def test_get_batches_two_regions():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line()
second_variant = get_variant_line(pos="2", info="Annotation=DDD;Exonic")
variants.append(first_variant)
variants.append(second_variant)
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
chromosomes = get_batches(variants=variants, batch_queue=batch_queue,
header=header)
batch_1 = batch_queue.get()
batch_queue.task_done()
batch_2 = batch_queue.get()
batch_queue.task_done()
assert chromosomes == ['1']
assert len(batch_1) == 1
assert len(batch_2) == 1
def test_get_batches_vep():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line(info="MQ;CSQ=G|ADK")
second_variant = get_variant_line(pos="2", info="MQ;CSQ=G|ADK")
variants.append(first_variant)
variants.append(second_variant)
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
header.vep_columns = ['Allele', 'SYMBOL']
chromosomes = get_batches(variants=variants,
batch_queue=batch_queue,
header=header)
batch_1 = batch_queue.get()
batch_queue.task_done()
batch_2 = batch_queue.get()
batch_queue.task_done()
assert chromosomes == ['1']
assert len(batch_1) == 1
assert len(batch_2) == 1
def test_get_batches_new_chromosome():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line()
second_variant = get_variant_line(chrom="2")
variants.append(first_variant)
variants.append(second_variant)
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
chromosomes = get_batches(variants=variants, batch_queue=batch_queue,
header=header)
batch_1 = batch_queue.get()
batch_queue.task_done()
batch_2 = batch_queue.get()
batch_queue.task_done()
assert chromosomes == ['1', '2']
assert len(batch_1) == 1
assert len(batch_2) == 1
def test_get_batches_two_regions():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line()
second_variant = get_variant_line(pos="2", info="Annotation=DDD;Exonic")
variants.append(first_variant)
variants.append(second_variant)
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
chromosomes = get_batches(variants=variants,
batch_queue=batch_queue,
header=header)
batch_1 = batch_queue.get()
batch_queue.task_done()
batch_2 = batch_queue.get()
batch_queue.task_done()
assert chromosomes == ['1']
assert len(batch_1) == 1
assert len(batch_2) == 1
def test_get_batches_new_chromosome():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line()
second_variant = get_variant_line(chrom="2")
variants.append(first_variant)
variants.append(second_variant)
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
chromosomes = get_batches(variants=variants,
batch_queue=batch_queue,
header=header)
batch_1 = batch_queue.get()
batch_queue.task_done()
batch_2 = batch_queue.get()
batch_queue.task_done()
assert chromosomes == ['1', '2']
assert len(batch_1) == 1
assert len(batch_2) == 1
def test_get_batches_vep():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line(info="MQ;CSQ=G|ADK")
second_variant = get_variant_line(pos="2", info="MQ;CSQ=G|ADK")
variants.append(first_variant)
variants.append(second_variant)
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
header.vep_columns = ['Allele', 'SYMBOL']
chromosomes = get_batches(variants=variants, batch_queue=batch_queue,
header=header)
batch_1 = batch_queue.get()
batch_queue.task_done()
batch_2 = batch_queue.get()
batch_queue.task_done()
assert chromosomes == ['1']
assert len(batch_1) == 1
assert len(batch_2) == 1
def test_get_batches_one():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line()
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
variants.append(first_variant)
chromosomes = get_batches(variants=variants, batch_queue=batch_queue,
header=header)
batch = batch_queue.get()
assert chromosomes == ['1']
assert len(batch) == 1
def test_get_batches_one():
"""
Test to get a batch
"""
batch_queue = Queue()
variants = []
first_variant = get_variant_line()
header = HeaderParser()
header.parse_header_line("#{0}".format(HEADER))
variants.append(first_variant)
chromosomes = get_batches(variants=variants,
batch_queue=batch_queue,
header=header)
batch = batch_queue.get()
assert chromosomes == ['1']
assert len(batch) == 1
def test_variant_printer():
"""Test the variant printer"""
vcf_file = setup_vcf_file()
variant_queue = Manager().Queue()
head = HeaderParser()
outfile = NamedTemporaryFile(mode='w+t', delete=False, suffix='.vcf')
outfile.close()
variant_printer = VariantPrinter(
task_queue=variant_queue,
head=head,
mode='chromosome',
outfile = outfile.name
)
variant_printer.start()
batch = OrderedDict()
for line in open(vcf_file):
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
variant_dict = get_variant_dict(line, head.header)
variant_id = get_variant_id(variant_dict)
variant_dict['variant_id'] = variant_id
variant_dict['info_dict'] = get_info_dict(variant_dict['INFO'])
variant_queue.put(variant_dict)
variant_queue.put(None)
variant_printer.join()
variants = []
with open(outfile.name, 'r', 'utf-8-sig') as f:
for line in f:
variants.append(line.rstrip().split('\t'))
assert variants[0][0] == '1'
assert variants[0][2] == '11900'
def filter(variant_file, annotation, threshold, discard, greater, silent, outfile):
"""
Filter vcf variants.
