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_intron(pfD,pfA,queryorsbjct,verbose=False,**kwargs):
"""
Project splicesites from SBJCT intron on continious QUERY PacbPORFs
@type pfD: PacbPORF object
@param pfD: PacbPORF object that has to deliver (aligned) donor sites
@type pfA: PacbPORF object
@param pfA: PacbPORF object that has to deliver (aligned) acceptor sites
@type queryorsbjct: string
@param queryorsbjct: literal string 'query' or 'sbjct'
@type verbose: Boolean
@param verbose: print debugging info to STDOUT when True
@attention: see pacb.connecting.merge_orfs_with_intron for **kwargs)
@rtype: list
@return: list with ProjectedIntrons (from Sbjct on Query)
"""
# input validation
IsPacbPORF(pfD)
IsPacbPORF(pfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs,KWARGS_PROJECTED_INTRON)
### if not kwargs.has_key('projected_intron_max_nt_offset'):
### kwargs['projected_intron_max_nt_offset'] = PROJECTED_INTRON_MAX_NT_OFFSET
### if not kwargs.has_key('projected_intron_max_aa_offset'):
### kwargs['projected_intron_max_aa_offset'] = PROJECTED_INTRON_MAX_AA_OFFSET
# settings for minimal alignment entropy score
min_donor_site_alignment_entropy = 0.0
min_acceptor_site_alignment_entropy = 0.0
ELEGIABLE_SPLICE_SITE_AA_RANGE = 75
sposD = pfD._get_original_alignment_pos_start()
eposD = pfD._get_original_alignment_pos_end()
sposA = pfA._get_original_alignment_pos_start()
eposA = pfA._get_original_alignment_pos_end()
if queryorsbjct == "query":
# Orfs of SBJCT must be identical
IsIdenticalOrfs(pfD.orfS,pfA.orfS)
donorOrf = pfD.orfQ
accepOrf = pfA.orfQ
prjctOrf = pfD.orfS # pfD.orfS == pfA.orfS
dStart = sposD.query_dna_start # ALIGNED start of donorPacbPORF
dEnd = pfD.query_dna_end # ABSOLUTE end of donorPacbPORF
aStart = pfA.query_dna_start # ABSOLUTE start of acceptorPacbPORF
aEnd = eposA.query_dna_end # ALIGNED end of acceptorPacbPORF
outOfAlignmentAttribute = "sbjct_dna_start"
# calculate elegiable splice site range
qdr = pfD.alignment_dna_range_query()
qar = pfA.alignment_dna_range_query()
min_donor_pos = max([ min(qdr), max(qdr)-(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
max_accep_pos = min([ max(qar), min(qar)+(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
elif queryorsbjct == "sbjct":
# Orfs of QUERY must be identical
IsIdenticalOrfs(pfD.orfQ,pfA.orfQ)
donorOrf = pfD.orfS
accepOrf = pfA.orfS
prjctOrf = pfD.orfQ # pfD.orfQ == pfA.orfQ
dStart = sposD.sbjct_dna_start # ALIGNED start of donorPacbPORF
dEnd = pfD.sbjct_dna_end # ABSOLUTE end of donorPacbPORF
aStart = pfA.sbjct_dna_start # ABSOLUTE start of acceptorPacbPORF
aEnd = eposA.sbjct_dna_end # ALIGNED end of acceptorPacbPORF
outOfAlignmentAttribute = "query_dna_start"
# calculate elegiable splice site range
sdr = pfD.alignment_dna_range_sbjct()
sar = pfA.alignment_dna_range_sbjct()
min_donor_pos = max([ min(sdr), max(sdr)-(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
max_accep_pos = min([ max(sar), min(sar)+(ELEGIABLE_SPLICE_SITE_AA_RANGE*3) ])
else:
message = "'queryorsbjct' (%s), not 'query' or 'sbjct'" % queryorsbjct
raise InproperlyAppliedArgument, message
# predict introns only in `queryorsbjct` Orfs
# introns is a list of IntronConnectingOrfs objects
introns = merge_orfs_with_intron(donorOrf,accepOrf,
min_donor_pos=min_donor_pos,
max_acceptor_pos=max_accep_pos,
order_by='length',**kwargs)
# return list with projected introns
projected_introns = []
# gather unique donor and acceptor positions from list
# of IntronConnectingOrfs
for intron in introns:
# break if intron is to large
if kwargs['max_intron_nt_length'] and intron.length > kwargs['max_intron_nt_length']: break
# continue if intron is to small
if kwargs['min_intron_nt_length'] and intron.length < kwargs['min_intron_nt_length']: continue
# continue if intron has non-canonical features
# check if intron.start is on pfD;
# inframe-introns can be projected outside of pfD/pfA area
if intron.start <= dStart: continue
if intron.start >= dEnd: continue
# check if intron.end is on pfA;
# inframe-introns can be projected outside of pfD/pfA area
if intron.end <= aStart: continue
