def set_sam_tag(cls, count_tag, bamfile_name, tag_regex):
"""Add key:value pair to class variable: has_sam_tag.
Keyword Arguments:
count_tag -- boolean on whether to count with this tag
bamfile_name -- file to query for tag
tag_regex -- regular expression for the tag (eg 'NA:i:(\d+)')
"""
(run_pipe_worked, sam_sample) = run_pipe(['samtools view {}'.format(bamfile_name), 'head -n 10'])
if run_pipe_worked:
return cls.process_set_sam_tag(sam_sample, count_tag, tag_regex)
else:
raise MetageneError("Checking the bam file failed with error: {}".format(sam_sample))
def set_chromosome_sizes(cls, bamfile):
"""Set chromosome_sizes dictionary with BAM header.
Keyword Arguments:
bamfile -- name of bamfile
"""
(run_pipe_worked, header) = run_pipe(["samtools view -H {}".format(bamfile)])
if not run_pipe_worked:
raise MetageneError("Could not open BAM file {}".format(bamfile))
else:
try:
return cls.extract_chromosome_sizes(header)
except MetageneError as err:
raise MetageneError("Error processing {} header\n{}".format(bamfile, err.message))
def set_chromosome_sizes(cls, bamfile):
"""Set chromosome_sizes dictionary with BAM header.
Keyword Arguments:
bamfile -- name of bamfile
"""
(run_pipe_worked,
header) = run_pipe(["samtools view -H {}".format(bamfile)])
if not run_pipe_worked:
raise MetageneError("Could not open BAM file {}".format(bamfile))
else:
try:
return cls.extract_chromosome_sizes(header)
except MetageneError as err:
raise MetageneError("Error processing {} header\n{}".format(
bamfile, err.message))
def set_sam_tag(cls, count_tag, bamfile_name, tag_regex):
"""Add key:value pair to class variable: has_sam_tag.
Keyword Arguments:
count_tag -- boolean on whether to count with this tag
bamfile_name -- file to query for tag
tag_regex -- regular expression for the tag (eg 'NA:i:(\d+)')
"""
(run_pipe_worked, sam_sample) = run_pipe(
['samtools view {}'.format(bamfile_name), 'head -n 10'])
if run_pipe_worked:
return cls.process_set_sam_tag(sam_sample, count_tag, tag_regex)
else:
raise MetageneError(
"Checking the bam file failed with error: {}".format(
sam_sample))
def metagene_count():
"""Chain of command for metagene_count analysis."""
arguments = get_arguments()
# confirm BAM file and extract chromosome sizes
Read.set_chromosome_sizes(arguments.alignment)
##TODO: create a list of chromosomes to analyze and/or exclude
# create chromosome conversion dictionary for feature (GFF/BED) to alignment (BAM)
Feature.set_chromosome_conversion(arguments.chromosome_names, Read.chromosome_sizes.keys())
# define has_abundance and has_mappings tags for Read class
Read.set_sam_tag(arguments.extract_abundance, arguments.alignment, "NA:i:(\d+)")
Read.set_sam_tag(arguments.extract_mappings, arguments.alignment, "NH:i:(\d+)")
# define the metagene array shape (left padding, start, internal, end, right padding)
# metagene = padding ---- internal region ---- padding
try:
metagene = Metagene(arguments.interval_size, arguments.padding, arguments.padding)
print "Metagene definition:\t{}".format(metagene)
except MetageneError as err:
print err
raise MetageneError("Unable to create the metagene template")
try:
Feature.set_format(arguments.feature) # assign file format for the feature file
print "Reading feature file as {} format".format(Feature.format)
except MetageneError as err:
print err
raise MetageneError("Unable to create the feature object")