Filter variants based on their annotation
"""
logger.info("Running genmod filter version {0}".format(__version__))
start_time_analysis = datetime.now()
logger.info("Initializing a Header Parser")
head = HeaderParser()
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
#Add the first variant to the iterator
variant_file = itertools.chain([line], variant_file)
header_line = head.header
if not annotation in head.info_dict:
logger.warning("Annotation {0} not specified in header".format(annotation))
logger.info("Please check VCF file")
logger.info("Exiting...")
sys.exit(1)
logger.info("Building a plugin from extract_vcf for {0}".format(annotation))
annotation_plugin = Plugin(
name=annotation,
field='INFO',
info_key=annotation,
separators = [','],
record_rule = 'min',
data_type = 'float'
)
logger.debug("Plugin=(field={0},info_key={1},separators={2},record_rule={3}"\
",data_type={4})".format('INFO', annotation, "','", 'min', 'float'))
print_headers(head=head, outfile=outfile, silent=silent)
nr_of_variants = 0
nr_of_passed_variants = 0
for variant in variant_file:
nr_of_variants += 1
keep_variant = False
value = annotation_plugin.get_value(variant_line=variant)
logger.debug("Found value {0}".format(value))
if value:
if greater:
if value > threshold:
keep_variant = True
else:
if value < threshold:
keep_variant = True
else:
if not discard:
keep_variant = True
if keep_variant:
logger.debug("Keeping variant")
nr_of_passed_variants += 1
print_variant(
variant_line=variant,
outfile=outfile,
mode='vcf',
silent=silent
)
else:
logger.debug("Discarding variant")
logger.info("Number of variants in file {0}".format(nr_of_variants))
logger.info("Number of variants passing filter {0}".format(nr_of_passed_variants))
logger.info("Number of variants filtered {0}".format(
nr_of_variants - nr_of_passed_variants))
def models(context, variant_file, family_file, family_type, reduced_penetrance,
vep, keyword, phased, strict, silent, processes, outfile, temp_dir,
whole_gene):
"""
Annotate genetic models for vcf variants.
Checks what patterns of inheritance that are followed in a VCF file.
The analysis is family based so each family that are specified in the family
file and exists in the variant file will get it's own annotation.
"""
######### This is for logging the command line string #########
frame = inspect.currentframe()
args, _, _, values = inspect.getargvalues(frame)
argument_list = [
i + '=' + str(values[i]) for i in values
if values[i] and i not in ['frame']
]
variant_file = get_file_handle(variant_file)
###########################################################################
logger.info(
"Running GENMOD annotate models version {0}".format(__version__))
logger.debug("Arguments: {0}".format(', '.join(argument_list)))
reduced_penetrance_genes = set()
nr_reduced_penetrance_genes = 0
if reduced_penetrance:
logger.info("Found file with genes that have reduced penetrance")
for line in reduced_penetrance:
if not line.startswith('#'):
nr_reduced_penetrance_genes += 1
gene_id = line.rstrip().split()[0]
logger.debug(
"Adding gene {0} to reduced penetrance genes".format(
gene_id))
reduced_penetrance_genes.add(gene_id)
logger.info("Found {0} genes with reduced penetrance".format(
nr_reduced_penetrance_genes))
if not family_file:
logger.warning("Please provide a family file with -f/--family_file")
context.abort()
logger.info("Setting up a family parser")
family_parser = FamilyParser(family_file, family_type)
logger.debug("Family parser done")
families = {}
logger.info("Check if the familys have any affected")
for family_id in family_parser.families:
found_affected = False
family_obj = family_parser.families[family_id]
for ind_id in family_obj.individuals:
ind_obj = family_obj.individuals[ind_id]
if ind_obj.affected:
found_affected = True
if found_affected:
families[family_id] = family_obj
else:
logger.warning("No affected individuals found for family {0}."\
" Skipping family.".format(family_id))
if not families:
logger.warning(
"Please provide at least one family with affected individuals")
context.abort()
# The individuals in the ped file must be present in the variant file:
logger.info("Families used in analysis: {0}".format(','.join(
list(families.keys()))))
logger.info("Individuals included in analysis: {0}".format(','.join(
list(family_parser.individuals.keys()))))
head = HeaderParser()
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
#Add the first variant to the iterator
if not line.startswith('#'):
variant_file = itertools.chain([line], variant_file)
else:
print_headers(head=head, outfile=outfile, silent=silent)
sys.exit(0)
if vep:
if not "CSQ" in head.info_dict:
logger.warning(
"vep flag is used but there is no CSQ field specified in header"
)
logger.info("Please check VCF file")
context.abort()
else:
logger.info("Using VEP annotation")
else:
if not keyword in head.info_dict:
logger.warning(
"Annotation key {0} could not be found in VCF header".format(
keyword))
logger.info("Please check VCF file")
context.abort()
else:
logger.info("Using {0} annotation".format(keyword))
if "GeneticModels" in head.info_dict:
logger.warning("Genetic models are already annotated according to vcf"\
" header.")
