def test_run_process(self):
filepath = "{0}/tests/data/test_file.txt".format(MODULE_DIR)
remove_file(filepath)
# simple test to create file
args = ["touch", filepath]
run_process(args)
self.assertEqual(os.path.isfile(filepath), True)
def intersect_with_bed(input_bam, input_bed, output_file, force_strand=True):
"""
Given a bam file, intersect with a bed file to leave only thse reads that
correspond to entries in the bed file
Parameters
---------
input_bam : str
Path to the .bam file containing the reads
input_bed : str
Path to the .bed file containing the intervals
output_file : str
Path to the output file
force_strand : bool
If set, ensure the reads map to the same strand
Examples
---------
>>> from bioUtilities.bam import intersect_with_bed
>>> intersect_with_bed("reads.bam", "exons.bed", "exon_reads.bam")
"""
# now use bedtools to count the reads
args = ["bedtools", "intersect", "-s", "-abam", input_bam, "-b", input_bed]
if not force_strand:
del args[2]
run_process(args, file_for_output=output_file)
def nm_filter(input_file, output_file, lower_limit = 0, upper_limit = 10):
"""
Filter a .bam/.sam by NM value
Parameters
---------
input_file : str
Path to the file to be filtered
output_file : str
Path to the output
lower_limit : int
If set, the lower boundary of NM values, for which all reads have to
be greater than or equal to
upper_limit : int
If set, the upper boundary of NM values, for which all reads have to
be less than or equal to
Examples
---------
>>> from bioUtilities.bam import nm_filter
>>> nm_filter("test.bam", "test_output.bam", lower_limit = 2)
>>> nm_filter("test.bam", "test_output.bam", upper_limit = 6)
>>> nm_filter("test.bam", "test_output.bam", lower_limit = 2, upper_limit = 6)
"""
# if neither thresholds are specified
if not lower_limit and not upper_limit:
raise Exception("\nERROR: You must specify at least one of the lower_limit or upper_limit thresholds.\n")
if not lower_limit:
print("Using the default lower limit of 0.")
if not upper_limit:
print("Using the default uppper limit of 10.")
#create output file
if output_file[-4:] == ".bam":
temp_output_file = "{0}.sam".format(output_file[:-4])
else:
temp_output_file = output_file
grep_args = ["^@"]
for i in range(lower_limit, upper_limit + 1):
grep_args.append("\|\tNM:i:{0}\t".format(i))
grep_args = "".join(grep_args)
# read in the file
file_reads = run_process(["samtools", "view", "-h", input_file])
# now run the grep args
output = run_process(["grep", grep_args], input_to_pipe = file_reads, file_for_output = temp_output_file)
# if the output file is in bam format
if output_file != temp_output_file:
samtools_args = ["samtools", "view", "-bh", temp_output_file]
run_process(samtools_args, file_for_output = output_file)
remove_file(temp_output_file)
def test_mapq_filter_lower_limit(self):
input_file = "{0}/tests/data/input.bam".format(MODULE_DIR)
expected_file = "{0}/tests/data/expected_mapq_filter_1.sam".format(MODULE_DIR)
observed_file = "{0}/tests/data/observed_mapq_filter_1.bam".format(MODULE_DIR)
mapq_filter(input_file, observed_file, lower_limit = 200)
expected = read_many_fields(expected_file, "\t")
# convert bam to sam to check correct output
# use samtools to extract in the same format as sam
temp_observed = "{0}/tests/data/observed_mapq_filter_1.sam".format(MODULE_DIR)
samtools_args = ["samtools", "view", observed_file]
run_process(samtools_args, file_for_output = temp_observed)
observed = read_many_fields(temp_observed, "\t")
self.assertEqual(expected, observed)
remove_file(temp_observed)
remove_file(observed_file)
def read_count(input_file):
"""
Get the number of reads from a .bam/.sam file
Parameters
---------
input_file : str
Path to the file to be counted
Returns
---------
read_count : int
The number of reads in the specified file
Examples
---------
>>> from bioUtilities.bam import read_count
>>> reads = read_count("test.bam")
>>> print(reads)
>>> 15
"""
raw_read_count = run_process(["samtools", "view", "-c", input_file])
read_count = int(re.findall("(\d+)", raw_read_count)[0])
return read_count
