def merge_pacbporfs_with_query_intron_bridgeing(pacbporfD,
pacbporfA,
verbose=False,
**kwargs):
"""
Merge query Orfs in PacbPORF by **best** intron
@attention: see orfs.merge_orfs_with_intron for **kwargs
@type pacbporfD: PacbPORF object
@param pacbporfD: PacbPORF object that has to deliver PSSM donor objects
@type pacbporfA: PacbPORF object
@param pacbporfA: PacbPORF object that has to deliver PSSM acceptor objects
@type verbose: Boolean
@param verbose: print status/debugging messages to STDOUT
@rtype: list
@return: list with ( intron, intron ), in query and sbjct
"""
# input validation
IsPacbPORF(pacbporfD)
IsPacbPORF(pacbporfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs, KWARGS_MAPPED_INTRON)
if not kwargs.has_key('aligned_site_max_triplet_distance'):
kwargs['aligned_site_max_triplet_distance'] = kwargs['max_aa_offset']
# calculate maximal/minimal donor/acceptor site position based on alignment
ELEGIABLE_SPLICE_SITE_AA_RANGE = 75
qdr = pacbporfD.alignment_dna_range_query()
qar = pacbporfA.alignment_dna_range_query()
min_donor_query_pos = max(
[min(qdr), max(qdr) - (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
max_accep_query_pos = min(
[max(qar), min(qar) + (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
# get list of introns
intronlist = merge_orfs_with_intron(pacbporfD.orfQ,
pacbporfA.orfQ,
min_donor_pos=min_donor_query_pos,
max_acceptor_pos=max_accep_query_pos,
**kwargs)
# filter on entropy
# settings for minimal alignment entropy score
if min([pacbporfD.identityscore, pacbporfA.identityscore]) > 0.55:
min_donor_site_entropy = 0.01
min_acceptor_site_entropy = 0.01
intronlist = _filter_introns_on_entropy(
intronlist,
pacbporfD,
pacbporfA,
min_donor_site_entropy=min_donor_site_entropy,
min_acceptor_site_entropy=min_acceptor_site_entropy)
else:
# do not filter, but do not forget to store apps data to intron(s)
for intron in intronlist:
succes = set_apps_intron_query(intron, pacbporfD, pacbporfA)
for intron in intronlist:
intron._distance = 0 # ??
# set GFF fsource attribute for recognition of intron sources
intron._gff['fsource'] = 'ABGPbridgeing'
# get unique list of donors & acceptors
donor = olba(list(Set([intron.donor for intron in intronlist])),
order_by='pos')
accep = olba(list(Set([intron.acceptor for intron in intronlist])),
order_by='pos')
############################################################################
if verbose:
print "dQ1", [d.pos for d in donor], "aQ1", [a.pos for a in accep]
############################################################################
intronlist = _filter_introns_on_pssm_entropy_combination(intronlist)
# get unique list of donors & acceptors
donor = olba(list(Set([intron.donor for intron in intronlist])),
order_by='pos')
accep = olba(list(Set([intron.acceptor for intron in intronlist])),
order_by='pos')
############################################################################
if verbose:
print "dQ1", [d.pos for d in donor], "aQ1", [a.pos for a in accep]
############################################################################
filtered_intron_list = []
for intron in intronlist:
intron.assign_bp_and_ppts()
if intron.branchpoint and (intron.ppt5p or intron.ppt3p):
filtered_intron_list.append(intron)
else:
pass
# check if list is emptied due to branchpoint filtering
# in that case, filter for either branchpoint OR polyppt
if not filtered_intron_list and intronlist:
for intron in intronlist:
if intron.branchpoint or (intron.ppt5p or intron.ppt3p):
filtered_intron_list.append(intron)
# return list of filtered introns
return filtered_intron_list
def merge_pacbporfs_with_query_intron_bridgeing(pacbporfD,pacbporfA,verbose=False,**kwargs):
"""
Merge query Orfs in PacbPORF by **best** intron
@attention: see orfs.merge_orfs_with_intron for **kwargs
@type pacbporfD: PacbPORF object
@param pacbporfD: PacbPORF object that has to deliver PSSM donor objects
@type pacbporfA: PacbPORF object
@param pacbporfA: PacbPORF object that has to deliver PSSM acceptor objects
@type verbose: Boolean
@param verbose: print status/debugging messages to STDOUT
@rtype: list
@return: list with ( intron, intron ), in query and sbjct
"""
# input validation
IsPacbPORF(pacbporfD)
IsPacbPORF(pacbporfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs,KWARGS_MAPPED_INTRON)
if not kwargs.has_key('aligned_site_max_triplet_distance'):
kwargs['aligned_site_max_triplet_distance'] = kwargs['max_aa_offset']
# calculate maximal/minimal donor/acceptor site position based on alignment
ELEGIABLE_SPLICE_SITE_AA_RANGE = 75
qdr = pacbporfD.alignment_dna_range_query()
qar = pacbporfA.alignment_dna_range_query()
min_donor_query_pos = max([ min(qdr), max(qdr)-(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
max_accep_query_pos = min([ max(qar), min(qar)+(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
# get list of introns
intronlist = merge_orfs_with_intron(pacbporfD.orfQ,pacbporfA.orfQ,
min_donor_pos =min_donor_query_pos,
max_acceptor_pos=max_accep_query_pos,**kwargs)
# filter on entropy
# settings for minimal alignment entropy score
if min([pacbporfD.identityscore,pacbporfA.identityscore]) > 0.55:
min_donor_site_entropy = 0.01
min_acceptor_site_entropy = 0.01
intronlist = _filter_introns_on_entropy(intronlist,pacbporfD,pacbporfA,
min_donor_site_entropy=min_donor_site_entropy,
min_acceptor_site_entropy=min_acceptor_site_entropy)
else:
# do not filter, but do not forget to store apps data to intron(s)
for intron in intronlist:
succes = set_apps_intron_query(intron,pacbporfD,pacbporfA)
for intron in intronlist:
intron._distance = 0 # ??
