def alignSeqs(seqs,*rest):
distXY,X,Y=alignSeq(seqs[0],seqs[1],*rest)
return (distXY,[X,Y])
def formatalign2D(alignment,seq=None):
"Formats the alignment for human use"
if not alignment:
return "No alignment\n"
outStrIO=StringIO()
try:
from editdist import alignSeq
except ImportError:
print "Using subb alignment"
def alignSeq(xseq,yseq,*rest):
"Stubb for alignment. Does nothing really"
l=max(len(xseq),len(yseq))
return (0,xseq.ljust(l).replace(" ","-"),yseq.rjust(l).replace(" ","-"))
def formatAlnSeq(xaln,yaln,xname,yname,xstart,ystart,motifAln="",linelen=60):
"""Formats the DNA sequence alignment output"""
outstr="Sequence 1: %s\nSequence 2: %s\n\n"%(xname,yname)
n=len(xaln)
xpos,ypos=xstart+1,ystart+1
for i in range(0,n,linelen):
xline,yline,mline=xaln[i:i+linelen],yaln[i:i+linelen],motifAln[i:i+linelen]
outstr+="%7d : %s\n%7d : %s\n %s\n\n"%(xpos,xline,ypos,yline,mline)
xpos+=len(xline)-xline.count("-")
ypos+=len(yline)-yline.count("-")
return outstr
outStrIO.write("### lambda=%g mu=%g nu=%g xi=%g Nucleotides per rotation=%g\n"%(alignment.Lambda,alignment.Mu,alignment.Nu,alignment.Xi,alignment.nuc_per_rotation))
outStrIO.write("### D[%s][%s]\nNote! First nucleotide at position 1 (one) and binding site at zero!\n"%(alignment.x_name,alignment.y_name))
xname,yname=alignment.x_name,alignment.y_name
xe,ye=0,0
if seq and not (seq.has_key(xname) and seq.has_key(yname)):
seq=None
if seq:
outStrIO.write("Sequence %s:\n%s\n\nSequence %s:\n%s\n"%(xname,seq.describe(xname),yname,seq.describe(yname)))
# goodAlign= [ (x,y,Score,Motif,(startX,endX),(startY,endY),Strand) ]
for i,goodAlign in zip(range(1,len(alignment.bestAlignments)+1),alignment.bestAlignments):
if len(goodAlign)==0:
continue
if seq:
xstart=max(goodAlign[0].beginX-10,0)
ystart=max(goodAlign[0].beginY-10,0)
xend=min(goodAlign[-1].endX+10,len(seq[xname]))
yend=min(goodAlign[-1].endY+10,len(seq[yname]))
xseq=seq[xname][xstart:xend]
yseq=seq[yname][ystart:yend]
xaln=""
yaln=""
maln=""
xadded=xstart
yadded=ystart
outStrIO.write("\n### Alignment No %d ###\n"%(i,))
#for (x,y,score,motif,xcoord,ycoord,strand) in goodAlign:
for alns in goodAlign:
outStrIO.write("D[%d][%d]=%.2f %s (%d,%d) <=> (%d,%d) %s\n"%(alns.seqX,alns.seqY,alns.score,alns.motif,alns.beginX,alns.endX,alns.beginY,alns.endY,alns.strand))
if seq:
y2add=yseq[yadded-ystart:alns.beginY-1-ystart].lower()
x2add=xseq[xadded-xstart:alns.beginX-1-xstart].lower()
#alnFmt="%%s%%-%ds%%s"%(max(len(y2add),len(x2add)))
#alnFmt="%s%s%s"
siteLen=alns.endX-alns.beginX+1
alnFmt="%%s%%s%%-%ds"%(siteLen)
assert(len(xseq[alns.beginX-1-xstart:alns.endX-xstart])==siteLen)
distYX,y2add,x2add=alignSeq(y2add,x2add,1,1)
yaln=alnFmt%(yaln,y2add,yseq[alns.beginY-1-ystart:alns.endY-ystart].upper())
xaln=alnFmt%(xaln,x2add,xseq[alns.beginX-1-xstart:alns.endX-xstart].upper())
maln=alnFmt%(maln," "*len(y2add),alns.motif[-siteLen:])
xadded,yadded=alns.endX,alns.endY
if seq:
distYX,y2add,x2add=alignSeq(yseq[yadded-ystart:yend-ystart].lower(),xseq[xadded-xstart:xend-xstart].lower())
yaln+=y2add
xaln+=x2add
outStrIO.write("\n"+formatAlnSeq(xaln.replace(" ","-"),yaln.replace(" ","-"),xname,yname,xstart,ystart,maln))
#outstr+="%s\n%s\n"%(xaln.replace(" ","-"),yaln.replace(" ","-"))
outStrIO.write("### Alignment took %.1f CPU seconds.\n"%(alignment.secs_to_align))
return outStrIO.getvalue()