def test_xt_filter(self):
input_file = "{0}/tests/data/input.bam".format(MODULE_DIR)
expected_file = "{0}/tests/data/expected_xt_filter.sam".format(
MODULE_DIR)
observed_file = "{0}/tests/data/observed_xt_filter.bam".format(
MODULE_DIR)
xt_filter(input_file, observed_file, filter="XT:A:U")
#convert bam to sam to check correct output
temp_observed = "{0}/tests/data/observed_xt_filter.sam".format(
MODULE_DIR)
samtools_args = ["samtools", "view", observed_file]
run_process(samtools_args, file_for_output=temp_observed)
expected = read_many_fields(expected_file, "\t")
observed = read_many_fields(temp_observed, "\t")
self.assertEqual(expected, observed)
remove_file(temp_observed)
remove_file(observed_file)
def xt_filter(input_file, output_file, filter=None):
"""
Filter a .bam/.sam file by the XT tag
Parameters
---------
input_file : str
Path to the file to be filtered
output_file : str
Path to the output
filter : str
Filter than reads should contain
Examples
---------
>>> from bioUtilities.bam import xt_filter
>>> xt_filter("test.bam", "test_xt_filtered.bam", filter = "XT:A:U")
"""
if not xt_filter:
raise Exception('\nXT filter not specified.\n')
# if the output format is .bam, temporarily create .sam output file
if output_file[-4:] == ".bam":
temp_output_file = "{0}.sam".format(output_file[:-4])
else:
temp_output_file = output_file
# get the header of the file
sam_output = run_process(["samtools", "view", "-h", input_file])
grep_args = []
# get header lines
grep_args.append("^@")
# get XT values matching the filter
grep_args.append("\|\t{0}\t".format(filter))
grep_args = "".join(grep_args)
# run the filter
run_process(["grep", grep_args],
input_to_pipe=sam_output,
file_for_output=temp_output_file)
# if wanting to create bam, create bam and delete sam
if output_file != temp_output_file:
samtools_args = ["samtools", "view", "-bh", temp_output_file]
run_process(samtools_args, file_for_output=output_file)
remove_file(temp_output_file)
def test_intersect_with_bed(self):
input_bed = "{0}/tests/data/input.bed".format(MODULE_DIR)
input_bam = "{0}/tests/data/input3.bam".format(MODULE_DIR)
observed_file = "{0}/tests/data/observed_intersect_with_bed.bam".format(
MODULE_DIR)
expected_file = "{0}/tests/data/expected_intersect_with_bed.sam".format(
MODULE_DIR)
remove_file(observed_file)
intersect_with_bed(input_bam, input_bed, observed_file)
observed_sam = "{0}/tests/data/observed_intersect_with_bed.sam".format(
MODULE_DIR)
args = ["samtools", "view", "-h", observed_file]
run_process(args, file_for_output=observed_sam)
observed = read_many_fields(observed_sam)
expected = read_many_fields(expected_file)
self.assertEqual(expected, observed)
remove_file(observed_file)
remove_file(observed_sam)
def line_count(filepath):
"""
Count the number of lines in a file
Parameters
---------
filepath : str
Path to the file
Returns
---------
line_count : int
The number of lines in the file
Examples
---------
>>> from bioUtilities.files import line_count
>>> line_count("test_file.txt")
3
"""
count_output = run_process(["wc", "-l", filepath])
line_count = int(re.findall("\s+(\d+)", count_output)[0])
return line_count
def mapq_filter(input_file, output_file, lower_limit=None, upper_limit=None):
"""
Filter a .bam/.sam by MAPQ value
Parameters
---------
input_file : str
Path to the file to be filtered
output_file : str
Path to the output
lower_limit : int
If set, the lower boundary of mapq values, for which all reads have to
be greater than or equal to
upper_limit : int
If set, the upper boundary of mapq values, for which all reads have to
be less than or equal to
Examples
---------
>>> from bioUtilities.bam import xt_filter
>>> mapq_filter("test.bam", "test_output.bam", lower_limit = 100)
>>> mapq_filter("test.bam", "test_output.bam", upper_limit = 250)
>>> mapq_filter("test.bam", "test_output.bam", lower_limit = 100, upper_limit = 250)
"""
#if neither thresholds are specified
if not lower_limit and not upper_limit:
raise Exception(
"ERROR: You must specify at least one of the lower_limit or upper_limit thresholds."