context.abort()
logger.info("Adding genmod version to vcf header")
head.add_version_tracking(info_id='genmod',
version=__version__,
date=datetime.now().strftime("%Y-%m-%d %H:%M"),
command_line=' '.join(argument_list))
logger.debug("Version added")
logger.info("Adding genetic models to vcf header")
add_metadata(
head,
'info',
'GeneticModels',
annotation_number='.',
entry_type='String',
description="':'-separated list of genetic models for this variant.")
logger.debug("Genetic models added")
logger.info("Adding model score to vcf header")
add_metadata(head,
'info',
'ModelScore',
annotation_number='.',
entry_type='String',
description="PHRED score for genotype models.")
logger.debug("Model score added")
logger.info("Adding Compounds to vcf header")
add_metadata(
head,
'info',
'Compounds',
annotation_number='.',
entry_type='String',
description=(
"List of compound pairs for this variant."
"The list is splitted on ',' family id is separated with compounds"
"with ':'. Compounds are separated with '|'."))
logger.debug("Compounds added")
vcf_individuals = head.individuals
logger.debug("Individuals found in vcf file: {}".format(
', '.join(vcf_individuals)))
try:
check_individuals(family_parser.individuals, vcf_individuals)
except IOError as e:
logger.error(e)
logger.info("Individuals in PED file: {0}".format(', '.join(
family_parser.individuals)))
logger.info("Individuals in VCF file: {0}".format(
', '.join(vcf_individuals)))
context.abort()
start_time_analysis = datetime.now()
analysis_individuals = list(family_parser.individuals.keys())
logger.info("Individuals used in analysis: {0}".format(
', '.join(analysis_individuals)))
###################################################################
### The task queue is where all jobs(in this case batches that ###
### represents variants in a region) is put. The consumers will ###
### then pick their jobs from this queue. ###
###################################################################
logger.debug("Setting up a JoinableQueue for storing variant batches")
# One batch consists of all variants from one or several overlapping genes
# there can be a significant amount of variants in a batch for whole genome
# data...
variant_queue = JoinableQueue(maxsize=100)
logger.debug("Setting up a Queue for storing results from workers")
results = Manager().Queue()
num_model_checkers = processes
#Adapt the number of processes to the machine that run the analysis
logger.info('Number of CPU:s {}'.format(cpu_count()))
logger.info('Number of model checkers: {}'.format(num_model_checkers))
# These are the workers that do the heavy part of the analysis
logger.info('Seting up the workers')
try:
model_checkers = [
VariantAnnotator(task_queue=variant_queue,
results_queue=results,
families=families,
individuals=analysis_individuals,
phased=phased,
strict=strict,
vep=vep,
reduced_penetrance_genes=reduced_penetrance_genes)
for i in range(num_model_checkers)
]
logger.info('Starting the workers')
for worker in model_checkers:
logger.debug('Starting worker {0}'.format(worker))
worker.start()
# This process prints the variants to temporary files
logger.info('Seting up the variant printer')
if len(model_checkers) == 1:
print_headers(head=head, outfile=outfile, silent=silent)
variant_printer = VariantPrinter(task_queue=results,
head=head,
mode='normal',
outfile=outfile)
else:
# We use a temp file to store the processed variants
logger.debug("Build a tempfile for printing the variants")
if temp_dir:
temp_file = NamedTemporaryFile(delete=False, dir=temp_dir)
else:
temp_file = NamedTemporaryFile(delete=False)
temp_file.close()
variant_printer = VariantPrinter(task_queue=results,
head=head,
mode='chromosome',
outfile=temp_file.name)
logger.info('Starting the variant printer process')
variant_printer.start()
start_time_variant_parsing = datetime.now()
# This process parses the original vcf and create batches to put in the variant queue:
logger.info('Start parsing the variants')
chromosome_list = get_batches(variants=variant_file,
batch_queue=variant_queue,
header=head,
vep=vep,
annotation_keyword=keyword)
logger.debug("Put stop signs in the variant queue")
for i in range(num_model_checkers):
variant_queue.put(None)
variant_queue.join()
results.put(None)
variant_printer.join()
if len(model_checkers) > 1:
sort_variants(infile=temp_file.name, mode='chromosome')
