def __main__():
#
# Parse options, args.
#
options, args = doc_optparse.parse( __doc__ )
try:
if len(options.cols.split(',')) == 5:
# BED file
chrom_col, start_col, end_col, strand_col, name_col = parse_cols_arg( options.cols )
else:
# gff file
chrom_col, start_col, end_col, strand_col = parse_cols_arg( options.cols )
name_col = False
dbkey = options.dbkey
output_format = options.output_format
gff_format = options.gff
interpret_features = options.interpret_features
GALAXY_DATA_INDEX_DIR = options.GALAXY_DATA_INDEX_DIR
fasta_file = options.fasta
input_filename, output_filename = args
except:
doc_optparse.exception()
includes_strand_col = strand_col >= 0
strand = None
nibs = {}
#
# Set path to sequence data.
#
if fasta_file:
# Need to create 2bit file from fasta file.
try:
seq_path = tempfile.NamedTemporaryFile( dir="." ).name
cmd = "faToTwoBit %s %s" % ( fasta_file, seq_path )
tmp_name = tempfile.NamedTemporaryFile( dir="." ).name
tmp_stderr = open( tmp_name, 'wb' )
proc = subprocess.Popen( args=cmd, shell=True, stderr=tmp_stderr.fileno() )
returncode = proc.wait()
tmp_stderr.close()
# Get stderr, allowing for case where it's very large.
tmp_stderr = open( tmp_name, 'rb' )
stderr = ''
buffsize = 1048576
try:
while True:
stderr += tmp_stderr.read( buffsize )
if not stderr or len( stderr ) % buffsize != 0:
break
except OverflowError:
pass
tmp_stderr.close()
# Error checking.
if returncode != 0:
raise Exception(stderr)
except Exception, e:
stop_err( 'Error running faToTwoBit. ' + str( e ) )
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():
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
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():
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():
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():
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():
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 stop_err(msg):
sys.stderr.write(msg)
sys.exit()
if len(sys.argv) < 3:
stop_err("Incorrect number of arguments.")
inp_file = sys.argv[1]
out_file = sys.argv[2]
fout = open(out_file, 'w')
int_file = sys.argv[3]
if int_file != "None": #The user has specified an interval file
dbkey_i = sys.argv[4]
chr_col_i, start_col_i, end_col_i, strand_col_i = parse_cols_arg( sys.argv[5] )
def rateEstimator(block):
global alignlen, mismatches
src1 = block.components[0].src
sequence1 = block.components[0].text
start1 = block.components[0].start
end1 = block.components[0].end
len1 = int(end1)-int(start1)
len1_withgap = len(sequence1)
mismatch = 0.0
for seq in range (1, len(block.components)):
src2 = block.components[seq].src
def __main__():
#
# Parse options, args.
#
options, args = doc_optparse.parse(__doc__)
try:
if len(options.cols.split(',')) == 5:
# BED file
chrom_col, start_col, end_col, strand_col, name_col = parse_cols_arg(
options.cols)
else:
# gff file
chrom_col, start_col, end_col, strand_col = parse_cols_arg(
options.cols)
name_col = False
dbkey = options.dbkey
output_format = options.output_format
gff_format = options.gff
interpret_features = options.interpret_features
GALAXY_DATA_INDEX_DIR = options.GALAXY_DATA_INDEX_DIR
fasta_file = options.fasta
input_filename, output_filename = args
except:
doc_optparse.exception()
