextpos) in [(pos - 1, pos)
for pos in range(1, len(thepacbporfs))]:
prevPF = thepacbporfs[prevpos]
nextPF = thepacbporfs[nextpos]
if prevPF.orfS.id == nextPF.orfS.id:
# check if PacbPORFs are positioned more or less okay
if prevPF.distance_towards(nextPF) > 20: continue
# check if exonQ is positioned ~between these PacbPORFs
if exonQ.orf.dnapos2aapos(exonQ.end) < max(
prevPF.alignment_protein_range_query()) - 12:
continue
if exonQ.orf.dnapos2aapos(exonQ.start) > min(
nextPF.alignment_protein_range_query()) + 12:
continue
# check if gap can be projected already by a perfect intron
introns = merge_pacbporfs_by_intron_in_query(
prevPF, nextPF, max_aa_offset=1)
# if introns found => continue
if introns: continue
# orfObj is the orfS of prevPF or nextPF (just take any)
orfObj = prevPF.orfS
# assign elegiable range of tinyexon match on SBJCT
aapos_sbjct_range = range(
max(prevPF.alignment_protein_range_sbjct()) - 12,
min(nextPF.alignment_protein_range_sbjct()) + 12)
tinyexonmatches = _find_match_on_orfobj(exonQ, orfObj)
for (aaseq, aapos) in tinyexonmatches:
# check if the match is obtained in the expected
# sbjct AA range; if not, ignore the match
if aapos not in aapos_sbjct_range: continue
# make pacbporf object
pacbpobj = PacbP(
input=(exonQ.proteinsequence(), aaseq,
exonQ.orf.dnapos2aapos(exonQ.start), aapos))
pacbporfobj = pacbp2pacbporf(pacbpobj, exonQ.orf,
orfObj)
pacbporfobj.extend_pacbporf_after_stops()
# remove included pacbporfs
is_suborsuperset = False
for accepted_pacbporf in thepacbporfs:
if pacbporfobj.issubsetorsuperset(
accepted_pacbporf):
is_suborsuperset = True
break
if is_suborsuperset:
continue
# check if 2 (perfect) introns can be projected
introns5p = merge_pacbporfs_by_intron_in_query(
prevPF,
pacbporfobj,
max_aa_offset=1,
max_intron_nt_length=None)
#max_intron_nt_length=140)
introns3p = merge_pacbporfs_by_intron_in_query(
pacbporfobj,
nextPF,
max_aa_offset=1,
max_intron_nt_length=None)
#max_intron_nt_length=140)
# continue if not is_confirmed_by_intron_projection
if not introns5p or not introns3p: continue
# check if placeable in PCG/pacbporflist
distPrev = prevPF.distance_towards(pacbporfobj)
distNext = pacbporfobj.distance_towards(nextPF)
ovrlPrev = pacbporfobj.overlap(prevPF)
ovrlNext = pacbporfobj.overlap(nextPF)
if distPrev and distNext:
rejected = False
elif not distPrev and ovrlPrev:
rejected = False
elif not distNext and ovrlNext:
rejected = False
elif ovrlPrev and ovrlNext:
rejected = False
else:
rejected = True
print "OKAY", exonQ.proteinsequence(
), aaseq, rejected, informant, (distPrev, distNext,
ovrlPrev, ovrlNext)
# label pacbporf as found by tinyexon QQ
pacbporfobj._tinyexon_label = "QQ"
# store to target_tinyexon_pacbporf_data
key = (exonQ.proteinsequence(), exonQ.start)
_update_tinyexon_pacbporf_dict(
target_tinyexon_pacbporf_data, key, pacbporfobj,
rejected, informant)
# cleanup tinyexon protein matches that have been observed to litte
_remove_dict_elements_with_short_value_list(
target_tinyexon_pacbporf_data, min_value_list_size=min_discovery_count)
# return target_tinyexon_pacbporf_data
return target_tinyexon_pacbporf_data
def _find_qp_and_pq_tinyexons_as_pacbporfs(target,
tinyexondata,
PCG,
min_discovery_count=2):
""" """
target_tinyexon_pacbporf_data = {}
for informant in tinyexondata.keys():
if informant == target: continue
thepacbporfs = order_pacbporf_list(
PCG.get_pacbps_by_organisms(target, informant))
for exonQ in tinyexondata[target]:
if exonQ.orf.id in [pf.orfQ.id for pf in thepacbporfs]: continue
for orfObj in PCG.get_orfs_of_graph(organism=informant):
tinyexonmatches = _find_qp_or_pq_match_on_orfobj(exonQ, orfObj)
for (aaseq, aapos) in tinyexonmatches:
# make pacbporf object
pacbpobj = PacbP(
input=(exonQ.proteinsequence(), aaseq,
exonQ.orf.dnapos2aapos(exonQ.start), aapos))
pacbporfobj = pacbp2pacbporf(pacbpobj, exonQ.orf, orfObj)
pacbporfobj.extend_pacbporf_after_stops()
# remove included pacbporfs
is_suborsuperset = False
for accepted_pacbporf in thepacbporfs:
if pacbporfobj.issubsetorsuperset(accepted_pacbporf):
is_suborsuperset = True
break
if is_suborsuperset:
continue
# check if a (perfect) intron can be projected
is_confirmed_by_intron_projection = False
for accepted_pacbporf in thepacbporfs:
if accepted_pacbporf.orfS.id == pacbporfobj.orfS.id:
if min(accepted_pacbporf.alignment_dna_range_query(
)) > min(pacbporfobj.alignment_dna_range_query()):
try:
introns = merge_pacbporfs_by_intron_in_query(
pacbporfobj,
accepted_pacbporf,
max_aa_offset=0,
max_intron_nt_length=None)
#max_intron_nt_length=140)
except IndexError:
# unexpected event: TODO: solve in merge_pacbporfs_by_intron_in_query
introns = []
else:
try:
introns = merge_pacbporfs_by_intron_in_query(
accepted_pacbporf,
pacbporfobj,
max_aa_offset=0,
max_intron_nt_length=None)
#max_intron_nt_length=140)
except IndexError:
# unexpected event: TODO: solve in merge_pacbporfs_by_intron_in_query
introns = []
if len(introns) >= 1:
is_confirmed_by_intron_projection = True
break
# continue if not is_confirmed_by_intron_projection
if not is_confirmed_by_intron_projection: continue
# check if placeable in PCG/pacbporflist
rejected = [
pf.is_postioned_compatibly(pacbporfobj)
for pf in thepacbporfs
].count(False) > 0
# label pacbporf as found by tinyexon QP
pacbporfobj._tinyexon_label = "QP"
# store to target_tinyexon_pacbporf_data
key = (exonQ.proteinsequence(), exonQ.start)
_update_tinyexon_pacbporf_dict(
target_tinyexon_pacbporf_data, key, pacbporfobj,
rejected, informant)
# cleanup tinyexon protein matches that have been observed to litte
_remove_dict_elements_with_short_value_list(
target_tinyexon_pacbporf_data, min_value_list_size=min_discovery_count)
# return target_tinyexon_pacbporf_data
return target_tinyexon_pacbporf_data
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