# set GFF fsource attribute for recognition of intron sources
intron._gff['fsource'] = 'ABGPbridgeing'
# get unique list of donors & acceptors
donor = olba( list(Set([intron.donor for intron in intronlist ])), order_by='pos')
accep = olba( list(Set([intron.acceptor for intron in intronlist ])), order_by='pos')
############################################################################
if verbose: print "dQ1",[d.pos for d in donor],"aQ1",[a.pos for a in accep]
############################################################################
intronlist = _filter_introns_on_pssm_entropy_combination(intronlist)
# get unique list of donors & acceptors
donor = olba( list(Set([intron.donor for intron in intronlist ])), order_by='pos')
accep = olba( list(Set([intron.acceptor for intron in intronlist ])), order_by='pos')
############################################################################
if verbose: print "dQ1",[d.pos for d in donor],"aQ1",[a.pos for a in accep]
############################################################################
filtered_intron_list = []
for intron in intronlist:
intron.assign_bp_and_ppts()
if intron.branchpoint and (intron.ppt5p or intron.ppt3p):
filtered_intron_list.append( intron )
else:
pass
# check if list is emptied due to branchpoint filtering
# in that case, filter for either branchpoint OR polyppt
if not filtered_intron_list and intronlist:
for intron in intronlist:
if intron.branchpoint or (intron.ppt5p or intron.ppt3p):
filtered_intron_list.append( intron )
# return list of filtered introns
return filtered_intron_list
def merge_pacbporfs_by_tinyexons(pacbporfD,pacbporfA,
orfSetObjQ,orfSetObjS,verbose=False,**kwargs):
""" """
# input validation
IsPacbPORF(pacbporfD)
IsPacbPORF(pacbporfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs,KWARGS_MAPPED_INTRON)
if not kwargs.has_key('aligned_site_max_triplet_distance'):
kwargs['aligned_site_max_triplet_distance'] = kwargs['max_aa_offset']
# settings for minimal alignment entropy score
min_donor_site_alignment_entropy = 0.0
min_acceptor_site_alignment_entropy = 0.0
resultlistQ = merge_orfs_with_tinyexon(
pacbporfD.orfQ,pacbporfA.orfQ,
preceding_donor_sites=pacbporfD.orfQ._donor_sites,
subsequent_acceptor_sites=pacbporfA.orfQ._acceptor_sites,
orflist=orfSetObjQ.orfs,**kwargs)
resultlistS = merge_orfs_with_tinyexon(
pacbporfD.orfS,pacbporfA.orfS,
preceding_donor_sites=pacbporfD.orfS._donor_sites,
subsequent_acceptor_sites=pacbporfA.orfS._acceptor_sites,
orflist=orfSetObjS.orfs,**kwargs)
# translate resultlists to dict: key == exon, value = [ {intronsD},{intronsS} ]
resultdictQ,key2exonQ = _tinyexon_list_2_dict(resultlistQ)
resultdictS,key2exonS = _tinyexon_list_2_dict(resultlistS)
# get unique list of donors & acceptors
donorQ = olba( list(Set([inD.donor for inD,te,inA in resultlistQ ])), order_by='pos')
donorS = olba( list(Set([inD.donor for inD,te,inA in resultlistS ])), order_by='pos')
accepQ = olba( list(Set([inA.acceptor for inD,te,inA in resultlistQ ])), order_by='pos')
accepS = olba( list(Set([inA.acceptor for inD,te,inA in resultlistS ])), order_by='pos')
## filter for alignable donor & acceptor sites
kwargs['allow_non_canonical'] = True # True
kwargs['aligned_site_max_triplet_distance'] = 0 # 2
algdonors = _filter_for_alignable_splice_sites(donorQ,donorS,pacbporfD,**kwargs)
algacceps = _filter_for_alignable_splice_sites(accepQ,accepS,pacbporfA,**kwargs)
# settings for minimal alignment entropy score
# TODO TODO -> THIS MUST BE FIXED TO A NICE THRESHOLD VALUE!!!
min_donor_site_alignment_entropy = 0.1
min_acceptor_site_alignment_entropy = 0.1
# remove sites with to low alignment entropy
algdonors = _filter_for_entropy(algdonors,pacbporfD,'donor',
min_alignment_entropy=min_donor_site_alignment_entropy)
algacceps = _filter_for_entropy(algacceps,pacbporfA,'acceptor',
min_alignment_entropy=min_acceptor_site_alignment_entropy)
# return list: intronQD,intronSD,tinyexon,intronAQ,intronAS
return_list = []
############################################################################
if verbose:
print "bridges constructed: ORFS:",
print (pacbporfD.orfQ.id,pacbporfA.orfQ.id),
print (pacbporfD.orfS.id,pacbporfA.orfS.id),
print len(resultdictQ), len(resultdictS),
print ( len(resultlistQ), len(donorQ), len(accepQ) ),
print ( len(resultlistS), len(donorS), len(accepS) ),
print ( len(algdonors), len(algacceps) )
############################################################################
for keyQ,tinyexonQ in key2exonQ.iteritems():
for keyS,tinyexonS in key2exonS.iteritems():
if tinyexonQ.donor.phase != tinyexonS.donor.phase:
continue
if tinyexonQ.acceptor.phase != tinyexonS.acceptor.phase:
continue
if tinyexonQ.length != tinyexonS.length:
continue
# if here, then tinyexons of identical structure
####################################################################
if verbose:
print tinyexonQ.length, tinyexonQ.donor.phase,
print ( len(resultdictQ[keyQ][0]), len(resultdictQ[keyQ][1]) ),
print ( len(resultdictS[keyS][0]), len(resultdictS[keyS][1]) ),
print tinyexonQ,
print tinyexonQ.proteinsequence(), tinyexonS.proteinsequence(),
print tinyexonS.acceptor.pssm_score + tinyexonS.donor.pssm_score
####################################################################
donor_introns = []
acceptor_introns = []
for intronDQkey, intronDQ in resultdictQ[keyQ][0].iteritems():
if intronDQ.donor.pos not in [ dQ.pos for dQ,dS in algdonors ]:
continue
for intronDSkey, intronDS in resultdictS[keyS][0].iteritems():
if intronDS.donor.pos not in [ dS.pos for dQ,dS in algdonors ]:
continue
# check if they exists as aligned sites
alignedkey = ( intronDQ.donor.pos, intronDS.donor.pos )
if alignedkey not in [ (dQ.pos, dS.pos) for dQ,dS in algdonors ]:
continue
# if here, we have a set of introns 5' of the tinyexon
# which are perfectly alignable!
donor_introns.append((intronDQ,intronDS))
for intronAQkey, intronAQ in resultdictQ[keyQ][1].iteritems():
if intronAQ.acceptor.pos not in [ aQ.pos for aQ,aS in algacceps ]:
continue
for intronASkey, intronAS in resultdictS[keyS][1].iteritems():
if intronAS.acceptor.pos not in [ aS.pos for aQ,aS in algacceps ]:
continue
# check if they exists as aligned sites
alignedkey = ( intronAQ.acceptor.pos, intronAS.acceptor.pos )
if alignedkey not in [ (aQ.pos, aS.pos) for aQ,aS in algacceps ]:
continue
# if here, we have a set of introns 3' of the tinyexon
# which are perfectly alignable!
acceptor_introns.append((intronAQ,intronAS))
if not len(donor_introns) or not len(acceptor_introns):
# no aligned 5' && aligned 3' introns
continue
# initialize extended tinyexon PacbPORF
from pacb import PacbP
pacbp = PacbP(input=(
tinyexonQ.proteinsequence(),
tinyexonS.proteinsequence(),
tinyexonQ.protein_start(),
tinyexonS.protein_start(),
) )
pacbp.strip_unmatched_ends()
# continue if no fraction could be aligned
if len(pacbp) == 0: continue
tinypacbporf = pacbp2pacbporf(pacbp,tinyexonQ.orf,tinyexonS.orf)
tinypacbporf.extend_pacbporf_after_stops()
####################################################################
if verbose:
print tinypacbporf
tinypacbporf.print_protein_and_dna()
print len(donor_introns), len(acceptor_introns),
print max([ dQ.donor.pssm_score+dS.donor.pssm_score for dQ,dS in donor_introns]),
print max([ aQ.acceptor.pssm_score+aS.acceptor.pssm_score for aQ,aS in acceptor_introns])
####################################################################
# if here, we have accepted tinyexon bridges!
# gather them and store to return_list
for intronDQkey, intronDQ in resultdictQ[keyQ][0].iteritems():
if intronDQ.donor.pos not in [ dQ.pos for dQ,dS in algdonors ]:
continue
for intronDSkey, intronDS in resultdictS[keyS][0].iteritems():
if intronDS.donor.pos not in [ dS.pos for dQ,dS in algdonors ]:
continue
for intronAQkey, intronAQ in resultdictQ[keyQ][1].iteritems():
if intronAQ.acceptor.pos not in [ aQ.pos for aQ,aS in algacceps ]:
continue
for intronASkey, intronAS in resultdictS[keyS][1].iteritems():
if intronAS.acceptor.pos not in [ aS.pos for aQ,aS in algacceps ]:
continue
####################################################
# set some meta-data properties to the intron objects
####################################################
_score_introns_obtained_by_mapping(
intronDQ,intronDS,pacbporfD,
tinypacbporf,source='ABGPmappingTE')
_score_introns_obtained_by_mapping(
intronAQ,intronAS,tinypacbporf,
pacbporfA,source='ABGPmappingTE')
# create _linked_to_xxx attributes
intronDQ._linked_to_pacbporfs = [ tinypacbporf ]
intronAQ._linked_to_pacbporfs = [ tinypacbporf ]
intronDS._linked_to_pacbporfs = [ tinypacbporf ]
intronAS._linked_to_pacbporfs = [ tinypacbporf ]
intronDQ._linked_to_introns = [ intronAQ ]
intronAQ._linked_to_introns = [ intronDQ ]
intronDS._linked_to_introns = [ intronAS ]
intronAS._linked_to_introns = [ intronDS ]
# append to tmp result list
return_list.append(
(intronDQ,intronDS,tinypacbporf,intronAQ,intronAS)
)
# check if there are >1 candidate tiny exons
# currently, we choose only to return the **best** mapped tinyexon
if len(return_list) == 0:
pass
elif len(return_list) == 1:
pass
else:
# only take the highest scoring candidate here
min_distance = min([ (a._distance+d._distance) for a,b,c,d,e in return_list ])
pos2score = []
for (intronDQ,intronDS,tinypacbporf,intronAQ,intronAS) in return_list:
if (intronDQ._distance + intronAQ._distance) > min_distance:
pos2score.append( 0.0 )
else:
# calculate overall pssm score
total_pssm = 0.0
total_pssm += intronDQ.donor.pssm_score
total_pssm += intronDQ.acceptor.pssm_score
total_pssm += intronDS.donor.pssm_score
total_pssm += intronDS.acceptor.pssm_score
total_pssm += intronAQ.donor.pssm_score
total_pssm += intronAQ.acceptor.pssm_score
total_pssm += intronAS.donor.pssm_score
total_pssm += intronAS.acceptor.pssm_score
pos2score.append( total_pssm )
# get highest score and linked tinyexon
max_score = max(pos2score)
return_list = [ return_list[pos2score.index(max_score)] ]
############################################################################
# some printing in verbose mode
if verbose and return_list:
(intronDQ,intronDS,tinypacbporf,intronAQ,intronAS) = return_list[0]
print "BEST MAPPED TINYEXON:"
print tinypacbporf
print tinypacbporf.query, intronDQ._distance, intronAQ._distance,
print ( intronDQ.donor.pos, intronDQ.acceptor.pos ),
print ( intronDS.donor.pos, intronDS.acceptor.pos ),
print ( intronAQ.donor.pos, intronAQ.acceptor.pos ),
print ( intronAS.donor.pos, intronAS.acceptor.pos )
############################################################################
# return the result list
return return_list
def merge_pacbporfs_with_introns(pacbporfD,pacbporfA,verbose=False,**kwargs):
"""
Merge 2 PacbPORF objects by introns