if intron.end >= aEnd: continue
if queryorsbjct == "sbjct":
# get positions of donor & acceptor in the PacbPORF alignment
donorPositionPos, phaseD = pfD.dnaposition_sbjct(intron.donor.pos,forced_return=True)
accepPositionPos, phaseA = pfA.dnaposition_sbjct(intron.acceptor.pos,forced_return=True)
# calculate projected distance on QUERY
posDposQuery = pfD._positions[donorPositionPos].query_pos
posAposQuery = pfA._positions[accepPositionPos].query_pos
aaDistance = posAposQuery - posDposQuery
else:
# get positions of donor & acceptor in the PacbPORF alignment
donorPositionPos, phaseD = pfD.dnaposition_query(intron.donor.pos,forced_return=True)
accepPositionPos, phaseA = pfA.dnaposition_query(intron.acceptor.pos,forced_return=True)
# calculate binary entropy from projected position on SBJCT
posDposSbjct = pfD._positions[donorPositionPos].sbjct_pos
posAposSbjct = pfA._positions[accepPositionPos].sbjct_pos
aaDistance = posAposSbjct - posDposSbjct
# calculate binary entropy score
entropyDonorSbjct = pfD.alignment_entropy(donorPositionPos,method='donor')
entropyAcceptorSbjct= pfA.alignment_entropy(accepPositionPos,method='acceptor')
# do distance check upon (projected) intron acceptance
if abs(aaDistance) <= kwargs['max_aa_offset']:
# check if we've runned out of the aligned part
outofalignedpacbporf = False
# get the projected donor position; mind the gap on this spot ;-)
while pfD._positions[donorPositionPos].isa_gap and donorPositionPos > 0 :
donorPositionPos -= 1
else:
projected_donor_position = getattr(pfD._positions[donorPositionPos],outOfAlignmentAttribute) + phaseD
if donorPositionPos == 0 and pfD._positions[donorPositionPos].isa_gap:
print "WarningThatIsTackled::outofalignedpacbporf::donor"
outofalignedpacbporf = True
# get the projected acceptor position; mind the gap on this spot ;-)
while pfA._positions[accepPositionPos].isa_gap and len(pfA._positions) > accepPositionPos+1:
accepPositionPos += 1
else:
projected_accep_position = getattr(pfA._positions[accepPositionPos],outOfAlignmentAttribute) + phaseA
if accepPositionPos == len(pfA._positions)-1 and pfA._positions[accepPositionPos].isa_gap:
print "WarningThatIsTackled::outofalignedpacbporf::acceptor"
outofalignedpacbporf = True
if not outofalignedpacbporf:
################################################################
# set some meta-data properties to the intron object
################################################################
# add distance score to intron
intron._distance = abs(aaDistance)*3
# add Alignment Positional Periphery Score into objects
if queryorsbjct == "query":
succes = set_apps_intron_query(intron,pfD,pfA)
else:
succes = set_apps_intron_sbjct(intron,pfD,pfA)
# set GFF fsource attribute for recognition of intron sources
intron._gff['fsource'] = "ABGPprojecting"
# make a ProjectedIntronConnectingOrfs object
pico = ProjectedIntronConnectingOrfs(prjctOrf,
projected_donor_position,
projected_accep_position)
intron.binary_entropy_donor = entropyDonorSbjct
intron.binary_entropy_acceptor = entropyAcceptorSbjct
pico.add_projected_intron( intron )
pico.phase = intron.phase
projected_introns.append( pico )
################################################################
if verbose:
print "PROJ::", intron._distance,
print (pfD.orfQ.id, pfA.orfQ.id),
print (pfD.orfS.id, pfA.orfS.id),
print "%s-%snt" % (intron.donor.pos, intron.acceptor.pos),
print "%2.1f,%2.1f" % (intron.donor.pssm_score, intron.acceptor.pssm_score),
print "%2.1f,%2.1f" % (intron.binary_entropy_donor,intron.binary_entropy_acceptor)
################################################################
if aaDistance > kwargs['max_aa_offset']:
# break out; ordered by length can never result in
# a proper projected intron
break
# filter out less relevant ones compared to complete set of results
projected_introns = _filter_projected_introns(projected_introns)
# and return a list of ProjectedIntronConnectingOrfs
return projected_introns
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_intron(pfD,