includes_strand_col = strand_col >= 0
strand = None
nibs = {}
#
# Set path to sequence data.
#
if fasta_file:
# Need to create 2bit file from fasta file.
try:
seq_path = tempfile.NamedTemporaryFile(dir=".").name
cmd = "faToTwoBit %s %s" % (fasta_file, seq_path)
tmp_name = tempfile.NamedTemporaryFile(dir=".").name
tmp_stderr = open(tmp_name, 'wb')
proc = subprocess.Popen(args=cmd,
shell=True,
stderr=tmp_stderr.fileno())
returncode = proc.wait()
tmp_stderr.close()
# Get stderr, allowing for case where it's very large.
tmp_stderr = open(tmp_name, 'rb')
stderr = ''
buffsize = 1048576
try:
while True:
stderr += tmp_stderr.read(buffsize)
if not stderr or len(stderr) % buffsize != 0:
break
except OverflowError:
pass
tmp_stderr.close()
# Error checking.
if returncode != 0:
raise Exception(stderr)
except Exception as e:
stop_err('Error running faToTwoBit. ' + str(e))
else:
seq_path = check_seq_file(dbkey, GALAXY_DATA_INDEX_DIR)
if not os.path.exists(seq_path):
# If this occurs, we need to fix the metadata validator.
stop_err(
"No sequences are available for '%s', request them by reporting this error."
% dbkey)
#
# Fetch sequences.
#
# Get feature's line(s).
def get_lines(feature):
if isinstance(feature, gff_util.GFFFeature):
return feature.lines()
else:
return [feature.rstrip('\r\n')]
skipped_lines = 0
first_invalid_line = 0
invalid_lines = []
fout = open(output_filename, "w")
warnings = []
warning = ''
twobitfile = None
file_iterator = open(input_filename)
if gff_format and interpret_features:
file_iterator = gff_util.GFFReaderWrapper(file_iterator,
fix_strand=False)
line_count = 1
for feature in file_iterator:
# Ignore comments, headers.
if isinstance(feature, (Header, Comment)):
line_count += 1
continue
name = ""
if gff_format and interpret_features:
# Processing features.
gff_util.convert_gff_coords_to_bed(feature)
chrom = feature.chrom
start = feature.start
end = feature.end
strand = feature.strand
else:
# Processing lines, either interval or GFF format.
line = feature.rstrip('\r\n')
if line and not line.startswith("#"):
fields = line.split('\t')
try:
chrom = fields[chrom_col]
start = int(fields[start_col])
end = int(fields[end_col])
if name_col:
name = fields[name_col]
if gff_format:
start, end = gff_util.convert_gff_coords_to_bed(
[start, end])
if includes_strand_col:
strand = fields[strand_col]
except:
warning = "Invalid chrom, start or end column values. "
warnings.append(warning)
if not invalid_lines:
invalid_lines = get_lines(feature)
first_invalid_line = line_count
skipped_lines += len(invalid_lines)
continue
if start > end:
warning = "Invalid interval, start '%d' > end '%d'. " % (
start, end)
warnings.append(warning)
if not invalid_lines:
invalid_lines = get_lines(feature)
first_invalid_line = line_count
skipped_lines += len(invalid_lines)
continue
if strand not in ['+', '-']:
strand = '+'
sequence = ''
else:
continue
# Open sequence file and get sequence for feature/interval.
if seq_path and os.path.exists("%s/%s.nib" % (seq_path, chrom)):
# TODO: improve support for GFF-nib interaction.
if chrom in nibs:
nib = nibs[chrom]
else:
nibs[chrom] = nib = bx.seq.nib.NibFile(
open("%s/%s.nib" % (seq_path, chrom)))
try:
sequence = nib.get(start, end - start)
except Exception as e:
warning = "Unable to fetch the sequence from '%d' to '%d' for build '%s'. " % (
start, end - start, dbkey)
warnings.append(warning)
if not invalid_lines:
invalid_lines = get_lines(feature)
first_invalid_line = line_count
skipped_lines += len(invalid_lines)
continue
elif seq_path and os.path.isfile(seq_path):
if not (twobitfile):
twobitfile = bx.seq.twobit.TwoBitFile(open(seq_path))
try:
if options.gff and interpret_features:
# Create sequence from intervals within a feature.
sequence = ''
for interval in feature.intervals:
sequence += twobitfile[
interval.chrom][interval.start:interval.end]
else:
sequence = twobitfile[chrom][start:end]
except:
warning = "Unable to fetch the sequence from '%d' to '%d' for chrom '%s'. " % (
start, end - start, chrom)
warnings.append(warning)
if not invalid_lines:
invalid_lines = get_lines(feature)
first_invalid_line = line_count
skipped_lines += len(invalid_lines)
continue
else:
warning = "Chromosome by name '%s' was not found for build '%s'. " % (
chrom, dbkey)
warnings.append(warning)
if not invalid_lines:
invalid_lines = get_lines(feature)
first_invalid_line = line_count
skipped_lines += len(invalid_lines)
continue
if sequence == '':
warning = "Chrom: '%s', start: '%s', end: '%s' is either invalid or not present in build '%s'. " % \
( chrom, start, end, dbkey )
warnings.append(warning)
if not invalid_lines:
invalid_lines = get_lines(feature)
first_invalid_line = line_count
skipped_lines += len(invalid_lines)
continue
if includes_strand_col and strand == "-":
sequence = reverse_complement(sequence)
if output_format == "fasta":
l = len(sequence)
c = 0
if gff_format:
start, end = gff_util.convert_bed_coords_to_gff([start, end])
fields = [dbkey, str(chrom), str(start), str(end), strand]
meta_data = "_".join(fields)
if name.strip():
fout.write(">%s %s\n" % (meta_data, name))
else:
fout.write(">%s\n" % meta_data)
while c < l:
b = min(c + 50, l)
fout.write("%s\n" % str(sequence[c:b]))
c = b
else: # output_format == "interval"
if gff_format and interpret_features:
# TODO: need better GFF Reader to capture all information needed
# to produce this line.
meta_data = "\t".join([
feature.chrom, "galaxy_extract_genomic_dna", "interval",
str(feature.start),
str(feature.end), feature.score, feature.strand, ".",
gff_util.gff_attributes_to_str(feature.attributes, "GTF")
])
else:
meta_data = "\t".join(fields)
if gff_format:
format_str = "%s seq \"%s\";\n"
else:
format_str = "%s\t%s\n"
fout.write(format_str % (meta_data, str(sequence)))
# Update line count.
if isinstance(feature, gff_util.GFFFeature):
line_count += len(feature.intervals)
else:
line_count += 1
fout.close()
if warnings:
warn_msg = "%d warnings, 1st is: " % len(warnings)
warn_msg += warnings[0]
print warn_msg
if skipped_lines:
# Error message includes up to the first 10 skipped lines.
print 'Skipped %d invalid lines, 1st is #%d, "%s"' % (
skipped_lines, first_invalid_line, '\n'.join(invalid_lines[:10]))
# Clean up temp file.
if fasta_file:
os.remove(seq_path)
os.remove(tmp_name)
def __main__():
#
# Parse options, args.
#
options, args = doc_optparse.parse( __doc__ )
try:
if len(options.cols.split(',')) == 5:
# BED file
chrom_col, start_col, end_col, strand_col, name_col = parse_cols_arg( options.cols )
else:
# gff file
chrom_col, start_col, end_col, strand_col = parse_cols_arg( options.cols )
name_col = False
dbkey = options.dbkey
output_format = options.output_format
gff_format = options.gff
interpret_features = options.interpret_features
GALAXY_DATA_INDEX_DIR = options.GALAXY_DATA_INDEX_DIR
fasta_file = options.fasta
input_filename, output_filename = args
except:
doc_optparse.exception()
