def convertMali2Alignlib( mali ):
'''convert a multiple alignment of type :class:`Mali`
into an alignlib multiple alignment object.
'''
import alignlib
m = alignlib.makeMultipleAlignment()
for identifier in mali.getIdentifiers():
a = alignlib.makeAlignatum( mali[identifier] )
m.add( a )
return m
def applyMethod(self, neighbours):
"""apply the method."""
# build multiple alignment
mali = alignlib.makeMultipleAlignment()
query_nid = neighbours.mQueryToken
sequence = self.mFasta.getSequence(query_nid)
mali.add(alignlib.makeAlignatum(sequence))
qseq = alignlib.makeSequence(sequence)
alignator = alignlib.makeAlignatorDPFull(alignlib.ALIGNMENT_GLOBAL,
-10.0, -1.0, True, True, True,
True)
for n in neighbours.mMatches:
if n.mSbjctToken == query_nid: continue
sequence = self.mFasta.getSequence(n.mSbjctToken)
blast_query2sbjct = n.getAlignment()
if blast_query2sbjct == None:
raise ValueError(
"AddaRealignment.py needs a reference alignment.")
realign_query2sbjct = alignlib.makeAlignmentVector()
sseq = alignlib.makeSequence(sequence)
qseq.useSegment(n.mQueryFrom, n.mQueryTo)
sseq.useSegment(n.mSbjctFrom, n.mSbjctTo)
realign_query2sbjct = alignlib.makeAlignmentVector()
alignator.align(realign_query2sbjct, qseq, sseq)
nidentical = alignlib.getAlignmentIdentity(realign_query2sbjct,
blast_query2sbjct,
alignlib.RR)
nblast = blast_query2sbjct.getNumAligned()
nrealigned = realign_query2sbjct.getNumAligned()
self.mOutfile.write( "%s\t%s\t%i\t%i\t%i\n" % \
(n.mQueryToken, n.mSbjctToken, nidentical, nblast, nrealigned ) )
if nidentical == nblast:
self.mNIdentical += 1
else:
self.mNDifferent += 1
def applyMethod(self, neighbours ):
"""apply the method."""
# build multiple alignment
mali = alignlib.makeMultipleAlignment()
query_nid = neighbours.mQueryToken
sequence = self.mFasta.getSequence( query_nid )
mali.add( alignlib.makeAlignatum( sequence ) )
qseq = alignlib.makeSequence( sequence )
alignator = alignlib.makeAlignatorDPFull( alignlib.ALIGNMENT_GLOBAL,
-10.0, -1.0, True, True, True, True)
for n in neighbours.mMatches:
if n.mSbjctToken == query_nid: continue
sequence = self.mFasta.getSequence( n.mSbjctToken )
blast_query2sbjct = n.getAlignment()
if blast_query2sbjct == None:
raise ValueError( "AddaRealignment.py needs a reference alignment.")
realign_query2sbjct = alignlib.makeAlignmentVector()
sseq = alignlib.makeSequence( sequence )
qseq.useSegment( n.mQueryFrom, n.mQueryTo )
sseq.useSegment( n.mSbjctFrom, n.mSbjctTo )
realign_query2sbjct = alignlib.makeAlignmentVector()
alignator.align( realign_query2sbjct, qseq, sseq )
nidentical = alignlib.getAlignmentIdentity( realign_query2sbjct, blast_query2sbjct, alignlib.RR )
nblast = blast_query2sbjct.getNumAligned()
nrealigned = realign_query2sbjct.getNumAligned()
self.mOutfile.write( "%s\t%s\t%i\t%i\t%i\n" % \
(n.mQueryToken, n.mSbjctToken, nidentical, nblast, nrealigned ) )
if nidentical == nblast:
self.mNIdentical += 1
else:
self.mNDifferent += 1
continue
if use_cds_id:
identifier = cds_identifier
else:
identifier = peptide_identifier
if options.output_format =="alignment":
options.stdout.write("\t".join( map(str, (identifier, alignlib.AlignmentFormatEmissions( map_p2c ),
len(cur_record.sequence), len(cds_sequences[identifier])) ) )+"\n")
elif options.output_format == "fasta":
map_p2c.switchRowCol()
alignatum = alignlib.makeAlignatum( c )
alignatum.mapOnAlignment( map_p2c, len(p) * 3 )
s = alignatum.getString()
if len(s) != len(p) * 3:
raise ValueError ("incomplete aligned string for %s: %s, cds=%s" % (cur_record.title, s, c ))
options.stdout.write( ">%s\n%s\n" % (identifier, s ))
noutput += 1
sys.stdout.flush()
E.info( "ninput=%i, noutput=%i, nnosequence=%i, nskipped=%i" % (ninput, noutput, nnosequence, nskipped) )
E.Stop()
def buildMali(self, query_nid, neighbours):
"""build a multiple alignment from a set of neighbours.