print_headers(head=head, outfile=outfile, silent=silent)
with open(temp_file.name, 'r', encoding='utf-8') as f:
for line in f:
print_variant(variant_line=line,
outfile=outfile,
mode='modified',
silent=silent)
except Exception as err:
logger.warning(err)
for worker in model_checkers:
worker.terminate()
variant_printer.terminate()
context.abort()
finally:
if len(model_checkers) > 1:
logger.info("Removing temp file")
os.remove(temp_file.name)
logger.debug("Temp file removed")
logger.info('Time for whole analyis: {0}'.format(
str(datetime.now() - start_time_analysis)))
def models(variant_file, family_file, family_type, reduced_penetrance, vep,
keyword, phased, strict, silent, processes, whole_gene, outfile, temp_dir):
"""
Annotate genetic models for vcf variants.
Checks what patterns of inheritance that are followed in a VCF file.
The analysis is family based so each family that are specified in the family
file and exists in the variant file will get it's own annotation.
"""
logger = logging.getLogger(__name__)
######### This is for logging the command line string #########
frame = inspect.currentframe()
args, _, _, values = inspect.getargvalues(frame)
argument_list = [
i+'='+str(values[i]) for i in values if values[i] and
i not in ['frame']
]
###########################################################################
logger.info("Running GENMOD annotate version {0}".format(__version__))
logger.debug("Arguments: {0}".format(', '.join(argument_list)))
reduced_penetrance_genes = set()
nr_reduced_penetrance_genes = 0
if reduced_penetrance:
logger.info("Found file with genes that have reduced penetrance")
for line in reduced_penetrance:
if not line.startswith('#'):
nr_reduced_penetrance_genes += 1
gene_id = line.rstrip().split()[0]
logger.debug("Adding gene {0} to reduced penetrance genes".format(
gene_id
))
reduced_penetrance_genes.add(
gene_id
)
logger.info("Found {0} genes with reduced penetrance".format(
nr_reduced_penetrance_genes))
if not family_file:
logger.warning("Please provide a family file with -f/--family_file")
logger.info("Exiting")
sys.exit(1)
logger.info("Setting up a family parser")
family_parser = FamilyParser(family_file, family_type)
logger.debug("Family parser done")
families = {}
logger.info("Check if the familys have any affected")
for family_id in family_parser.families:
found_affected = False
family_obj = family_parser.families[family_id]
for ind_id in family_obj.individuals:
ind_obj = family_obj.individuals[ind_id]
if ind_obj.affected:
found_affected = True
if found_affected:
families[family_id] = family_obj
else:
logger.warning("No affected individuals found for family {0}."\
" Skipping family.".format(family_id))
if not families:
logger.warning("Please provide at least one family with affected individuals")
sys.exit(0)
# The individuals in the ped file must be present in the variant file:
logger.info("Families used in analysis: {0}".format(
','.join(list(families.keys()))))
logger.info("Individuals included in analysis: {0}".format(
','.join(list(family_parser.individuals.keys()))))
head = HeaderParser()
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
#Add the first variant to the iterator
variant_file = itertools.chain([line], variant_file)
if vep:
if not "CSQ" in head.info_dict:
logger.warning("vep flag is used but there is no CSQ field specified in header")
logger.info("Please check VCF file")
logger.info("Exiting...")
sys.exit(1)
else:
logger.info("Using VEP annotation")
else:
if not keyword in head.info_dict:
logger.warning("Annotation key {0} could not be found in VCF header".format(keyword))
logger.info("Please check VCF file")
logger.info("Exiting...")
sys.exit(1)
else:
logger.info("Using {0} annotation".format(keyword))
if "GeneticModels" in head.info_dict:
logger.warning("Genetic models are already annotated according to vcf"\
" header.")
logger.info("Exiting...")
sys.exit(1)
logger.info("Adding genmod version to vcf header")
head.add_version_tracking(
info_id='genmod',
version=__version__,
date=datetime.now().strftime("%Y-%m-%d %H:%M"),
command_line=' '.join(argument_list)
)
logger.debug("Version added")
logger.info("Adding genetic models to vcf header")
add_metadata(
head,
'info',
'GeneticModels',
annotation_number='.',
entry_type='String',
description="':'-separated list of genetic models for this variant."