pfA,
queryorsbjct,
verbose=False,
**kwargs):
"""
Project splicesites from SBJCT intron on continious QUERY PacbPORFs
@type pfD: PacbPORF object
@param pfD: PacbPORF object that has to deliver (aligned) donor sites
@type pfA: PacbPORF object
@param pfA: PacbPORF object that has to deliver (aligned) acceptor sites
@type queryorsbjct: string
@param queryorsbjct: literal string 'query' or 'sbjct'
@type verbose: Boolean
@param verbose: print debugging info to STDOUT when True
@attention: see pacb.connecting.merge_orfs_with_intron for **kwargs)
@rtype: list
@return: list with ProjectedIntrons (from Sbjct on Query)
"""
# input validation
IsPacbPORF(pfD)
IsPacbPORF(pfA)
# edit **kwargs dictionary for some forced attributes
_update_kwargs(kwargs, KWARGS_PROJECTED_INTRON)
### if not kwargs.has_key('projected_intron_max_nt_offset'):
### kwargs['projected_intron_max_nt_offset'] = PROJECTED_INTRON_MAX_NT_OFFSET
### if not kwargs.has_key('projected_intron_max_aa_offset'):
### kwargs['projected_intron_max_aa_offset'] = PROJECTED_INTRON_MAX_AA_OFFSET
# settings for minimal alignment entropy score
min_donor_site_alignment_entropy = 0.0
min_acceptor_site_alignment_entropy = 0.0
ELEGIABLE_SPLICE_SITE_AA_RANGE = 75
sposD = pfD._get_original_alignment_pos_start()
eposD = pfD._get_original_alignment_pos_end()
sposA = pfA._get_original_alignment_pos_start()
eposA = pfA._get_original_alignment_pos_end()
if queryorsbjct == "query":
# Orfs of SBJCT must be identical
IsIdenticalOrfs(pfD.orfS, pfA.orfS)
donorOrf = pfD.orfQ
accepOrf = pfA.orfQ
prjctOrf = pfD.orfS # pfD.orfS == pfA.orfS
dStart = sposD.query_dna_start # ALIGNED start of donorPacbPORF
dEnd = pfD.query_dna_end # ABSOLUTE end of donorPacbPORF
aStart = pfA.query_dna_start # ABSOLUTE start of acceptorPacbPORF
aEnd = eposA.query_dna_end # ALIGNED end of acceptorPacbPORF
outOfAlignmentAttribute = "sbjct_dna_start"
# calculate elegiable splice site range
qdr = pfD.alignment_dna_range_query()
qar = pfA.alignment_dna_range_query()
min_donor_pos = max(
[min(qdr),
max(qdr) - (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
max_accep_pos = min(
[max(qar),
min(qar) + (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
elif queryorsbjct == "sbjct":
# Orfs of QUERY must be identical
IsIdenticalOrfs(pfD.orfQ, pfA.orfQ)
donorOrf = pfD.orfS
accepOrf = pfA.orfS
prjctOrf = pfD.orfQ # pfD.orfQ == pfA.orfQ
dStart = sposD.sbjct_dna_start # ALIGNED start of donorPacbPORF
dEnd = pfD.sbjct_dna_end # ABSOLUTE end of donorPacbPORF
aStart = pfA.sbjct_dna_start # ABSOLUTE start of acceptorPacbPORF
aEnd = eposA.sbjct_dna_end # ALIGNED end of acceptorPacbPORF
outOfAlignmentAttribute = "query_dna_start"
# calculate elegiable splice site range
sdr = pfD.alignment_dna_range_sbjct()
sar = pfA.alignment_dna_range_sbjct()
min_donor_pos = max(
[min(sdr),
max(sdr) - (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
max_accep_pos = min(
[max(sar),
min(sar) + (ELEGIABLE_SPLICE_SITE_AA_RANGE * 3)])
else:
message = "'queryorsbjct' (%s), not 'query' or 'sbjct'" % queryorsbjct
raise InproperlyAppliedArgument, message
# predict introns only in `queryorsbjct` Orfs
# introns is a list of IntronConnectingOrfs objects
introns = merge_orfs_with_intron(donorOrf,
accepOrf,
min_donor_pos=min_donor_pos,
max_acceptor_pos=max_accep_pos,
order_by='length',
**kwargs)
# return list with projected introns
projected_introns = []
# gather unique donor and acceptor positions from list
# of IntronConnectingOrfs
for intron in introns:
# break if intron is to large
if kwargs['max_intron_nt_length'] and intron.length > kwargs[
'max_intron_nt_length']:
break
# continue if intron is to small
if kwargs['min_intron_nt_length'] and intron.length < kwargs[
'min_intron_nt_length']:
continue
# continue if intron has non-canonical features
# check if intron.start is on pfD;
# inframe-introns can be projected outside of pfD/pfA area
if intron.start <= dStart: continue
if intron.start >= dEnd: continue
# check if intron.end is on pfA;
# inframe-introns can be projected outside of pfD/pfA area
if intron.end <= aStart: continue