@attention: see orfs.merge_orfs_with_intron for **kwargs
@attention: see functions._filter_for_alignable_splice_sites for **kwargs
@attention: see functions._filter_for_entropy for **kwargs
@type pacbporfD: PacbPORF object
@param pacbporfD: PacbPORF object that has to deliver PSSM donor objects
@type pacbporfA: PacbPORF object
@param pacbporfA: PacbPORF object that has to deliver PSSM acceptor objects
@type verbose: Boolean
@param verbose: print status/debugging messages to STDOUT
@rtype: list
@return: list with ( intron, intron ), in query and sbjct
"""
# input validation
IsPacbPORF(pacbporfD)
IsPacbPORF(pacbporfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs,KWARGS_MAPPED_INTRON)
if not kwargs.has_key('aligned_site_max_triplet_distance'):
kwargs['aligned_site_max_triplet_distance'] = kwargs['max_aa_offset']
# settings for minimal alignment entropy score
min_donor_site_alignment_entropy = 0.0
min_acceptor_site_alignment_entropy = 0.0
# calculate maximal/minimal donor/acceptor site position based on alignment
ELEGIABLE_SPLICE_SITE_AA_RANGE = 75
qdr = pacbporfD.alignment_dna_range_query()
qar = pacbporfA.alignment_dna_range_query()
min_donor_query_pos = max([ min(qdr), max(qdr)-(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
max_accep_query_pos = min([ max(qar), min(qar)+(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
sdr = pacbporfD.alignment_dna_range_sbjct()
sar = pacbporfA.alignment_dna_range_sbjct()
min_donor_sbjct_pos = max([ min(sdr), max(sdr)-(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
max_accep_sbjct_pos = min([ max(sar), min(sar)+(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
# get list of introns
#intronsQ = merge_orfs_with_intron(pacbporfD.orfQ,pacbporfA.orfQ,
# min_donor_pos =min_donor_query_pos,
# max_acceptor_pos=max_accep_query_pos,**kwargs)
#intronsS = merge_orfs_with_intron(pacbporfD.orfS,pacbporfA.orfS,
# min_donor_pos =min_donor_sbjct_pos,
# max_acceptor_pos=max_accep_sbjct_pos,**kwargs)
# get list of introns
intronsQ = merge_orfs_with_intron(pacbporfD.orfQ,pacbporfA.orfQ,**kwargs)
intronsS = merge_orfs_with_intron(pacbporfD.orfS,pacbporfA.orfS,**kwargs)
# get unique list of donors & acceptors
donorQ = olba( list(Set([inQ.donor for inQ in intronsQ ])), order_by='pos')
donorS = olba( list(Set([inS.donor for inS in intronsS ])), order_by='pos')
accepQ = olba( list(Set([inQ.acceptor for inQ in intronsQ ])), order_by='pos')
accepS = olba( list(Set([inS.acceptor for inS in intronsS ])), order_by='pos')
############################################################################
if verbose:
print "dQ1", [ d.pos for d in donorQ ], "aQ1", [ a.pos for a in accepQ ]
print "dS1", [ d.pos for d in donorS ], "aS1", [ a.pos for a in accepS ]
############################################################################
# filter for alignable donor & acceptor sites
kwargs['allow_non_canonical'] = kwargs['allow_non_canonical_donor']
algdonors = _filter_for_alignable_splice_sites(donorQ,donorS,pacbporfD,**kwargs)
kwargs['allow_non_canonical'] = kwargs['allow_non_canonical_acceptor']
algacceps = _filter_for_alignable_splice_sites(accepQ,accepS,pacbporfA,**kwargs)
############################################################################
if verbose:
print "dQ2", [ _dq.pos for (_dq,_ds) in algdonors ],
print "aQ2", [ _aq.pos for (_aq,_as) in algacceps ]
print "dS2", [ _ds.pos for (_dq,_ds) in algdonors ],
print "aS2", [ _as.pos for (_aq,_as) in algacceps ]
############################################################################
# remove sites with to low alignment entropy
algdonors = _filter_for_entropy(algdonors,pacbporfD,'donor',
min_alignment_entropy=min_donor_site_alignment_entropy)
algacceps = _filter_for_entropy(algacceps,pacbporfA,'acceptor',
min_alignment_entropy=min_acceptor_site_alignment_entropy)
############################################################################
if verbose:
print "dQ3", [ _dq.pos for (_dq,_ds) in algdonors ],
print "aQ3", [ _aq.pos for (_aq,_as) in algacceps ]
print "dS3", [ _ds.pos for (_dq,_ds) in algdonors ],
print "aS3", [ _as.pos for (_aq,_as) in algacceps ]
############################################################################
# make unique position lists for quick lookup in intron lists
dQpl = Set([ dQ.pos for dQ,dS in algdonors ])
dSpl = Set([ dS.pos for dQ,dS in algdonors ])
aQpl = Set([ aQ.pos for aQ,aS in algacceps ])
aSpl = Set([ aS.pos for aQ,aS in algacceps ])
# check exterior boundaries of PacbPORFs
sposD = pacbporfD._get_original_alignment_pos_start()
eposD = pacbporfD._get_original_alignment_pos_end()
sposA = pacbporfA._get_original_alignment_pos_start()
eposA = pacbporfA._get_original_alignment_pos_end()
# now make list of aligable introns
algintrons = []
for intQ in intronsQ:
# check if intron falls within the PacbPORF aligned area
if intQ.donor.pos <= sposD.query_dna_start: continue
if intQ.acceptor.pos >= eposA.query_dna_end: continue
if intQ.donor.pos in dQpl and intQ.acceptor.pos in aQpl:
# Query intron occurs in list of alignable splice sites!
for intS in intronsS:
# check if intron falls within the PacbPORF aligned area
if intS.donor.pos <= sposD.sbjct_dna_start: continue
if intS.acceptor.pos >= eposA.sbjct_dna_end: continue
if intS.donor.pos in dSpl and intS.acceptor.pos in aSpl:
# Sbjct intron occurs as well in alignable splice sites!
if (intQ.donor,intS.donor) in algdonors and\
(intQ.acceptor,intS.acceptor) in algacceps:
# Sbjct & Query Donor & Acceptor are alignable!
algintrons.append( ( intQ, intS ) )
############################################################################
# set some meta-data properties to the intron objects
############################################################################
for intQ,intS in algintrons:
distDnt = pacbporfD.get_distance_aligned_nucleotide_positions(
query = intQ.donor.pos, sbjct = intS.donor.pos
)
distAnt = pacbporfA.get_distance_aligned_nucleotide_positions(
query = intQ.acceptor.pos, sbjct = intS.acceptor.pos
)
# final distance check. kwargs['aligned_site_max_triplet_distance']
# is applied on donor and acceptor site. This distance measured on the
# protein sequence can be DOUBLED in case distDnt / distAnt are
# opposite (+ and -). Check here if the protein sequence gap is
# as well <= kwargs['aligned_site_max_triplet_distance'].
if abs(distAnt - distDnt) > kwargs['aligned_site_max_triplet_distance']*3:
continue
# add distance score to introns
intQ._distance = abs(distDnt) + abs(distAnt)
intS._distance = abs(distDnt) + abs(distAnt)
# add Alignment Positional Periphery Score into objects
succes = set_apps_intron_query(intQ,pacbporfD,pacbporfA)
succes = set_apps_intron_sbjct(intS,pacbporfD,pacbporfA)
# set GFF fsource attribute for recognition of intron sources
intQ._gff['fsource'] = "ABGPmapping"
intS._gff['fsource'] = "ABGPmapping"
########################################################################
if verbose:
# some printing....
print "Aligned introns:", ( intQ.donor.pos, intQ.acceptor.pos ) ,
print ( intS.donor.pos, intS.acceptor.pos ),
print "DIST:", distDnt, distAnt,
print "[%s]" % kwargs['aligned_site_max_triplet_distance'],
print "ENTROPY: %1.2f %1.2f" % (intQ._apps_donor, intQ._apps_accep),
print "PSSM: (%1.2f %1.2f) (%1.2f %1.2f)" % (
intQ.donor.pssm_score, intS.donor.pssm_score,
intQ.acceptor.pssm_score, intS.acceptor.pssm_score,
)
########################################################################
# return lists of aligned introns
return algintrons
def merge_pacbporfs_by_tinyexons(pacbporfD,
pacbporfA,
orfSetObjQ,
orfSetObjS,
verbose=False,
**kwargs):
""" """
# input validation
IsPacbPORF(pacbporfD)
IsPacbPORF(pacbporfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs, KWARGS_MAPPED_INTRON)
if not kwargs.has_key('aligned_site_max_triplet_distance'):
kwargs['aligned_site_max_triplet_distance'] = kwargs['max_aa_offset']
# settings for minimal alignment entropy score
min_donor_site_alignment_entropy = 0.0
min_acceptor_site_alignment_entropy = 0.0
resultlistQ = merge_orfs_with_tinyexon(
pacbporfD.orfQ,
pacbporfA.orfQ,
preceding_donor_sites=pacbporfD.orfQ._donor_sites,
subsequent_acceptor_sites=pacbporfA.orfQ._acceptor_sites,
orflist=orfSetObjQ.orfs,
**kwargs)
resultlistS = merge_orfs_with_tinyexon(
pacbporfD.orfS,
pacbporfA.orfS,
preceding_donor_sites=pacbporfD.orfS._donor_sites,
subsequent_acceptor_sites=pacbporfA.orfS._acceptor_sites,
orflist=orfSetObjS.orfs,
**kwargs)
# translate resultlists to dict: key == exon, value = [ {intronsD},{intronsS} ]
resultdictQ, key2exonQ = _tinyexon_list_2_dict(resultlistQ)
resultdictS, key2exonS = _tinyexon_list_2_dict(resultlistS)
# get unique list of donors & acceptors
donorQ = olba(list(Set([inD.donor for inD, te, inA in resultlistQ])),
order_by='pos')
donorS = olba(list(Set([inD.donor for inD, te, inA in resultlistS])),
order_by='pos')
accepQ = olba(list(Set([inA.acceptor for inD, te, inA in resultlistQ])),
order_by='pos')
accepS = olba(list(Set([inA.acceptor for inD, te, inA in resultlistS])),
order_by='pos')
