def count_coverage(readers, comments=True):
primary = readers[0]
secondary = readers[1]
secondary_copy = readers[2]
rightTree = quicksect.IntervalTree()
for item in secondary:
if type(item) is GenomicInterval:
rightTree.insert(item, secondary.linenum, item.fields)
bitsets = secondary_copy.binned_bitsets()
global full, partial
for interval in primary:
if type(interval) is Header:
yield interval
if type(interval) is Comment and comments:
yield interval
elif type(interval) == GenomicInterval:
chrom = interval.chrom
start = int(interval.start)
end = int(interval.end)
full = 0
partial = 0
if chrom not in bitsets:
bases_covered = 0
percent = 0.0
full = 0
partial = 0
else:
bases_covered = bitsets[chrom].count_range(start, end - start)
if (end - start) == 0:
percent = 0
else:
percent = float(bases_covered) / float(end - start)
if bases_covered:
root = rightTree.chroms[
chrom] #root node for the chrom tree
counter(root, start, end)
interval.fields.append(str(bases_covered))
interval.fields.append(str(percent))
interval.fields.append(str(full))
interval.fields.append(str(partial))
yield interval
def proximal_region_finder(readers, region, comments=True):
"""
Returns an iterator that yields elements of the form [ <original_interval>, <closest_feature> ].
Intervals are GenomicInterval objects.
"""
primary = readers[0]
features = readers[1]
either = False
if region == 'Upstream':
up, down = True, False
elif region == 'Downstream':
up, down = False, True
else:
up, down = True, True
if region == 'Either':
either = True
# Read features into memory:
rightTree = quicksect.IntervalTree()
for item in features:
if type(item) is GenomicInterval:
rightTree.insert(item, features.linenum, item)
for interval in primary:
if type(interval) is Header:
yield interval
if type(interval) is Comment and comments:
yield interval
elif type(interval) == GenomicInterval:
chrom = interval.chrom
start = int(interval.start)
end = int(interval.end)
strand = interval.strand
if chrom not in rightTree.chroms:
continue
else:
root = rightTree.chroms[chrom] #root node for the chrom tree
result_up = []
result_down = []
if (strand == '+' and up) or (strand == '-' and down):
#upstream +ve strand and downstream -ve strand cases
get_closest_feature(root, 1, start, None,
lambda node: result_up.append(node),
None)
if (strand == '+' and down) or (strand == '-' and up):
#downstream +ve strand and upstream -ve strand case
get_closest_feature(root, 0, None, end - 1, None,
lambda node: result_down.append(node))
if result_up:
if len(
result_up
) > 1: #The results_up list has a list of intervals upstream to the given interval.
ends = []
for n in result_up:
ends.append(n.end)
res_ind = ends.index(
max(ends)
) #fetch the index of the closest interval i.e. the interval with the max end from the results_up list
else:
res_ind = 0
if not (either):
yield [interval, result_up[res_ind].other]
if result_down:
if not (either):
#The last element of result_down will be the closest element to the given interval
yield [interval, result_down[-1].other]
if either and (result_up or result_down):
iter_val = []
if result_up and result_down:
if abs(start - int(result_up[res_ind].end)) <= abs(
end - int(result_down[-1].start)):
iter_val = [interval, result_up[res_ind].other]
else:
#The last element of result_down will be the closest element to the given interval
iter_val = [interval, result_down[-1].other]
elif result_up:
iter_val = [interval, result_up[res_ind].other]
elif result_down:
#The last element of result_down will be the closest element to the given interval
iter_val = [interval, result_down[-1].other]
yield iter_val
sp_file.seek(0)
win = NiceReaderWrapper( fileinput.FileInput( int_file ),
chrom_col=chr_col_i,
start_col=start_col_i,
end_col=end_col_i,
strand_col=strand_col_i,
fix_strand=True)
indel = NiceReaderWrapper( fileinput.FileInput( sp_file.name ),
chrom_col=1,
start_col=sort_col,
end_col=sort_col+1,
strand_col=-1,
fix_strand=True)
indelTree = quicksect.IntervalTree()
for item in indel:
if type( item ) is GenomicInterval:
indelTree.insert( item, indel.linenum, item.fields )
result=[]
global full, blk_len, blk_list
for interval in win:
if type( interval ) is Header:
pass
if type( interval ) is Comment:
pass
elif type( interval ) == GenomicInterval:
chrom = interval.chrom
start = int(interval.start)
end = int(interval.end)
def main():
infile = sys.argv[1]
for i, line in enumerate(file(infile)):
line = line.rstrip('\r\n')
if len(line) > 0 and not line.startswith('#'):
elems = line.split('\t')
break
if i == 30:
break # Hopefully we'll never get here...
if len(elems) != 15:
stop_err(
"This tool only works on tabular data output by 'Extract Orthologous Microsatellites from pair-wise alignments' tool. The data in your input dataset is either missing or not formatted properly."
)
global winspecies, speciesind
if region == 'win':
if dbkey_i in elems[1]:
winspecies = 1
speciesind = 1
elif dbkey_i in elems[8]:
winspecies = 2
speciesind = 8
else:
stop_err(
"The species build corresponding to your interval file is not present in the Microsatellite file."
)
fin = open(infile, 'r')
skipped = 0
blk = 0
win = 0
linestr = ""
if region == 'win':
msats = NiceReaderWrapper(fileinput.FileInput(infile),
chrom_col=speciesind,
start_col=speciesind + 1,
end_col=speciesind + 2,
strand_col=-1,
fix_strand=True)
msatTree = quicksect.IntervalTree()
for item in msats:
if type(item) is GenomicInterval:
msatTree.insert(item, msats.linenum, item.fields)
for iline in fint:
try:
iline = iline.rstrip('\r\n')
if not (iline) or iline == "":
continue
ielems = iline.strip("\r\n").split('\t')
ichr = ielems[chr_col_i]
istart = int(ielems[start_col_i])
iend = int(ielems[end_col_i])
isrc = "%s.%s" % (dbkey_i, ichr)
if isrc not in msatTree.chroms:
continue
result = []
root = msatTree.chroms[isrc] #root node for the chrom
counter(root, istart, iend, lambda node: result.append(node))
if not (result):
continue
tmpfile1 = tempfile.NamedTemporaryFile('wb+')
for node in result:
tmpfile1.write("%s\n" % "\t".join(node.other))
tmpfile1.seek(0)
output_writer(iline, tmpfile1.readlines())
except:
skipped += 1
if skipped:
print "Skipped %d intervals as invalid." % (skipped)
elif region == 'align':
if s_group_cols[0] != -1:
print >> fout, "#Window\tSpecies_1\tSpecies_2\tGroupby_Feature\tSubGroupby_Feature\tMutability\tCount"
else:
print >> fout, "#Window\tSpecies_1\tWindow_Start\tWindow_End\tSpecies_2\tGroupby_Feature\tMutability\tCount"
prev_bnum = -1
try:
for line in fin:
line = line.strip("\r\n")
if not (line) or line == "":
continue
elems = line.split('\t')
try:
assert int(elems[0])
assert len(elems) == 15
except:
continue
new_bnum = int(elems[0])
if new_bnum != prev_bnum:
if prev_bnum != -1:
output_writer(
prev_bnum,
linestr.strip().replace('\r', '\n').split('\n'))
linestr = line + "\n"
else:
linestr += line
linestr += "\n"
prev_bnum = new_bnum
output_writer(prev_bnum,
linestr.strip().replace('\r', '\n').split('\n'))
except Exception, ea:
print >> sys.stderr, ea
skipped += 1
if skipped:
print "Skipped %d lines as invalid." % (skipped)