def test_snv_annot_without_rg():
'It tests that we can do snv calling with a bam without rg info'
test_dir = NamedTemporaryDir()
project_name = 'backbone'
configuration = {'Snvs':{'default_bam_platform':'sanger'},
'General_settings':{'threads':THREADS}}
settings_path = create_project(directory=test_dir.name,
name=project_name,
configuration=configuration)
project_dir = join(test_dir.name, project_name)
#the reference
reference_dir = join(project_dir, 'mapping/reference')
os.makedirs(reference_dir)
reference_fpath = join(reference_dir, 'reference.fasta')
out = open(reference_fpath, 'w')
for line in open(join(TEST_DATA_DIR, 'blast/arabidopsis_genes')):
out.write(line)
bams_dir = join(project_dir, 'mapping', 'bams')
os.makedirs(bams_dir)
bam_fpath = join(bams_dir, 'merged.0.bam')
shutil.copy(join(TEST_DATA_DIR, 'merged.0.bam'), bam_fpath)
create_bam_index(bam_fpath)
annot_input_dir = join(project_dir, 'annotations', 'input')
os.makedirs(annot_input_dir)
os.symlink(reference_fpath, join(annot_input_dir, 'reference.fasta'))
do_analysis(project_settings=settings_path, kind='annotate_snvs', silent=True)
def run(self):
'It runs the analysis.'
inputs, output_dirs = self._get_inputs_and_prepare_outputs()
output_dir = output_dirs['result']
merged_bam = inputs['merged_bam']
create_bam_index(merged_bam.last_version)
pipeline = 'snv_bam_annotator'
bam_fpath = merged_bam.last_version
configuration = {'snv_bam_annotator': {'bam_fhand':bam_fpath}}
settings = self._project_settings
if 'Snvs' in settings:
snv_settings = settings['Snvs']
# read egde conf
read_edge_conf = self._configure_read_edge_conf(snv_settings)
configuration['snv_bam_annotator']['read_edge_conf'] = read_edge_conf
for config_param in ('min_quality', 'min_mapq', 'min_num_alleles',
'max_maf', 'min_num_reads_for_allele'):
if snv_settings[config_param] is not None:
param_value = float(snv_settings[config_param])
else:
param_value = None
configuration['snv_bam_annotator'][config_param] = param_value
if 'default_bam_platform' in snv_settings:
configuration['snv_bam_annotator']['default_bam_platform'] = \
snv_settings['default_bam_platform']
return self._run_annotation(pipeline=pipeline,
configuration=configuration,
inputs=inputs,
output_dir=output_dir)
def create_snv_annotator(bam_fhand, min_quality=45, default_sanger_quality=25,
min_mapq=15, min_num_alleles=1, max_maf=None,
read_edge_conf=None, default_bam_platform=None,
min_num_reads_for_allele=None, ploidy=2):
'It creates an annotator capable of annotating the snvs in a SeqRecord'
#the bam should have an index, does the index exists?
bam_fhand = get_fhand(bam_fhand)
create_bam_index(bam_fpath=bam_fhand.name)
read_edge_conf = _normalize_read_edge_conf(read_edge_conf)
bam = pysam.Samfile(bam_fhand.name, 'rb')
# default min num_reads per allele and ploidy
if min_num_reads_for_allele is None:
min_num_reads_for_allele = DEFAUL_MIN_NUM_READS_PER_ALLELE
if ploidy is None:
ploidy = DEFAULT_PLOIDY
def annotate_snps(sequence):
'It annotates the snvs found in the sequence'
for snv in _snvs_in_bam(bam, reference=sequence,
min_quality=min_quality,
default_sanger_quality=default_sanger_quality,
min_mapq=min_mapq,
min_num_alleles=min_num_alleles,
max_maf=max_maf,
read_edge_conf=read_edge_conf,
default_bam_platform=default_bam_platform,
min_num_reads_for_allele=min_num_reads_for_allele):
snv = _summarize_snv(snv)
location = snv['ref_position']
type_ = 'snv'
qualifiers = {'alleles':snv['alleles'],
'reference_allele':snv['reference_allele'],
'read_groups':snv['read_groups'],
'mapping_quality': snv['mapping_quality'],
'quality': snv['quality']}
snv_feat = SeqFeature(location=FeatureLocation(location, location),
type=type_,
qualifiers=qualifiers)
annotate_pic(snv_feat)
annotate_heterozygosity(snv_feat, ploidy=ploidy)
sequence.features.append(snv_feat)
return sequence
return annotate_snps
def run(self):
'''It runs the analysis.'''
