continue
##create ranges of accepted alignments
accept_alignment_ranges = [None] * pblen
#alignments[pbname].sort(key=lambda a: (a.send-a.sstart) * pow(a.pctid/100.0,2))
alignments[pbname].sort(key=lambda a: (a.send-a.sstart))
for alignment in alignments[pbname]:
for p in range(alignment.sstart-1,alignment.send):
accept_alignment_ranges[p] = alignment.qname
##
##find clr ranges
##
#find ranges
covered_ranges = map(lambda (s,e): CoverageRange(s,e,1.0,CovStat["COVERED"]),
getMarkedRanges(map(lambda c: 1 if not c == None else 0 , accept_alignment_ranges)))
uncovered_ranges = map(lambda (s,e): CoverageRange(s,e,0.7,CovStat["UNCOVERED"]),
getMarkedRanges(map(lambda c: 1 if c == None else 0 , accept_alignment_ranges)))
#remove uncorrected ends
uncovered_ranges = filter(lambda x: not (x.begin == 0 or x.end == pblen-1),uncovered_ranges)
joined_ranges = sorted(covered_ranges + uncovered_ranges, key=lambda x: x.begin)
#find the clr ranges
while True:
clr_ranges = get_contiguous_ranges(joined_ranges)
if( all(map(lambda y: y.pctid > CLR_ID_CUTOFF,clr_ranges))):
break
for cr in clr_ranges:
#skip clr ranges that are ok
if cr.pctid > CLR_ID_CUTOFF:
continue
##create ranges of accepted alignments
accept_alignment_ranges = [None] * pblen
#alignments[pbname].sort(key=lambda a: (a.send-a.sstart) * pow(a.pctid/100.0,2))
alignments[pbname].sort(key=lambda a: (a.send-a.sstart))
for alignment in alignments[pbname]:
for p in range(alignment.sstart-1,alignment.send):
accept_alignment_ranges[p] = alignment.qname
##
##find clr ranges
##
#find ranges
covered_ranges = map(lambda (s,e): CoverageRange(s,e,1.0,CovStat["COVERED"]),
getMarkedRanges(map(lambda c: 1 if not c == None else 0 , accept_alignment_ranges)))
uncovered_ranges = map(lambda (s,e): CoverageRange(s,e,0.7,CovStat["UNCOVERED"]),
getMarkedRanges(map(lambda c: 1 if c == None else 0 , accept_alignment_ranges)))
#remove uncorrected ends
uncovered_ranges = filter(lambda x: not (x.begin == 0 or x.end == pblen-1),uncovered_ranges)
joined_ranges = sorted(covered_ranges + uncovered_ranges, key=lambda x: x.begin)
#find the clr ranges
while True:
clr_ranges = get_contiguous_ranges(joined_ranges)
if( all(map(lambda y: y.pctid > CLR_ID_CUTOFF,clr_ranges))):
break
for cr in clr_ranges:
#skip clr ranges that are ok
if cr.pctid > CLR_ID_CUTOFF:
alignmentIt = getNucmerAlignmentIterator(afh)
sys.stderr.write("Loaded Alignments\n");
counter = 0
for name,group in groupby(alignmentIt, lambda x: x.sname):
#build coverage vector
cov = getCoverageFromNucAlignments(group)
#mark the regions with 0 (no) coverage as 1 and change
#everything else to 0
cov_inv = map(lambda c: 1 if c == 0 else 0, cov)
#ranges with zero coverage
zero_cov_ranges = getMarkedRanges(cov_inv)
seq = reads[name]
#calculate GC % for windows of GC_WINDOW_SIZE
gc_sliding_window = getGCSlidingWindow(seq, GC_WINDOW_SIZE)
#filter gaps that are at > MIN_COV_GAP
#and have at least one base > GC_THRESHOLD
#take the sum of the lengths of all of the regions
gc_gap_bases = sum(map(lambda (s,e):
e-s if e-s > MIN_COV_GAP and any(map(lambda x: True if x > GC_THRESHOLD else False, gc_sliding_window[s:e])) else 0,
zero_cov_ranges))
ofh.write("%s\t%d\n" % (name,gc_gap_bases))
if counter % 10000 == 0:
ftbases = open(sys.argv[3] + ".uncov.total.bases", "w")
pcov = [] # pct cov
total_bases = 0
total_uncovered_bases = 0
for pbname, alignments in groupby(getNucmerAlignmentIterator(fh), lambda x: x.sname):
a = list(alignments)
cov = getCoverageFromNucAlignments(a)
# mark the 0 coverage regions
zcov = map(lambda c: 1 if c == 0 else 0, cov)
# ranges with 0 coverage
zcov_ranges = getMarkedRanges(zcov)
# only look at the gaps larger than the min gap size
zcov_ranges_filt = filter(lambda (x, y): y - x > COV_GAP_MIN, zcov_ranges)
# write out the regions that pass filter to region file
freg.write("\t".join([pbname, " ".join(map(lambda t: "%d,%d" % t, zcov_ranges_filt))]) + "\n")
total_bases += a[0].slen
for rbeg, rend in zcov_ranges_filt:
total_uncovered_bases += rend - rbeg
pct_ranges = map(
lambda (rb, re): (int(float(rb) / (a[0].slen - 1) * 100), int(float(re) / (a[0].slen - 1) * 100)),
zcov_ranges_filt,
)
pcov = [] #pct cov
total_bases = 0
total_uncovered_bases = 0
for pbname, alignments in groupby(getNucmerAlignmentIterator(fh),
lambda x: x.sname):
a = list(alignments)
cov = getCoverageFromNucAlignments(a)
#mark the 0 coverage regions
zcov = map(lambda c: 1 if c == 0 else 0, cov)
#ranges with 0 coverage
zcov_ranges = getMarkedRanges(zcov)
#only look at the gaps larger than the min gap size
zcov_ranges_filt = filter(lambda (x, y): y - x > COV_GAP_MIN, zcov_ranges)
#write out the regions that pass filter to region file
freg.write("\t".join(
[pbname, " ".join(map(lambda t: "%d,%d" % t, zcov_ranges_filt))]) +
"\n")
total_bases += a[0].slen
for rbeg, rend in zcov_ranges_filt:
total_uncovered_bases += rend - rbeg
pct_ranges = map(
lambda (rb, re):
alignmentIt = getNucmerAlignmentIterator(afh)
sys.stderr.write("Loaded Alignments\n")
counter = 0
for name, group in groupby(alignmentIt, lambda x: x.sname):
#build coverage vector
cov = getCoverageFromNucAlignments(group)
#mark the regions with 0 (no) coverage as 1 and change
#everything else to 0
cov_inv = map(lambda c: 1 if c == 0 else 0, cov)
#ranges with zero coverage
zero_cov_ranges = getMarkedRanges(cov_inv)
seq = reads[name]
#calculate GC % for windows of GC_WINDOW_SIZE
gc_sliding_window = getGCSlidingWindow(seq, GC_WINDOW_SIZE)
#filter gaps that are at > MIN_COV_GAP
#and have at least one base > GC_THRESHOLD
#take the sum of the lengths of all of the regions
gc_gap_bases = sum(
map(
lambda (s, e): e - s if e - s > MIN_COV_GAP and any(
map(lambda x: True if x > GC_THRESHOLD else False,
gc_sliding_window[s:e])) else 0, zero_cov_ranges))