"""
# build multiple alignment
mali = alignlib.makeMultipleAlignment()
query_sequence = self.mFasta.getSequence(query_nid)
mali.add(alignlib.makeAlignatum(query_sequence))
qseq = alignlib.makeSequence(query_sequence)
alignator = alignlib.makeAlignatorDPFull(alignlib.ALIGNMENT_LOCAL, -10,
-2)
nskipped = 0
for n in neighbours[:self.mMaxNumNeighbours]:
if n.mSbjctToken == query_nid: continue
if n.mEvalue > self.mMaxEvalue:
nskipped += 1
continue
sequence = self.mFasta.getSequence(n.mSbjctToken)
E.debug("adding %s" % str(n))
map_query2sbjct = n.getAlignment()
if map_query2sbjct == None:
sseq = alignlib.makeSequence(sequence)
qseq.useSegment(n.mQueryFrom, n.mQueryTo)
sseq.useSegment(n.mSbjctFrom, n.mSbjctTo)
map_query2sbjct = alignlib.makeAlignmentVector()
alignator.align(map_query2sbjct, qseq, sseq)
if map_query2sbjct.getLength() == 0:
self.warn("empty alignment: %s" % str(n))
nskipped += 1
continue
if map_query2sbjct.getRowTo() > len(query_sequence):
self.warn( "alignment out of bounds for query: %i>%i, line=%s" %\
(map_query2sbjct.getRowTo(), len(query_sequence), str(n)))
nskipped += 1
continue
elif map_query2sbjct.getColTo() > len(sequence):
self.warn( "alignment out of bounds for sbjct: %i>%i, line=%s" %\
(map_query2sbjct.getColTo(), len(sequence), str(n)))
nskipped += 1
continue
try:
mali.add(alignlib.makeAlignatum(sequence),
map_query2sbjct,
mali_is_in_row=True,
insert_gaps_mali=False,
insert_gaps_alignatum=True,
use_end_mali=True,
use_end_alignatum=False)
except RuntimeError, msg:
self.warn("problem when building alignment for %s: msg=%s" %
(str(n), msg))
nskipped += 1
continue
def buildMali(self, query_nid, neighbours ):
"""build a multiple alignment from a set of neighbours.
"""
# build multiple alignment
mali = alignlib.makeMultipleAlignment()
query_sequence = self.mFasta.getSequence( query_nid )
mali.add( alignlib.makeAlignatum( query_sequence ) )
qseq = alignlib.makeSequence( query_sequence )
alignator = alignlib.makeAlignatorDPFull( alignlib.ALIGNMENT_LOCAL,
-10, -2)
nskipped = 0
for n in neighbours[:self.mMaxNumNeighbours]:
if n.mSbjctToken == query_nid: continue
if n.mEvalue > self.mMaxEvalue:
nskipped += 1
continue
sequence = self.mFasta.getSequence( n.mSbjctToken )
E.debug( "adding %s" % str(n) )
map_query2sbjct = n.getAlignment()
if map_query2sbjct == None:
sseq = alignlib.makeSequence( sequence )
qseq.useSegment( n.mQueryFrom, n.mQueryTo )
sseq.useSegment( n.mSbjctFrom, n.mSbjctTo )
map_query2sbjct = alignlib.makeAlignmentVector()
alignator.align( map_query2sbjct, qseq, sseq )
if map_query2sbjct.getLength() == 0:
self.warn( "empty alignment: %s" % str( n ) )
nskipped += 1
continue
if map_query2sbjct.getRowTo() > len(query_sequence):
self.warn( "alignment out of bounds for query: %i>%i, line=%s" %\
(map_query2sbjct.getRowTo(), len(query_sequence), str(n)))
nskipped += 1
continue
elif map_query2sbjct.getColTo() > len(sequence):
self.warn( "alignment out of bounds for sbjct: %i>%i, line=%s" %\
(map_query2sbjct.getColTo(), len(sequence), str(n)))
nskipped += 1
continue
try:
mali.add( alignlib.makeAlignatum( sequence ),
map_query2sbjct,
mali_is_in_row = True,
insert_gaps_mali = False,
insert_gaps_alignatum = True,
use_end_mali = True,
use_end_alignatum = False )
except RuntimeError, msg:
self.warn( "problem when building alignment for %s: msg=%s" % (str(n), msg))
nskipped += 1
continue