## filter for alignable donor & acceptor sites
kwargs['allow_non_canonical'] = True # True
kwargs['aligned_site_max_triplet_distance'] = 0 # 2
algdonors = _filter_for_alignable_splice_sites(donorQ, donorS, pacbporfD,
**kwargs)
algacceps = _filter_for_alignable_splice_sites(accepQ, accepS, pacbporfA,
**kwargs)
# settings for minimal alignment entropy score
# TODO TODO -> THIS MUST BE FIXED TO A NICE THRESHOLD VALUE!!!
min_donor_site_alignment_entropy = 0.1
min_acceptor_site_alignment_entropy = 0.1
# remove sites with to low alignment entropy
algdonors = _filter_for_entropy(
algdonors,
pacbporfD,
'donor',
min_alignment_entropy=min_donor_site_alignment_entropy)
algacceps = _filter_for_entropy(
algacceps,
pacbporfA,
'acceptor',
min_alignment_entropy=min_acceptor_site_alignment_entropy)
# return list: intronQD,intronSD,tinyexon,intronAQ,intronAS
return_list = []
############################################################################
if verbose:
print "bridges constructed: ORFS:",
print(pacbporfD.orfQ.id, pacbporfA.orfQ.id),
print(pacbporfD.orfS.id, pacbporfA.orfS.id),
print len(resultdictQ), len(resultdictS),
print(len(resultlistQ), len(donorQ), len(accepQ)),
print(len(resultlistS), len(donorS), len(accepS)),
print(len(algdonors), len(algacceps))
############################################################################
for keyQ, tinyexonQ in key2exonQ.iteritems():
for keyS, tinyexonS in key2exonS.iteritems():
if tinyexonQ.donor.phase != tinyexonS.donor.phase:
continue
if tinyexonQ.acceptor.phase != tinyexonS.acceptor.phase:
continue
if tinyexonQ.length != tinyexonS.length:
continue
# if here, then tinyexons of identical structure
####################################################################
if verbose:
print tinyexonQ.length, tinyexonQ.donor.phase,
print(len(resultdictQ[keyQ][0]), len(resultdictQ[keyQ][1])),
print(len(resultdictS[keyS][0]), len(resultdictS[keyS][1])),
print tinyexonQ,
print tinyexonQ.proteinsequence(), tinyexonS.proteinsequence(),
print tinyexonS.acceptor.pssm_score + tinyexonS.donor.pssm_score
####################################################################
donor_introns = []
acceptor_introns = []
for intronDQkey, intronDQ in resultdictQ[keyQ][0].iteritems():
if intronDQ.donor.pos not in [dQ.pos for dQ, dS in algdonors]:
continue
for intronDSkey, intronDS in resultdictS[keyS][0].iteritems():
if intronDS.donor.pos not in [
dS.pos for dQ, dS in algdonors
]:
continue
# check if they exists as aligned sites
alignedkey = (intronDQ.donor.pos, intronDS.donor.pos)
if alignedkey not in [(dQ.pos, dS.pos)
for dQ, dS in algdonors]:
continue
# if here, we have a set of introns 5' of the tinyexon
# which are perfectly alignable!
donor_introns.append((intronDQ, intronDS))
for intronAQkey, intronAQ in resultdictQ[keyQ][1].iteritems():
if intronAQ.acceptor.pos not in [
aQ.pos for aQ, aS in algacceps
]:
continue
for intronASkey, intronAS in resultdictS[keyS][1].iteritems():
if intronAS.acceptor.pos not in [
aS.pos for aQ, aS in algacceps
]:
continue
# check if they exists as aligned sites
alignedkey = (intronAQ.acceptor.pos, intronAS.acceptor.pos)
if alignedkey not in [(aQ.pos, aS.pos)
for aQ, aS in algacceps]:
continue
# if here, we have a set of introns 3' of the tinyexon
# which are perfectly alignable!
acceptor_introns.append((intronAQ, intronAS))
if not len(donor_introns) or not len(acceptor_introns):
# no aligned 5' && aligned 3' introns
continue
# initialize extended tinyexon PacbPORF
from pacb import PacbP
pacbp = PacbP(input=(
tinyexonQ.proteinsequence(),
tinyexonS.proteinsequence(),
tinyexonQ.protein_start(),
tinyexonS.protein_start(),
))
pacbp.strip_unmatched_ends()
# continue if no fraction could be aligned
if len(pacbp) == 0: continue
tinypacbporf = pacbp2pacbporf(pacbp, tinyexonQ.orf, tinyexonS.orf)
tinypacbporf.extend_pacbporf_after_stops()
####################################################################
if verbose:
print tinypacbporf
tinypacbporf.print_protein_and_dna()
print len(donor_introns), len(acceptor_introns),
print max([
dQ.donor.pssm_score + dS.donor.pssm_score
for dQ, dS in donor_introns
]),
print max([
aQ.acceptor.pssm_score + aS.acceptor.pssm_score
for aQ, aS in acceptor_introns
])
