def test_does_not_readd_headers(self):
# Make sure headers that exist are not duplicated
from ngs_mapper.tagreads import get_bam_header
from ngs_mapper import samtools
self.temp_copy_files()
# Create a new bam file that the header already has a read group in it
# as well as a new header
hdr = get_bam_header(self.bam)
hdr += '\n' + '@RG\tID:Test\tCN:cn\tSM:sm\tPL:ILLUMINA\n'
hdr += '@RG\tID:Roche454\tSM:312\tPL:L454\n'
hdr += '@RG\tID:IonTorrent\tSM:312\tPL:IONTORRENT\n'
hdr += '@RG\tID:MiSeq\tSM:312\tPL:ILLUMINA\n'
hdr += '@RG\tID:Sanger\tSM:312\tPL:CAPILLARY\n'
# Read the pipe which should be sam input and output bam
s = samtools.view(self.bam)
# Put in new header and sam output after it
with open('t.sam', 'w') as fh:
# New header
fh.write(hdr)
# Then reads
fh.write(s.read())
# Now convert that pipe to bam
b = samtools.view('t.sam', h=True, S=True, b=True)
# Write the bam output
with open('hasrg.bam', 'wb') as bamfh:
bamfh.write(b.read())
# Close the file handles
b.close()
s.close()
# Now we have a bamfile with an existing header that we can test
r = self._C('hasrg.bam', 'sm', 'cn')
# Make sure that the new header made it in and that the MiSeq header was not duplicated
header_lines = r.splitlines(True)
read_groups = [rg for rg in header_lines if rg.startswith('@RG')]
num_miseq = 0
num_test = 0
for rg in read_groups:
if 'ID:Test\t' in rg:
num_test += 1
if 'ID:MiSeq\t' in rg:
num_miseq += 1
eq_(1, num_miseq, "Header was duplicated which is incorrect")
eq_(1, num_test, "Existing header was removed somehow")
# How many platform readgroups to expect(includes the MiSeq one that we are testing)
i = len(self.read_group_ids)
# Now increase that by HD, SQ and Test RG
i += 3
eq_(i, len(header_lines), "Incorrect number of header lines")
def test_header_correct(self):
hdr = self._C(self.bam)
with open('t.sam', 'w') as fh:
fh.write(hdr)
from ngs_mapper import samtools
h = samtools.view('t.sam', S=True, H=True)
eq_(hdr, h.read().rstrip())
def test_seqencingcenter_argument(self):
from ngs_mapper import samtools
self.temp_copy_files()
self._C([self.bam], ['-CN', 'seqcenter'])
s = samtools.view(self.bam, H=True)
rgs = s.readlines()
s.close()
# Ensure each read group contains the samplename set
for rg in [r for r in rgs if r.startswith('@RG')]:
ok_('CN:seqcenter\t' in rg,
"Sequencing center did not make it into the headers")
def count_rg(self, bam):
''' Count how many of each uniq read group id '''
from ngs_mapper import samtools
s = samtools.view(bam)
counts = {}
for read in s:
aread = samtools.SamRow(read)
tags = dict(aread.TAGS)
id = tags['RG']
if id not in counts:
counts[id] = 0
counts[id] += 1
return counts
def test_does_multiple_bams(self):
from ngs_mapper import samtools
self.temp_copy_files()
bam2 = join(self.tempdir, 'sample2.bam')
bai2 = join(self.tempdir, 'sample2.bam.bai')
shutil.copy(self.bam, bam2)
shutil.copy(self.bai, bai2)
self._C([self.bam, bam2])
for b in [self.bam, bam2]:
self.check_tagreadcounts(b)
n = basename(b).replace('.bam', '')
s = samtools.view(b, H=True)
rg = [
header.split('\t') for header in s if header.startswith('@RG')
]
for rgline in rg:
eq_(
'SM:' + n, rgline[2],
"Did not set {0} as SM for {1}. Header: {2}".format(
n, b, rgline))