if intron.end >= aEnd: continue
if queryorsbjct == "sbjct":
# get positions of donor & acceptor in the PacbPORF alignment
donorPositionPos, phaseD = pfD.dnaposition_sbjct(
intron.donor.pos, forced_return=True)
accepPositionPos, phaseA = pfA.dnaposition_sbjct(
intron.acceptor.pos, forced_return=True)
# calculate projected distance on QUERY
posDposQuery = pfD._positions[donorPositionPos].query_pos
posAposQuery = pfA._positions[accepPositionPos].query_pos
aaDistance = posAposQuery - posDposQuery
else:
# get positions of donor & acceptor in the PacbPORF alignment
donorPositionPos, phaseD = pfD.dnaposition_query(
intron.donor.pos, forced_return=True)
accepPositionPos, phaseA = pfA.dnaposition_query(
intron.acceptor.pos, forced_return=True)
# calculate binary entropy from projected position on SBJCT
posDposSbjct = pfD._positions[donorPositionPos].sbjct_pos
posAposSbjct = pfA._positions[accepPositionPos].sbjct_pos
aaDistance = posAposSbjct - posDposSbjct
# calculate binary entropy score
entropyDonorSbjct = pfD.alignment_entropy(donorPositionPos,
method='donor')
entropyAcceptorSbjct = pfA.alignment_entropy(accepPositionPos,
method='acceptor')
# do distance check upon (projected) intron acceptance
if abs(aaDistance) <= kwargs['max_aa_offset']:
# check if we've runned out of the aligned part
outofalignedpacbporf = False
# get the projected donor position; mind the gap on this spot ;-)
while pfD._positions[
donorPositionPos].isa_gap and donorPositionPos > 0:
donorPositionPos -= 1
else:
projected_donor_position = getattr(
pfD._positions[donorPositionPos],
outOfAlignmentAttribute) + phaseD
if donorPositionPos == 0 and pfD._positions[
donorPositionPos].isa_gap:
print "WarningThatIsTackled::outofalignedpacbporf::donor"
outofalignedpacbporf = True
# get the projected acceptor position; mind the gap on this spot ;-)
while pfA._positions[accepPositionPos].isa_gap and len(
pfA._positions) > accepPositionPos + 1:
accepPositionPos += 1
else:
projected_accep_position = getattr(
pfA._positions[accepPositionPos],
outOfAlignmentAttribute) + phaseA
if accepPositionPos == len(
pfA._positions
) - 1 and pfA._positions[accepPositionPos].isa_gap:
print "WarningThatIsTackled::outofalignedpacbporf::acceptor"
outofalignedpacbporf = True
if not outofalignedpacbporf:
################################################################
# set some meta-data properties to the intron object
################################################################
# add distance score to intron
intron._distance = abs(aaDistance) * 3
# add Alignment Positional Periphery Score into objects
if queryorsbjct == "query":
succes = set_apps_intron_query(intron, pfD, pfA)
else:
succes = set_apps_intron_sbjct(intron, pfD, pfA)
# set GFF fsource attribute for recognition of intron sources
intron._gff['fsource'] = "ABGPprojecting"
# make a ProjectedIntronConnectingOrfs object
pico = ProjectedIntronConnectingOrfs(prjctOrf,
projected_donor_position,
projected_accep_position)
intron.binary_entropy_donor = entropyDonorSbjct
intron.binary_entropy_acceptor = entropyAcceptorSbjct
pico.add_projected_intron(intron)
pico.phase = intron.phase
projected_introns.append(pico)
################################################################
if verbose:
print "PROJ::", intron._distance,
print(pfD.orfQ.id, pfA.orfQ.id),
print(pfD.orfS.id, pfA.orfS.id),
print "%s-%snt" % (intron.donor.pos, intron.acceptor.pos),
print "%2.1f,%2.1f" % (intron.donor.pssm_score,
intron.acceptor.pssm_score),
print "%2.1f,%2.1f" % (intron.binary_entropy_donor,
intron.binary_entropy_acceptor)
################################################################
if aaDistance > kwargs['max_aa_offset']:
# break out; ordered by length can never result in
# a proper projected intron
break
# filter out less relevant ones compared to complete set of results
projected_introns = _filter_projected_introns(projected_introns)
# and return a list of ProjectedIntronConnectingOrfs
return projected_introns
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