# print out the header line...
if not arguments.interval_variable:
with open("{}.metagene_counts.csv".format(arguments.output_prefix), 'w') as output_file:
output_file.write("# Metagene:\t{}\n".format(metagene)) # define for plotting later
output_file.write(metagene.print_full())
# for each feature
with open(arguments.feature, 'r') as feature_file:
for feature_line in read_chunk(feature_file, 1024):
if feature_line[0] != "#": # skip comment lines
# change creation with feature_method
feature = Feature.create(arguments.feature_count, metagene, feature_line, arguments.count_splicing,
arguments.ignore_strand)
# pull out sam file lines; it is important to use Feature.get_samtools_region(chromosome_lengths) rather
# than Feature.get_chromosome_region() because only the first ensures that the interval does not
# extend beyond the length of the chromosome which makes samtools view return no reads
(run_pipe_worked, sam_sample) = run_pipe(['samtools view {} {}'.format(
arguments.alignment,
feature.get_samtools_region())])
if run_pipe_worked:
for samline in sam_sample:
if len(samline) > 0:
# create Read feature
(created_read, read) = Read.create_from_sam(samline,
Feature.chromosome_conversion.values(),
arguments.count_method,
arguments.uniquely_mapping,
arguments.ignore_strand,
arguments.count_secondary_alignments,
arguments.count_failed_quality_control,
arguments.count_PCR_optical_duplicate,
arguments.count_supplementary_alignment)
# count read (if it exists)
if created_read:
feature.count_read(read, arguments.count_method, arguments.count_splicing,
arguments.count_partial_reads, arguments.ignore_strand)
# output the resulting metagene
with open("{}.metagene_counts.csv".format(arguments.output_prefix), 'a') as output_file:
output_file.write(
"{}\n".format(feature.print_metagene(interval_override=arguments.interval_variable)))
else:
raise MetageneError("Could not pull chromosomal region {} for feature {} from BAM file {}.".format(
feature.get_chromosome_region(),
feature.name,
arguments.alignment))
def metagene_count():
"""Chain of command for metagene_count analysis."""
arguments = get_arguments()
# confirm BAM file and extract chromosome sizes
Read.set_chromosome_sizes(arguments.alignment)
##TODO: create a list of chromosomes to analyze and/or exclude
# create chromosome conversion dictionary for feature (GFF/BED) to alignment (BAM)
Feature.set_chromosome_conversion(arguments.chromosome_names,
Read.chromosome_sizes.keys())
# define has_abundance and has_mappings tags for Read class
Read.set_sam_tag(arguments.extract_abundance, arguments.alignment,
"NA:i:(\d+)")
Read.set_sam_tag(arguments.extract_mappings, arguments.alignment,
"NH:i:(\d+)")
# define the metagene array shape (left padding, start, internal, end, right padding)
# metagene = padding ---- internal region ---- padding
try:
metagene = Metagene(arguments.interval_size, arguments.padding,
arguments.padding)
print "Metagene definition:\t{}".format(metagene)
except MetageneError as err:
print err
raise MetageneError("Unable to create the metagene template")
try:
Feature.set_format(
arguments.feature) # assign file format for the feature file
print "Reading feature file as {} format".format(Feature.format)
except MetageneError as err:
print err
raise MetageneError("Unable to create the feature object")
# print out the header line...
if not arguments.interval_variable:
with open("{}.metagene_counts.csv".format(arguments.output_prefix),
'w') as output_file:
output_file.write("# Metagene:\t{}\n".format(
metagene)) # define for plotting later
output_file.write(metagene.print_full())
# for each feature
with open(arguments.feature, 'r') as feature_file:
for feature_line in read_chunk(feature_file, 1024):
if feature_line[0] != "#": # skip comment lines
# change creation with feature_method
feature = Feature.create(arguments.feature_count, metagene,
feature_line,
arguments.count_splicing,
arguments.ignore_strand)
# pull out sam file lines; it is important to use Feature.get_samtools_region(chromosome_lengths) rather
# than Feature.get_chromosome_region() because only the first ensures that the interval does not
# extend beyond the length of the chromosome which makes samtools view return no reads
(run_pipe_worked, sam_sample) = run_pipe([
'samtools view {} {}'.format(arguments.alignment,
feature.get_samtools_region())
])
if run_pipe_worked:
for samline in sam_sample:
if len(samline) > 0:
# create Read feature
(created_read, read) = Read.create_from_sam(
samline,
Feature.chromosome_conversion.values(),
arguments.count_method,
arguments.uniquely_mapping,
arguments.ignore_strand,
arguments.count_secondary_alignments,
arguments.count_failed_quality_control,
arguments.count_PCR_optical_duplicate,
arguments.count_supplementary_alignment)
# count read (if it exists)
if created_read:
feature.count_read(
read, arguments.count_method,
arguments.count_splicing,
arguments.count_partial_reads,
arguments.ignore_strand)
# output the resulting metagene
with open(
"{}.metagene_counts.csv".format(
arguments.output_prefix), 'a') as output_file:
output_file.write("{}\n".format(
feature.print_metagene(interval_override=arguments.
interval_variable)))
else:
raise MetageneError(
"Could not pull chromosomal region {} for feature {} from BAM file {}."
.format(feature.get_chromosome_region(), feature.name,
arguments.alignment))