)
samtools_args = ["samtools", "view", "-h"]
# if both thresholds are specified, we want the reads with values between these
if lower_limit and upper_limit:
# create temp file
temp_directory = "temp_mapq_filter.{0}".format(random.random())
create_directory(temp_directory)
temp_file = "{0}/{1}".format(temp_directory,
output_file.split("/")[-1])
# first get everything above the lower limit
temp_args = samtools_args.copy()
temp_args.extend(["-q", lower_limit, input_file])
run_process(temp_args, file_for_output=temp_file)
# now get everything below the upper limit. need to account for
# samtools removing everything below threshold
# so when inversing need to add 1 to total
temp_args = samtools_args.copy()
upper_limit = upper_limit + 1
temp_args.extend(["-q", upper_limit, temp_file, "-U", output_file])
run_process(temp_args)
# cleanup the temp files
remove_directory(temp_directory)
# if only a lower limit is specified
elif lower_limit and not upper_limit:
samtools_args.extend(["-bq", lower_limit, input_file])
run_process(samtools_args, file_for_output=output_file)
#if only the upper threshold is specified
elif upper_limit and not lower_limit:
# account for inverse by adding 1
upper_limit = upper_limit + 1
samtools_args.extend(
["-q", upper_limit, input_file, "-U", output_file])
run_process(samtools_args)
def count_interval_reads(input_file,
input_bam,
output_file,
paired_end=False,
min_qual=None,
min_length=50):
"""
For each interval in bed format, count the number of reads in the bam file
Parameters
---------
input_file : str
Path to the file containing the intervals
input_bam : str
Path to the .bam file containing the reads
output_file : str
Path to the output file
Dependencies
---------
featureCounts v1.6.4
Examples
---------
>>> from bioUtilities.bam import count_interval_reads
>>> count_interval_reads("exon_junctions.bed", "reads.bam", "exon_junction_reads.bed")
"""
# check that featureCounts command exists
if not shutil.which('featureCounts'):
raise Exception('\nERROR: featureCounts must be installed.\n')
# if input_file is in bed format, need to convert to .saf format
# .saf format its 1-based
if get_extension(input_file) == ".bed":
base_input_file = input_file
working_input_file = "{0}.saf".format(input_file[:-4])
bed_to_saf(old_input_file, input_file)
else:
working_input_file = input_file
if get_extension(output_file) == ".bed":
working_output_file = "{0}.saf".format(output_file[:-4])
else:
working_output_file = output_file
# now can use featureCounts to count reads
# this return the file in 'saf' format
args = ["featureCounts", "-fO", "-F", "SAF", "-g", "ID"]
if paired_end:
args.append("-p")
if min_qual:
args.extend(["-Q", min_qual])
if min_length:
args.extend(["-d", min_length])
args.extend(
["-a", working_input_file, "-o", working_output_file, input_bam])
# now run the count
run_process(args)
# if the output format is bed, convert the saf output to bed
if get_extension(output_file) == ".bed":
entries = read_many_fields(working_output_file)[2:]
with open(output_file, "w") as outfile:
for entry in entries:
output = [
entry[1],
str(int(entry[2]) - 1),
str(int(entry[3]) - 1), entry[0], ".", entry[4]
]
output.extend(entry[5:])
outfile.write("{0}\n".format("\t".join(output)))
# now clean up the files
if working_input_file != input_file:
remove_file(working_input_file)
if working_output_file != output_file:
remove_file(output_file)