self._log({'analysis_started':True})
inputs = self._get_input_fpaths()
bam_path = inputs['bam']
bam_fpath = bam_path.last_version
reference_fpath = inputs['reference'].last_version
out_fhand = open(bam_path.next_version, 'w')
cmd = ['samtools', 'calmd', '-Abr', bam_fpath, reference_fpath]
call(cmd, raise_on_error=True, stdout=out_fhand)
create_bam_index(out_fhand.name)
out_fhand.close()
self._log({'analysis_finished':True})
def run(self):
'''It runs the analysis.'''
self._log({'analysis_started':True})
settings = self._project_settings
project_path = settings['General_settings']['project_path']
tmp_dir = settings['General_settings']['tmpdir']
inputs = self._get_input_fpaths()
bam_paths = inputs['bams']
reference_path = inputs['reference']
output_dir = self._create_output_dirs()['result']
merged_bam_path = VersionedPath(os.path.join(output_dir,
BACKBONE_BASENAMES['merged_bam']))
merged_bam_fpath = merged_bam_path.next_version
#Do we have to add the default qualities to the sam file?
#do we have characters different from ACTGN?
add_qualities = settings['Sam_processing']['add_default_qualities']
#memory for the java programs
java_mem = settings['Other_settings']['java_memory']
picard_path = settings['Other_settings']['picard_path']
if add_qualities:
default_sanger_quality = settings['Other_settings']['default_sanger_quality']
default_sanger_quality = int(default_sanger_quality)
else:
default_sanger_quality = None
temp_dir = NamedTemporaryDir()
for bam_path in bam_paths:
bam_basename = bam_path.basename
temp_sam = NamedTemporaryFile(prefix='%s.' % bam_basename,
suffix='.sam')
sam_fpath = os.path.join(temp_dir.name, bam_basename + '.sam')
bam2sam(bam_path.last_version, temp_sam.name)
sam_fhand = open(sam_fpath, 'w')
# First we need to create the sam with added tags and headers
add_header_and_tags_to_sam(temp_sam, sam_fhand)
temp_sam.close()
sam_fhand.close()
#the standardization
temp_sam2 = NamedTemporaryFile(prefix='%s.' % bam_basename,
suffix='.sam', delete=False)
standardize_sam(open(sam_fhand.name), temp_sam2,
default_sanger_quality,
add_def_qual=add_qualities,
only_std_char=True)
temp_sam2.flush()
shutil.move(temp_sam2.name, sam_fhand.name)
temp_sam2.close()
get_sam_fpaths = lambda dir_: [os.path.join(dir_, fname) for fname in os.listdir(dir_) if fname.endswith('.sam')]
# Once the headers are ready we are going to merge
sams = get_sam_fpaths(temp_dir.name)
sams = [open(sam) for sam in sams]
temp_sam = NamedTemporaryFile(suffix='.sam')
reference_fhand = open(reference_path.last_version)
try:
merge_sam(sams, temp_sam, reference_fhand)
except Exception:
if os.path.exists(merged_bam_fpath):
os.remove(merged_bam_fpath)
raise
reference_fhand.close()
# close files
for sam in sams:
sam.close()
# Convert sam into a bam,(Temporary)
temp_bam = NamedTemporaryFile(suffix='.bam')
sam2bam(temp_sam.name, temp_bam.name)
# finally we need to order the bam
#print 'unsorted.bam', temp_bam.name
#raw_input()
sort_bam_sam(temp_bam.name, merged_bam_fpath,
java_conf={'java_memory':java_mem,
'picard_path':picard_path}, tmp_dir=tmp_dir )
temp_bam.close()
temp_sam.close()
create_bam_index(merged_bam_fpath)
self._log({'analysis_finished':True})