####################################################################
# if here, we have accepted tinyexon bridges!
# gather them and store to return_list
for intronDQkey, intronDQ in resultdictQ[keyQ][0].iteritems():
if intronDQ.donor.pos not in [dQ.pos for dQ, dS in algdonors]:
continue
for intronDSkey, intronDS in resultdictS[keyS][0].iteritems():
if intronDS.donor.pos not in [
dS.pos for dQ, dS in algdonors
]:
continue
for intronAQkey, intronAQ in resultdictQ[keyQ][
1].iteritems():
if intronAQ.acceptor.pos not in [
aQ.pos for aQ, aS in algacceps
]:
continue
for intronASkey, intronAS in resultdictS[keyS][
1].iteritems():
if intronAS.acceptor.pos not in [
aS.pos for aQ, aS in algacceps
]:
continue
####################################################
# set some meta-data properties to the intron objects
####################################################
_score_introns_obtained_by_mapping(
intronDQ,
intronDS,
pacbporfD,
tinypacbporf,
source='ABGPmappingTE')
_score_introns_obtained_by_mapping(
intronAQ,
intronAS,
tinypacbporf,
pacbporfA,
source='ABGPmappingTE')
# create _linked_to_xxx attributes
intronDQ._linked_to_pacbporfs = [tinypacbporf]
intronAQ._linked_to_pacbporfs = [tinypacbporf]
intronDS._linked_to_pacbporfs = [tinypacbporf]
intronAS._linked_to_pacbporfs = [tinypacbporf]
intronDQ._linked_to_introns = [intronAQ]
intronAQ._linked_to_introns = [intronDQ]
intronDS._linked_to_introns = [intronAS]
intronAS._linked_to_introns = [intronDS]
# append to tmp result list
return_list.append(
(intronDQ, intronDS, tinypacbporf, intronAQ,
intronAS))
# check if there are >1 candidate tiny exons
# currently, we choose only to return the **best** mapped tinyexon
if len(return_list) == 0:
pass
elif len(return_list) == 1:
pass
else:
# only take the highest scoring candidate here
min_distance = min([(a._distance + d._distance)
for a, b, c, d, e in return_list])
pos2score = []
for (intronDQ, intronDS, tinypacbporf, intronAQ,
intronAS) in return_list:
if (intronDQ._distance + intronAQ._distance) > min_distance:
pos2score.append(0.0)
else:
# calculate overall pssm score
total_pssm = 0.0
total_pssm += intronDQ.donor.pssm_score
total_pssm += intronDQ.acceptor.pssm_score
total_pssm += intronDS.donor.pssm_score
total_pssm += intronDS.acceptor.pssm_score
total_pssm += intronAQ.donor.pssm_score
total_pssm += intronAQ.acceptor.pssm_score
total_pssm += intronAS.donor.pssm_score
total_pssm += intronAS.acceptor.pssm_score
pos2score.append(total_pssm)
# get highest score and linked tinyexon
max_score = max(pos2score)
return_list = [return_list[pos2score.index(max_score)]]
############################################################################
# some printing in verbose mode
if verbose and return_list:
(intronDQ, intronDS, tinypacbporf, intronAQ, intronAS) = return_list[0]
print "BEST MAPPED TINYEXON:"
print tinypacbporf
print tinypacbporf.query, intronDQ._distance, intronAQ._distance,
print(intronDQ.donor.pos, intronDQ.acceptor.pos),
print(intronDS.donor.pos, intronDS.acceptor.pos),
print(intronAQ.donor.pos, intronAQ.acceptor.pos),
print(intronAS.donor.pos, intronAS.acceptor.pos)
############################################################################
# return the result list
return return_list
def merge_pacbporfs_with_introns(pacbporfD,
pacbporfA,
verbose=False,
**kwargs):
"""
Merge 2 PacbPORF objects by introns
@attention: see orfs.merge_orfs_with_intron for **kwargs
@attention: see functions._filter_for_alignable_splice_sites for **kwargs
@attention: see functions._filter_for_entropy for **kwargs
@type pacbporfD: PacbPORF object
@param pacbporfD: PacbPORF object that has to deliver PSSM donor objects
@type pacbporfA: PacbPORF object
@param pacbporfA: PacbPORF object that has to deliver PSSM acceptor objects
@type verbose: Boolean
@param verbose: print status/debugging messages to STDOUT
@rtype: list
@return: list with ( intron, intron ), in query and sbjct
"""
# input validation
IsPacbPORF(pacbporfD)
IsPacbPORF(pacbporfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs, KWARGS_MAPPED_INTRON)
if not kwargs.has_key('aligned_site_max_triplet_distance'):
kwargs['aligned_site_max_triplet_distance'] = kwargs['max_aa_offset']
# settings for minimal alignment entropy score
min_donor_site_alignment_entropy = 0.0
min_acceptor_site_alignment_entropy = 0.0
# calculate maximal/minimal donor/acceptor site position based on alignment
ELEGIABLE_SPLICE_SITE_AA_RANGE = 75
qdr = pacbporfD.alignment_dna_range_query()
qar = pacbporfA.alignment_dna_range_query()
min_donor_query_pos = max(
[min(qdr), max(qdr) - (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
max_accep_query_pos = min(
[max(qar), min(qar) + (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
sdr = pacbporfD.alignment_dna_range_sbjct()
sar = pacbporfA.alignment_dna_range_sbjct()
min_donor_sbjct_pos = max(
[min(sdr), max(sdr) - (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
max_accep_sbjct_pos = min(
[max(sar), min(sar) + (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
# get list of introns
#intronsQ = merge_orfs_with_intron(pacbporfD.orfQ,pacbporfA.orfQ,
# min_donor_pos =min_donor_query_pos,
# max_acceptor_pos=max_accep_query_pos,**kwargs)
#intronsS = merge_orfs_with_intron(pacbporfD.orfS,pacbporfA.orfS,
# min_donor_pos =min_donor_sbjct_pos,
# max_acceptor_pos=max_accep_sbjct_pos,**kwargs)
# get list of introns
intronsQ = merge_orfs_with_intron(pacbporfD.orfQ, pacbporfA.orfQ, **kwargs)