includes_strand_col = strand_col >= 0
strand = None
nibs = {}
#
# Set path to sequence data.
#
if fasta_file:
# Need to create 2bit file from fasta file.
try:
seq_path = tempfile.NamedTemporaryFile( dir="." ).name
cmd = "faToTwoBit %s %s" % ( fasta_file, seq_path )
tmp_name = tempfile.NamedTemporaryFile( dir="." ).name
tmp_stderr = open( tmp_name, 'wb' )
proc = subprocess.Popen( args=cmd, shell=True, stderr=tmp_stderr.fileno() )
returncode = proc.wait()
tmp_stderr.close()
# Get stderr, allowing for case where it's very large.
tmp_stderr = open( tmp_name, 'rb' )
stderr = ''
buffsize = 1048576
try:
while True:
stderr += tmp_stderr.read( buffsize )
if not stderr or len( stderr ) % buffsize != 0:
break
except OverflowError:
pass
tmp_stderr.close()
# Error checking.
if returncode != 0:
raise Exception(stderr)
except Exception as e:
stop_err( 'Error running faToTwoBit. ' + str( e ) )
else:
seq_path = check_seq_file( dbkey, GALAXY_DATA_INDEX_DIR )
if not os.path.exists( seq_path ):
# If this occurs, we need to fix the metadata validator.
stop_err( "No sequences are available for '%s', request them by reporting this error." % dbkey )
#
# Fetch sequences.
#
# Get feature's line(s).
def get_lines( feature ):
if isinstance( feature, gff_util.GFFFeature ):
return feature.lines()
else:
return [ feature.rstrip( '\r\n' ) ]
skipped_lines = 0
first_invalid_line = 0
invalid_lines = []
fout = open( output_filename, "w" )
warnings = []
warning = ''
twobitfile = None
file_iterator = open( input_filename )
if gff_format and interpret_features:
file_iterator = gff_util.GFFReaderWrapper( file_iterator, fix_strand=False )
line_count = 1
for feature in file_iterator:
# Ignore comments, headers.
if isinstance( feature, ( Header, Comment ) ):
line_count += 1
continue
name = ""
if gff_format and interpret_features:
# Processing features.
gff_util.convert_gff_coords_to_bed( feature )
chrom = feature.chrom
start = feature.start
end = feature.end
strand = feature.strand
else:
# Processing lines, either interval or GFF format.
line = feature.rstrip( '\r\n' )
if line and not line.startswith( "#" ):
fields = line.split( '\t' )
try:
chrom = fields[chrom_col]
start = int( fields[start_col] )
end = int( fields[end_col] )
if name_col:
name = fields[name_col]
if gff_format:
start, end = gff_util.convert_gff_coords_to_bed( [start, end] )
if includes_strand_col:
strand = fields[strand_col]
except:
warning = "Invalid chrom, start or end column values. "
warnings.append( warning )
if not invalid_lines:
invalid_lines = get_lines( feature )
first_invalid_line = line_count
skipped_lines += len( invalid_lines )
continue
if start > end:
warning = "Invalid interval, start '%d' > end '%d'. " % ( start, end )
warnings.append( warning )
if not invalid_lines:
invalid_lines = get_lines( feature )
first_invalid_line = line_count
skipped_lines += len( invalid_lines )
continue
if strand not in ['+', '-']:
strand = '+'
sequence = ''
else:
continue
# Open sequence file and get sequence for feature/interval.
if seq_path and os.path.exists( "%s/%s.nib" % ( seq_path, chrom ) ):
# TODO: improve support for GFF-nib interaction.
if chrom in nibs:
nib = nibs[chrom]
else:
nibs[chrom] = nib = bx.seq.nib.NibFile( open( "%s/%s.nib" % ( seq_path, chrom ) ) )
try:
sequence = nib.get( start, end - start )
except Exception as e:
warning = "Unable to fetch the sequence from '%d' to '%d' for build '%s'. " % ( start, end - start, dbkey )
warnings.append( warning )
if not invalid_lines:
invalid_lines = get_lines( feature )
first_invalid_line = line_count
skipped_lines += len( invalid_lines )
continue
elif seq_path and os.path.isfile( seq_path ):
if not(twobitfile):
twobitfile = bx.seq.twobit.TwoBitFile( open( seq_path ) )
try:
if options.gff and interpret_features:
# Create sequence from intervals within a feature.
sequence = ''
for interval in feature.intervals:
sequence += twobitfile[interval.chrom][interval.start:interval.end]
else:
sequence = twobitfile[chrom][start:end]
except:
warning = "Unable to fetch the sequence from '%d' to '%d' for chrom '%s'. " % ( start, end - start, chrom )
warnings.append( warning )
if not invalid_lines:
invalid_lines = get_lines( feature )
first_invalid_line = line_count
skipped_lines += len( invalid_lines )
continue
else:
warning = "Chromosome by name '%s' was not found for build '%s'. " % ( chrom, dbkey )
warnings.append( warning )
if not invalid_lines:
invalid_lines = get_lines( feature )
first_invalid_line = line_count
skipped_lines += len( invalid_lines )
continue
if sequence == '':
warning = "Chrom: '%s', start: '%s', end: '%s' is either invalid or not present in build '%s'. " % \
( chrom, start, end, dbkey )
warnings.append( warning )
if not invalid_lines:
invalid_lines = get_lines( feature )
first_invalid_line = line_count
skipped_lines += len( invalid_lines )
continue
if includes_strand_col and strand == "-":
sequence = reverse_complement( sequence )
if output_format == "fasta":
l = len( sequence )
c = 0
if gff_format:
start, end = gff_util.convert_bed_coords_to_gff( [ start, end ] )
fields = [dbkey, str( chrom ), str( start ), str( end ), strand]
meta_data = "_".join( fields )
if name.strip():
fout.write( ">%s %s\n" % (meta_data, name) )
else:
fout.write( ">%s\n" % meta_data )
while c < l:
b = min( c + 50, l )
fout.write( "%s\n" % str( sequence[c:b] ) )
c = b
else: # output_format == "interval"
if gff_format and interpret_features:
# TODO: need better GFF Reader to capture all information needed
# to produce this line.
meta_data = "\t".join(
[feature.chrom, "galaxy_extract_genomic_dna", "interval",
str( feature.start ), str( feature.end ), feature.score, feature.strand,
".", gff_util.gff_attributes_to_str( feature.attributes, "GTF" ) ] )
else:
meta_data = "\t".join( fields )
if gff_format:
format_str = "%s seq \"%s\";\n"
else:
format_str = "%s\t%s\n"
fout.write( format_str % ( meta_data, str( sequence ) ) )
# Update line count.
if isinstance( feature, gff_util.GFFFeature ):
line_count += len( feature.intervals )
else:
line_count += 1
fout.close()
if warnings:
warn_msg = "%d warnings, 1st is: " % len( warnings )
warn_msg += warnings[0]
print(warn_msg)
if skipped_lines:
# Error message includes up to the first 10 skipped lines.
print('Skipped %d invalid lines, 1st is #%d, "%s"' % ( skipped_lines, first_invalid_line, '\n'.join( invalid_lines[:10] ) ))
# Clean up temp file.
if fasta_file:
os.remove( seq_path )
os.remove( tmp_name )
def main():
try:
if int( sys.argv[3] ) < 0:
raise Exception
except:
stop_err( "Length of flanking region(s) must be a non-negative integer." )
# Parsing Command Line here
options, args = doc_optparse.parse( __doc__ )
try:
chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols )
inp_file, out_file, size, direction, region = args
if strand_col_1 <= 0:
strand = "+" # if strand is not defined, default it to +
except:
stop_err( "Metadata issue, correct the metadata attributes by clicking on the pencil icon in the history item." )
try:
offset = int(options.off)
size = int(size)
except:
stop_err( "Invalid offset or length entered. Try again by entering valid integer values." )
fo = open(out_file, 'w')
skipped_lines = 0
first_invalid_line = 0
invalid_line = None
elems = []
j = 0
for i, line in enumerate( file( inp_file ) ):
line = line.strip()
if line and (not line.startswith( '#' )) and line != '':
j += 1
try:
elems = line.split('\t')
#if the start and/or end columns are not numbers, skip that line.
assert int(elems[start_col_1])
assert int(elems[end_col_1])
if strand_col_1 != -1:
strand = elems[strand_col_1]
#if the stand value is not + or -, skip that line.
