def main():
distance = 0
minregions = 2
output = 1
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
if options.distance:
distance = int( options.distance )
if options.overlap:
distance = -1 * int( options.overlap )
if options.output:
output = int( options.output )
if options.minregions:
minregions = int( options.minregions )
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
# Get the cluster tree
try:
clusters, extra = find_clusters( g1, mincols=distance, minregions=minregions)
except ParseError, exc:
fail( "Invalid file format: %s" % str( exc ) )
def main():
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg(
options.cols2)
in_fname, in2_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
g2 = NiceReaderWrapper(fileinput.FileInput(in2_fname),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True)
out_file = open(out_fname, "w")
try:
for line in coverage([g1, g2]):
if type(line) is GenomicInterval:
out_file.write("%s\n" % "\t".join(line.fields))
else:
out_file.write("%s\n" % line)
except ParseError, exc:
out_file.close()
fail("Invalid file format: %s" % str(exc))
def main():
mincols = 1
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg( options.cols2 )
if options.mincols:
mincols = int( options.mincols )
pieces = bool( options.pieces )
in1_gff_format = bool( options.gff1 )
in2_gff_format = bool( options.gff2 )
in_fname, in2_fname, out_fname = args
except:
doc_optparse.exception()
# Set readers to handle either GFF or default format.
if in1_gff_format:
in1_reader_wrapper = GFFReaderWrapper
else:
in1_reader_wrapper = NiceReaderWrapper
if in2_gff_format:
in2_reader_wrapper = GFFReaderWrapper
else:
in2_reader_wrapper = NiceReaderWrapper
g1 = in1_reader_wrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
if in1_gff_format:
# Intersect requires coordinates in BED format.
g1.convert_to_bed_coord = True
g2 = in2_reader_wrapper( fileinput.FileInput( in2_fname ),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True )
if in2_gff_format:
# Intersect requires coordinates in BED format.
g2.convert_to_bed_coord = True
out_file = open( out_fname, "w" )
try:
for feature in intersect( [g1, g2], pieces=pieces, mincols=mincols ):
if isinstance( feature, GFFFeature ):
# Convert back to GFF coordinates since reader converted automatically.
convert_bed_coords_to_gff( feature )
for interval in feature.intervals:
out_file.write( "%s\n" % "\t".join( interval.fields ) )
elif isinstance( feature, GenomicInterval ):
out_file.write( "%s\n" % "\t".join( feature.fields ) )
else:
out_file.write( "%s\n" % feature )
except ParseError, e:
out_file.close()
fail( "Invalid file format: %s" % str( e ) )
def main():
distance = 0
minregions = 2
output = 1
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
if options.distance:
distance = int(options.distance)
if options.overlap:
distance = -1 * int(options.overlap)
if options.output:
output = int(options.output)
if options.minregions:
minregions = int(options.minregions)
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
# Get the cluster tree
try:
clusters, extra = find_clusters(g1,
mincols=distance,
minregions=minregions)
except ParseError, exc:
fail("Invalid file format: %s" % str(exc))
def main():
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg( options.cols2 )
in_fname, in2_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
g2 = NiceReaderWrapper( fileinput.FileInput( in2_fname ),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True )
out_file = open( out_fname, "w" )
try:
for line in coverage( [g1, g2] ):
if type( line ) is GenomicInterval:
out_file.write( "%s\n" % "\t".join( line.fields ) )
else:
out_file.write( "%s\n" % line )
except ParseError, exc:
out_file.close()
fail( "Invalid file format: %s" % str( exc ) )
def subtract_files(cls, fn1, fn2, out_fn):
g1 = NiceReaderWrapper(fileinput.FileInput(fn1), fix_strand=True)
g2 = NiceReaderWrapper(fileinput.FileInput(fn2), fix_strand=True)
out_file = open(out_fn, "w")
try:
for feature in subtract([g1, g2], pieces=True, mincols=1):
out_file.write("%s\n" % feature)
except ParseError, exc:
out_file.close()
fail("Invalid file format: %s" % str(exc))
def main():
mincols = 1
leftfill = False
rightfill = False
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg(
options.cols2)
if options.mincols:
mincols = int(options.mincols)
if options.fill:
if options.fill == "both":
rightfill = leftfill = True
else:
rightfill = options.fill == "right"
leftfill = options.fill == "left"
in_fname, in2_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
g2 = NiceReaderWrapper(fileinput.FileInput(in2_fname),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True)
out_file = open(out_fname, "w")
try:
for outfields in join(g1,
g2,
mincols=mincols,
rightfill=rightfill,
leftfill=leftfill):
if type(outfields) is list:
out_file.write("%s\n" % "\t".join(outfields))
else:
out_file.write("%s\n" % outfields)
except ParseError, exc:
out_file.close()
fail("Invalid file format: %s" % str(exc))
def main():
mincols = 1
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
if options.mincols:
mincols = int(options.mincols)
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
out_file = open(out_fname, "w")
try:
for line in merge(g1, mincols=mincols):
if options.threecol:
if type(line) is GenomicInterval:
out_file.write(
"%s\t%s\t%s\n" %
(line.chrom, str(line.startCol), str(line.endCol)))
elif type(line) is list:
out_file.write("%s\t%s\t%s\n" %
(line[chr_col_1], str(line[start_col_1]),
str(line[end_col_1])))
else:
out_file.write("%s\n" % line)
else:
if type(line) is GenomicInterval:
out_file.write("%s\n" % "\t".join(line.fields))
elif type(line) is list:
out_file.write("%s\n" % "\t".join(line))
else:
out_file.write("%s\n" % line)
except ParseError as exc:
out_file.close()
fail("Invalid file format: %s" % str(exc))
out_file.close()
if g1.skipped > 0:
print(skipped(g1, filedesc=" of 1st dataset"))
def main():
sameformat = False
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg( options.cols2 )
if options.sameformat:
sameformat = True
in_file_1, in_file_2, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_file_1 ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
g2 = NiceReaderWrapper( fileinput.FileInput( in_file_2 ),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True )
out_file = open( out_fname, "w" )
try:
for line in concat( [g1, g2], sameformat=sameformat ):
if type( line ) is GenomicInterval:
out_file.write( "%s\n" % "\t".join( line.fields ) )
else:
out_file.write( "%s\n" % line )
except ParseError as exc:
out_file.close()
fail( "Invalid file format: %s" % str( exc ) )
out_file.close()
if g1.skipped > 0:
print(skipped( g1, filedesc=" of 1st dataset" ))
if g2.skipped > 0:
print(skipped( g2, filedesc=" of 2nd dataset" ))
def main():
mincols = 1
leftfill = False
rightfill = False
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg( options.cols2 )
if options.mincols:
mincols = int( options.mincols )
if options.fill:
if options.fill == "both":
rightfill = leftfill = True
else:
rightfill = options.fill == "right"
leftfill = options.fill == "left"
in_fname, in2_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
g2 = NiceReaderWrapper( fileinput.FileInput( in2_fname ),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True )
out_file = open( out_fname, "w" )
try:
for outfields in join(g1, g2, mincols=mincols, rightfill=rightfill, leftfill=leftfill):