intronsS = merge_orfs_with_intron(pacbporfD.orfS, pacbporfA.orfS, **kwargs)
# get unique list of donors & acceptors
donorQ = olba(list(Set([inQ.donor for inQ in intronsQ])), order_by='pos')
donorS = olba(list(Set([inS.donor for inS in intronsS])), order_by='pos')
accepQ = olba(list(Set([inQ.acceptor for inQ in intronsQ])),
order_by='pos')
accepS = olba(list(Set([inS.acceptor for inS in intronsS])),
order_by='pos')
############################################################################
if verbose:
print "dQ1", [d.pos for d in donorQ], "aQ1", [a.pos for a in accepQ]
print "dS1", [d.pos for d in donorS], "aS1", [a.pos for a in accepS]
############################################################################
# filter for alignable donor & acceptor sites
kwargs['allow_non_canonical'] = kwargs['allow_non_canonical_donor']
algdonors = _filter_for_alignable_splice_sites(donorQ, donorS, pacbporfD,
**kwargs)
kwargs['allow_non_canonical'] = kwargs['allow_non_canonical_acceptor']
algacceps = _filter_for_alignable_splice_sites(accepQ, accepS, pacbporfA,
**kwargs)
############################################################################
if verbose:
print "dQ2", [_dq.pos for (_dq, _ds) in algdonors],
print "aQ2", [_aq.pos for (_aq, _as) in algacceps]
print "dS2", [_ds.pos for (_dq, _ds) in algdonors],
print "aS2", [_as.pos for (_aq, _as) in algacceps]
############################################################################
# remove sites with to low alignment entropy
algdonors = _filter_for_entropy(
algdonors,
pacbporfD,
'donor',
min_alignment_entropy=min_donor_site_alignment_entropy)
algacceps = _filter_for_entropy(
algacceps,
pacbporfA,
'acceptor',
min_alignment_entropy=min_acceptor_site_alignment_entropy)
############################################################################
if verbose:
print "dQ3", [_dq.pos for (_dq, _ds) in algdonors],
print "aQ3", [_aq.pos for (_aq, _as) in algacceps]
print "dS3", [_ds.pos for (_dq, _ds) in algdonors],
print "aS3", [_as.pos for (_aq, _as) in algacceps]
############################################################################
# make unique position lists for quick lookup in intron lists
dQpl = Set([dQ.pos for dQ, dS in algdonors])
dSpl = Set([dS.pos for dQ, dS in algdonors])
aQpl = Set([aQ.pos for aQ, aS in algacceps])
aSpl = Set([aS.pos for aQ, aS in algacceps])
# check exterior boundaries of PacbPORFs
sposD = pacbporfD._get_original_alignment_pos_start()
eposD = pacbporfD._get_original_alignment_pos_end()
sposA = pacbporfA._get_original_alignment_pos_start()
eposA = pacbporfA._get_original_alignment_pos_end()
# now make list of aligable introns
algintrons = []
for intQ in intronsQ:
# check if intron falls within the PacbPORF aligned area
if intQ.donor.pos <= sposD.query_dna_start: continue
if intQ.acceptor.pos >= eposA.query_dna_end: continue
if intQ.donor.pos in dQpl and intQ.acceptor.pos in aQpl:
# Query intron occurs in list of alignable splice sites!
for intS in intronsS:
# check if intron falls within the PacbPORF aligned area
if intS.donor.pos <= sposD.sbjct_dna_start: continue
if intS.acceptor.pos >= eposA.sbjct_dna_end: continue
if intS.donor.pos in dSpl and intS.acceptor.pos in aSpl:
# Sbjct intron occurs as well in alignable splice sites!
if (intQ.donor,intS.donor) in algdonors and\
(intQ.acceptor,intS.acceptor) in algacceps:
# Sbjct & Query Donor & Acceptor are alignable!
algintrons.append((intQ, intS))
############################################################################
# set some meta-data properties to the intron objects
############################################################################
for intQ, intS in algintrons:
distDnt = pacbporfD.get_distance_aligned_nucleotide_positions(
query=intQ.donor.pos, sbjct=intS.donor.pos)
distAnt = pacbporfA.get_distance_aligned_nucleotide_positions(
query=intQ.acceptor.pos, sbjct=intS.acceptor.pos)
# final distance check. kwargs['aligned_site_max_triplet_distance']
# is applied on donor and acceptor site. This distance measured on the
# protein sequence can be DOUBLED in case distDnt / distAnt are
# opposite (+ and -). Check here if the protein sequence gap is
# as well <= kwargs['aligned_site_max_triplet_distance'].
if abs(distAnt -
distDnt) > kwargs['aligned_site_max_triplet_distance'] * 3:
continue
# add distance score to introns
intQ._distance = abs(distDnt) + abs(distAnt)
intS._distance = abs(distDnt) + abs(distAnt)
# add Alignment Positional Periphery Score into objects
succes = set_apps_intron_query(intQ, pacbporfD, pacbporfA)
succes = set_apps_intron_sbjct(intS, pacbporfD, pacbporfA)
# set GFF fsource attribute for recognition of intron sources
intQ._gff['fsource'] = "ABGPmapping"
intS._gff['fsource'] = "ABGPmapping"
########################################################################
if verbose:
# some printing....
print "Aligned introns:", (intQ.donor.pos, intQ.acceptor.pos),
print(intS.donor.pos, intS.acceptor.pos),
print "DIST:", distDnt, distAnt,
print "[%s]" % kwargs['aligned_site_max_triplet_distance'],
print "ENTROPY: %1.2f %1.2f" % (intQ._apps_donor,
intQ._apps_accep),
print "PSSM: (%1.2f %1.2f) (%1.2f %1.2f)" % (
intQ.donor.pssm_score,
intS.donor.pssm_score,
intQ.acceptor.pssm_score,
intS.acceptor.pssm_score,
)
########################################################################
# return lists of aligned introns
return algintrons