assert strand in ['+', '-']
if direction == 'Upstream':
if strand == '+':
if region == 'end':
elems[end_col_1] = str(int(elems[end_col_1]) + offset)
elems[start_col_1] = str( int(elems[end_col_1]) - size )
else:
elems[end_col_1] = str(int(elems[start_col_1]) + offset)
elems[start_col_1] = str( int(elems[end_col_1]) - size )
elif strand == '-':
if region == 'end':
elems[start_col_1] = str(int(elems[start_col_1]) - offset)
elems[end_col_1] = str(int(elems[start_col_1]) + size)
else:
elems[start_col_1] = str(int(elems[end_col_1]) - offset)
elems[end_col_1] = str(int(elems[start_col_1]) + size)
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elif direction == 'Downstream':
if strand == '-':
if region == 'start':
elems[end_col_1] = str(int(elems[end_col_1]) - offset)
elems[start_col_1] = str( int(elems[end_col_1]) - size )
else:
elems[end_col_1] = str(int(elems[start_col_1]) - offset)
elems[start_col_1] = str( int(elems[end_col_1]) - size )
elif strand == '+':
if region == 'start':
elems[start_col_1] = str(int(elems[start_col_1]) + offset)
elems[end_col_1] = str(int(elems[start_col_1]) + size)
else:
elems[start_col_1] = str(int(elems[end_col_1]) + offset)
elems[end_col_1] = str(int(elems[start_col_1]) + size)
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elif direction == 'Both':
if strand == '-':
if region == 'start':
start = str(int(elems[end_col_1]) - offset)
end1 = str(int(start) + size)
end2 = str(int(start) - size)
elems[start_col_1] = start
elems[end_col_1] = end1
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elems[start_col_1] = end2
elems[end_col_1] = start
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elif region == 'end':
start = str(int(elems[start_col_1]) - offset)
end1 = str(int(start) + size)
end2 = str(int(start) - size)
elems[start_col_1] = start
elems[end_col_1] = end1
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elems[start_col_1] = end2
elems[end_col_1] = start
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
else:
start1 = str(int(elems[end_col_1]) - offset)
end1 = str(int(start1) + size)
start2 = str(int(elems[start_col_1]) - offset)
end2 = str(int(start2) - size)
elems[start_col_1] = start1
elems[end_col_1] = end1
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elems[start_col_1] = end2
elems[end_col_1] = start2
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elif strand == '+':
if region == 'start':
start = str(int(elems[start_col_1]) + offset)
end1 = str(int(start) - size)
end2 = str(int(start) + size)
elems[start_col_1] = end1
elems[end_col_1] = start
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elems[start_col_1] = start
elems[end_col_1] = end2
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elif region == 'end':
start = str(int(elems[end_col_1]) + offset)
end1 = str(int(start) - size)
end2 = str(int(start) + size)
elems[start_col_1] = end1
elems[end_col_1] = start
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elems[start_col_1] = start
elems[end_col_1] = end2
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
else:
start1 = str(int(elems[start_col_1]) + offset)
end1 = str(int(start1) - size)
start2 = str(int(elems[end_col_1]) + offset)
end2 = str(int(start2) + size)
elems[start_col_1] = end1
elems[end_col_1] = start1
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
elems[start_col_1] = start2
elems[end_col_1] = end2
assert int(elems[start_col_1]) > 0 and int(elems[end_col_1]) > 0
fo.write( "%s\n" % '\t'.join( elems ) )
except:
skipped_lines += 1
if not invalid_line:
first_invalid_line = i + 1
invalid_line = line
fo.close()
if skipped_lines == j:
stop_err( "Data issue: click the pencil icon in the history item to correct the metadata attributes." )
if skipped_lines > 0:
print 'Skipped %d invalid lines starting with #%dL "%s"' % ( skipped_lines, first_invalid_line, invalid_line )
print 'Location: %s, Region: %s, Flank-length: %d, Offset: %d ' % ( direction, region, size, offset )
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():
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():
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="" ))
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=""))
