def clusterTEentries(entries, mmpenalty):
"""
cluster non-overlapping TE insertions of the same family.
a scoring system is used to decide whether non-overlapping TE insertions will be clustered
"""
clustered = []
tes = sorted(entries, key=lambda e: e.start)
while (len(tes) > 0):
a = tes.pop(0)
score = a.score
end = a.end
while (len(tes) > 0):
totest = tes[0]
gap = totest.start - end
gappen = gap * mmpenalty
scorewithgap = score - gappen
if scorewithgap < 0: # if the score with the gap reaches zero; break! similar dynamic programming
break
# novel high score:
totestscore = scorewithgap + totest.score
if (totestscore >= score):
tes.pop(0)
score = totestscore
end = totest.end
else:
#no new high score
break
ne = GTFEntry(a.chr, a.source, a.feature, a.start, end, score,
a.strand, a.frame, a.comment)
ne.target = a.target
clustered.append(ne)
return clustered
def clusterTEentries(entries,mmpenalty): """ cluster non-overlapping TE insertions of the same family. a scoring system is used to decide whether non-overlapping TE insertions will be clustered """ clustered=[] tes=sorted(entries,key=lambda e: e.start) while(len(tes)>0): a=tes.pop(0) score=a.score end=a.end while(len(tes)>0): totest=tes[0] gap=totest.start-end gappen=gap*mmpenalty scorewithgap=score-gappen if scorewithgap<0: # if the score with the gap reaches zero; break! similar dynamic programming break # novel high score: totestscore=scorewithgap+totest.score if(totestscore>=score): tes.pop(0) score=totestscore end=totest.end else: #no new high score break ne =GTFEntry(a.chr,a.source,a.feature,a.start,end,score,a.strand,a.frame,a.comment) ne.target=a.target clustered.append(ne) return clustered
def mergeTEentries(entries, matchscore):
"""
merge overlapping TE entries of the same family.
New score is the total length
"""
merged = []
tes = sorted(entries, key=lambda e: e.start)
while(len(tes) > 0):
a = tes.pop(0)
start = a.start
highestend = a.end
while(len(tes) > 0 and tes[0].start <= highestend):
b = tes.pop(0)
if(b.end > highestend):
highestend = b.end
ne = GTFEntry(a.chr, a.source, a.feature, start, highestend, matchscore * float(highestend - start), a.strand, a.frame, a.comment)
ne.target = a.target
merged.append(ne)
return merged
def mergeTEentries(entries,matchscore): """ merge overlapping TE entries of the same family. New score is the total length """ merged=[] tes=sorted(entries,key=lambda e: e.start) while(len(tes)>0): a=tes.pop(0) start=a.start highestend=a.end while(len(tes)>0 and tes[0].start<=highestend): b=tes.pop(0) if(b.end>highestend): highestend=b.end ne =GTFEntry(a.chr,a.source,a.feature,start,highestend,matchscore*float(highestend-start),a.strand,a.frame,a.comment) ne.target=a.target merged.append(ne) return merged
def get_querygtflist(nucmerentries):
#chr, source, feature, start, end, score, strand, frame, comment (comment is unparsed)
gtfes = []
for nm in nucmerentries:
comment = "{0}:{1}-{2} count={3}".format(nm.rname, nm.rstart, nm.rend,
nm.count)
gtfes.append(
GTFEntry(nm.qname, "nucmer", "orthologous", nm.qstart, nm.qend,
".", ".", ".", comment))
return gtfes