if type( outfields ) is list:
out_file.write( "%s\n" % "\t".join( outfields ) )
else:
out_file.write( "%s\n" % outfields )
except ParseError, exc:
out_file.close()
fail( "Invalid file format: %s" % str( exc ) )
def main():
mincols = 1
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
if options.mincols:
mincols = int( options.mincols )
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
out_file = open( out_fname, "w" )
try:
for line in merge(g1, mincols=mincols):
if options.threecol:
if type( line ) is GenomicInterval:
out_file.write( "%s\t%s\t%s\n" % ( line.chrom, str( line.startCol ), str( line.endCol ) ) )
elif type( line ) is list:
out_file.write( "%s\t%s\t%s\n" % ( line[chr_col_1], str( line[start_col_1] ), str( line[end_col_1] ) ) )
else:
out_file.write( "%s\n" % line )
else:
if type( line ) is GenomicInterval:
out_file.write( "%s\n" % "\t".join( line.fields ) )
elif type( line ) is list:
out_file.write( "%s\n" % "\t".join( line ) )
else:
out_file.write( "%s\n" % line )
except ParseError as exc:
out_file.close()
fail( "Invalid file format: %s" % str( exc ) )
out_file.close()
if g1.skipped > 0:
print(skipped( g1, filedesc=" of 1st dataset" ))
def main():
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
try:
bases = base_coverage(g1)
except ParseError, exc:
fail( "Invalid file format: %s" % str( exc ) )
def main():
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
try:
bases = base_coverage(g1)
except ParseError, exc:
fail("Invalid file format: %s" % str(exc))
def main():
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
try:
bases = base_coverage(g1)
except ParseError as exc:
fail( "Invalid file format: %s" % str( exc ) )
out_file = open( out_fname, "w" )
out_file.write( "%s\n" % str( bases ) )
out_file.close()
if g1.skipped > 0:
print(skipped( g1, filedesc="" ))
def main():
distance = 0
minregions = 2
output = 1
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
if options.distance:
distance = int( options.distance )
if options.overlap:
distance = -1 * int( options.overlap )
if options.output:
output = int( options.output )
if options.minregions:
minregions = int( options.minregions )
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
# Get the cluster tree
try:
clusters, extra = find_clusters( g1, mincols=distance, minregions=minregions)
except ParseError as exc:
fail( "Invalid file format: %s" % str( exc ) )
f1 = open( in_fname, "r" )
out_file = open( out_fname, "w" )
# If "merge"
if output == 1:
fields = ["." for x in range(max(g1.chrom_col, g1.start_col, g1.end_col) + 1)]
for chrom, tree in clusters.items():
for start, end, lines in tree.getregions():
fields[g1.chrom_col] = chrom
fields[g1.start_col] = str(start)
fields[g1.end_col] = str(end)
out_file.write( "%s\n" % "\t".join( fields ) )
# If "filtered" we preserve order of file and comments, etc.
if output == 2:
linenums = dict()
for chrom, tree in clusters.items():
for linenum in tree.getlines():
linenums[linenum] = 0
linenum = -1
f1.seek(0)
for line in f1.readlines():
linenum += 1
if linenum in linenums or linenum in extra:
out_file.write( "%s\n" % line.rstrip( "\n\r" ) )
# If "clustered" we output original intervals, but near each other (i.e. clustered)
if output == 3:
linenums = list()
f1.seek(0)
fileLines = f1.readlines()
for chrom, tree in clusters.items():
for linenum in tree.getlines():
out_file.write( "%s\n" % fileLines[linenum].rstrip( "\n\r" ) )
# If "minimum" we output the smallest interval in each cluster
if output == 4 or output == 5:
linenums = list()
f1.seek(0)
fileLines = f1.readlines()
for chrom, tree in clusters.items():
for start, end, lines in tree.getregions():
outsize = -1
outinterval = None
for line in lines:
# three nested for loops?
# should only execute this code once per line
fileline = fileLines[line].rstrip("\n\r")
try:
cluster_interval = GenomicInterval( g1, fileline.split("\t"),
g1.chrom_col,
g1.start_col,
g1.end_col,
g1.strand_col,
g1.default_strand,
g1.fix_strand )
except Exception as exc:
print(str( exc ), file=sys.stderr)
f1.close()
sys.exit()
interval_size = cluster_interval.end - cluster_interval.start
if outsize == -1 or \
( outsize > interval_size and output == 4 ) or \
( outsize < interval_size and output == 5 ):
outinterval = cluster_interval
outsize = interval_size
out_file.write( "%s\n" % outinterval )
f1.close()
out_file.close()
if g1.skipped > 0:
print(skipped( g1, filedesc="" ))
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True )
out_file = open( out_fname, "w" )
try:
for outfields in join(g1, g2, mincols=mincols, rightfill=rightfill, leftfill=leftfill):
if type( outfields ) is list:
out_file.write( "%s\n" % "\t".join( outfields ) )
else:
out_file.write( "%s\n" % outfields )
except ParseError, exc:
out_file.close()
fail( "Invalid file format: %s" % str( exc ) )
except MemoryError:
out_file.close()
fail( "Input datasets were too large to complete the join operation." )
out_file.close()
if g1.skipped > 0:
print skipped( g1, filedesc=" of 1st dataset" )
if g2.skipped > 0:
print skipped( g2, filedesc=" of 2nd dataset" )
if __name__ == "__main__":
main()
def main():
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg( options.cols2 )
in1_gff_format = bool( options.gff1 )
in2_gff_format = bool( options.gff2 )
in_fname, in2_fname, out_fname, direction = args
except:
doc_optparse.exception()
# Set readers to handle either GFF or default format.
if in1_gff_format:
in1_reader_wrapper = GFFIntervalToBEDReaderWrapper
else:
in1_reader_wrapper = NiceReaderWrapper
if in2_gff_format:
in2_reader_wrapper = GFFIntervalToBEDReaderWrapper
else:
in2_reader_wrapper = NiceReaderWrapper
g1 = in1_reader_wrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
g2 = in2_reader_wrapper( fileinput.FileInput( in2_fname ),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True )
# Find flanking features.
out_file = open( out_fname, "w" )
try:
for result in proximal_region_finder([g1, g2], direction):
if type( result ) is list:
line, closest_feature = result
# Need to join outputs differently depending on file types.
if in1_gff_format:
# Output is GFF with added attribute 'closest feature.'
# Invervals are in BED coordinates; need to convert to GFF.
line = convert_bed_coords_to_gff( line )
closest_feature = convert_bed_coords_to_gff( closest_feature )
# Replace double quotes with single quotes in closest feature's attributes.
out_file.write( "%s closest_feature \"%s\" \n" %
( "\t".join( line.fields ),
"\t".join( closest_feature.fields ).replace( "\"", "\\\"" )
) )
else:
# Output is BED + closest feature fields.
output_line_fields = []
output_line_fields.extend( line.fields )
output_line_fields.extend( closest_feature.fields )
out_file.write( "%s\n" % ( "\t".join( output_line_fields ) ) )
else:
out_file.write( "%s\n" % result )
except ParseError as exc:
fail( "Invalid file format: %s" % str( exc ) )
print("Direction: %s" % (direction))
if g1.skipped > 0:
print(skipped( g1, filedesc=" of 1st dataset" ))
if g2.skipped > 0:
print(skipped( g2, filedesc=" of 2nd dataset" ))
def main():
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg(
options.cols2)
in1_gff_format = bool(options.gff1)
in2_gff_format = bool(options.gff2)
in_fname, in2_fname, out_fname, direction = args
except:
doc_optparse.exception()
# Set readers to handle either GFF or default format.
if in1_gff_format:
in1_reader_wrapper = GFFIntervalToBEDReaderWrapper
else:
in1_reader_wrapper = NiceReaderWrapper
if in2_gff_format:
in2_reader_wrapper = GFFIntervalToBEDReaderWrapper
else:
in2_reader_wrapper = NiceReaderWrapper
g1 = in1_reader_wrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
g2 = in2_reader_wrapper(fileinput.FileInput(in2_fname),
chrom_col=chr_col_2,
start_col=start_col_2,
end_col=end_col_2,
strand_col=strand_col_2,
fix_strand=True)
# Find flanking features.
out_file = open(out_fname, "w")
try:
for result in proximal_region_finder([g1, g2], direction):
if type(result) is list:
line, closest_feature = result
# Need to join outputs differently depending on file types.
if in1_gff_format:
# Output is GFF with added attribute 'closest feature.'
# Invervals are in BED coordinates; need to convert to GFF.
line = convert_bed_coords_to_gff(line)
closest_feature = convert_bed_coords_to_gff(
closest_feature)
# Replace double quotes with single quotes in closest feature's attributes.
out_file.write(
"%s closest_feature \"%s\" \n" %
("\t".join(line.fields), "\t".join(
closest_feature.fields).replace("\"", "\\\"")))
else:
# Output is BED + closest feature fields.
output_line_fields = []
output_line_fields.extend(line.fields)
output_line_fields.extend(closest_feature.fields)
out_file.write("%s\n" % ("\t".join(output_line_fields)))
else:
out_file.write("%s\n" % result)
except ParseError as exc:
fail("Invalid file format: %s" % str(exc))
print("Direction: %s" % (direction))
if g1.skipped > 0:
print(skipped(g1, filedesc=" of 1st dataset"))
if g2.skipped > 0:
print(skipped(g2, filedesc=" of 2nd dataset"))
def main():
distance = 0
minregions = 2
output = 1
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
if options.distance:
distance = int(options.distance)
if options.overlap:
distance = -1 * int(options.overlap)
if options.output:
output = int(options.output)
if options.minregions:
minregions = int(options.minregions)
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
# Get the cluster tree
try:
clusters, extra = find_clusters(g1,
mincols=distance,
minregions=minregions)
except ParseError as exc:
fail("Invalid file format: %s" % str(exc))
f1 = open(in_fname, "r")
out_file = open(out_fname, "w")
# If "merge"
if output == 1:
fields = [
"."
for x in range(max(g1.chrom_col, g1.start_col, g1.end_col) + 1)
]
for chrom, tree in clusters.items():
for start, end, lines in tree.getregions():
fields[g1.chrom_col] = chrom
fields[g1.start_col] = str(start)
fields[g1.end_col] = str(end)
out_file.write("%s\n" % "\t".join(fields))
# If "filtered" we preserve order of file and comments, etc.
if output == 2:
linenums = dict()
for chrom, tree in clusters.items():
for linenum in tree.getlines():
linenums[linenum] = 0
linenum = -1
f1.seek(0)
for line in f1.readlines():
linenum += 1
if linenum in linenums or linenum in extra:
out_file.write("%s\n" % line.rstrip("\n\r"))
# If "clustered" we output original intervals, but near each other (i.e. clustered)
if output == 3:
linenums = list()
f1.seek(0)
fileLines = f1.readlines()
for chrom, tree in clusters.items():
for linenum in tree.getlines():
out_file.write("%s\n" % fileLines[linenum].rstrip("\n\r"))
# If "minimum" we output the smallest interval in each cluster
if output == 4 or output == 5:
linenums = list()
f1.seek(0)
fileLines = f1.readlines()
for chrom, tree in clusters.items():
for start, end, lines in tree.getregions():
outsize = -1
outinterval = None
for line in lines:
# three nested for loops?
# should only execute this code once per line
fileline = fileLines[line].rstrip("\n\r")
try:
cluster_interval = GenomicInterval(
g1, fileline.split("\t"), g1.chrom_col,
g1.start_col, g1.end_col, g1.strand_col,
g1.default_strand, g1.fix_strand)
except Exception as exc:
print(str(exc), file=sys.stderr)
f1.close()
sys.exit()
interval_size = cluster_interval.end - cluster_interval.start
if outsize == -1 or \
( outsize > interval_size and output == 4 ) or \
( outsize < interval_size and output == 5 ):
outinterval = cluster_interval
outsize = interval_size
out_file.write("%s\n" % outinterval)
f1.close()
out_file.close()
if g1.skipped > 0:
print(skipped(g1, filedesc=""))
def main():
allchroms = False
options, args = doc_optparse.parse(__doc__)
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg(
options.cols1)
lengths = options.lengths
if options.all:
allchroms = True
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper(fileinput.FileInput(in_fname),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True)
lens = dict()
chroms = list()
# dbfile is used to determine the length of each chromosome. The lengths
# are added to the lens dict and passed copmlement operation code in bx.
dbfile = fileinput.FileInput(lengths)
if dbfile:
if not allchroms:
try:
for line in dbfile:
fields = line.split("\t")
lens[fields[0]] = int(fields[1])
except:
# assume LEN doesn't exist or is corrupt somehow
pass
elif allchroms:
try:
for line in dbfile:
fields = line.split("\t")
end = int(fields[1])
chroms.append("\t".join([fields[0], "0", str(end)]))
except:
pass
# Safety...if the dbfile didn't exist and we're on allchroms, then
# default to generic complement
if allchroms and len(chroms) == 0:
allchroms = False
if allchroms:
chromReader = GenomicIntervalReader(chroms)
generator = subtract([chromReader, g1])
else:
generator = complement(g1, lens)
out_file = open(out_fname, "w")
try:
for interval in generator:
if type(interval) is GenomicInterval:
out_file.write("%s\n" % "\t".join(interval))
else:
out_file.write("%s\n" % interval)
except ParseError, exc:
out_file.close()
fail("Invalid file format: %s" % str(exc))
def main():
allchroms = False
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 )
lengths = options.lengths
if options.all:
allchroms = True
in_fname, out_fname = args
except:
doc_optparse.exception()
g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ),
chrom_col=chr_col_1,
start_col=start_col_1,
end_col=end_col_1,
strand_col=strand_col_1,
fix_strand=True )
lens = dict()
chroms = list()
# dbfile is used to determine the length of each chromosome. The lengths
# are added to the lens dict and passed copmlement operation code in bx.
dbfile = fileinput.FileInput( lengths )
if dbfile:
if not allchroms:
try:
for line in dbfile:
fields = line.split("\t")
lens[fields[0]] = int(fields[1])
except:
# assume LEN doesn't exist or is corrupt somehow
pass
elif allchroms:
try:
for line in dbfile:
fields = line.split("\t")
end = int(fields[1])
chroms.append("\t".join([fields[0], "0", str(end)]))
except:
pass
# Safety...if the dbfile didn't exist and we're on allchroms, then
# default to generic complement
if allchroms and len(chroms) == 0:
allchroms = False
if allchroms:
chromReader = GenomicIntervalReader(chroms)
generator = subtract([chromReader, g1])
else:
generator = complement(g1, lens)
out_file = open( out_fname, "w" )
try:
for interval in generator:
if type( interval ) is GenomicInterval:
out_file.write( "%s\n" % "\t".join( interval ) )
else:
out_file.write( "%s\n" % interval )
except ParseError as exc:
out_file.close()
fail( "Invalid file format: %s" % str( exc ) )
out_file.close()
if g1.skipped > 0:
print(skipped( g1, filedesc="" ))
out_file = open(out_fname, "w")
try:
for outfields in join(g1,
g2,
mincols=mincols,
rightfill=rightfill,
leftfill=leftfill):
if type(outfields) is list:
out_file.write("%s\n" % "\t".join(outfields))
else:
out_file.write("%s\n" % outfields)
except ParseError, exc:
out_file.close()
fail("Invalid file format: %s" % str(exc))
except MemoryError:
out_file.close()
fail("Input datasets were too large to complete the join operation.")
out_file.close()
if g1.skipped > 0:
print skipped(g1, filedesc=" of 1st dataset")
if g2.skipped > 0:
print skipped(g2, filedesc=" of 2nd dataset")
if __name__ == "__main__":
main()