)
logger.debug("Genetic models added")
logger.info("Adding model score to vcf header")
add_metadata(
head,
'info',
'ModelScore',
annotation_number='.',
entry_type='String',
description="PHRED score for genotype models."
)
logger.debug("Model score added")
logger.info("Adding Compounds to vcf header")
add_metadata(
head,
'info',
'Compounds',
annotation_number='.',
entry_type='String',
description=("List of compound pairs for this variant."
"The list is splitted on ',' family id is separated with compounds"
"with ':'. Compounds are separated with '|'.")
)
logger.debug("Compounds added")
vcf_individuals = head.individuals
logger.debug("Individuals found in vcf file: {}".format(', '.join(vcf_individuals)))
start_time_analysis = datetime.now()
try:
check_individuals(family_parser.individuals, vcf_individuals)
except IOError as e:
logger.error(e)
logger.info("Individuals in PED file: {0}".format(
', '.join(family_parser.individuals)))
logger.info("Individuals in VCF file: {0}".format(', '.join(vcf_individuals)))
logger.info("Exiting...")
sys.exit(1)
analysis_individuals = list(family_parser.individuals.keys())
logger.info("Individuals used in analysis: {0}".format(
', '.join(analysis_individuals)))
###################################################################
### The task queue is where all jobs(in this case batches that ###
### represents variants in a region) is put. The consumers will ###
### then pick their jobs from this queue. ###
###################################################################
logger.debug("Setting up a JoinableQueue for storing variant batches")
variant_queue = JoinableQueue(maxsize=1000)
logger.debug("Setting up a Queue for storing results from workers")
results = Manager().Queue()
num_model_checkers = processes
#Adapt the number of processes to the machine that run the analysis
logger.info('Number of CPU:s {}'.format(cpu_count()))
logger.info('Number of model checkers: {}'.format(num_model_checkers))
# These are the workers that do the heavy part of the analysis
logger.info('Seting up the workers')
model_checkers = [
VariantAnnotator(
task_queue=variant_queue,
results_queue=results,
families=families,
individuals=analysis_individuals,
phased=phased,
strict=strict,
whole_gene=whole_gene,
vep=vep,
reduced_penetrance_genes = reduced_penetrance_genes
)
for i in range(num_model_checkers)
]
logger.info('Starting the workers')
for worker in model_checkers:
logger.debug('Starting worker {0}'.format(worker))
worker.start()
# This process prints the variants to temporary files
logger.info('Seting up the variant printer')
if len(model_checkers) == 1:
print_headers(head=head, outfile=outfile, silent=silent)
variant_printer = VariantPrinter(
task_queue=results,
head=head,
mode='normal',
outfile = outfile
)
else:
# We use a temp file to store the processed variants
logger.debug("Build a tempfile for printing the variants")
if temp_dir:
temp_file = NamedTemporaryFile(delete=False, dir=temp_dir)
else:
temp_file = NamedTemporaryFile(delete=False)
temp_file.close()
variant_printer = VariantPrinter(
task_queue=results,
head=head,
mode='chromosome',
outfile = temp_file.name
)
logger.info('Starting the variant printer process')
variant_printer.start()
start_time_variant_parsing = datetime.now()
# This process parses the original vcf and create batches to put in the variant queue:
logger.info('Start parsing the variants')
chromosome_list = get_batches(
variants = variant_file,
batch_queue = variant_queue,
header = head,
vep = vep,
annotation_keyword = keyword
)
logger.debug("Put stop signs in the variant queue")
for i in range(num_model_checkers):
variant_queue.put(None)
variant_queue.join()
results.put(None)
variant_printer.join()
if len(model_checkers) > 1:
sort_variants(infile=temp_file.name, mode='chromosome')
print_headers(head=head, outfile=outfile, silent=silent)
with open(temp_file.name, 'r', encoding='utf-8') as f:
for line in f:
print_variant(
variant_line=line,
outfile=outfile,
mode='modified',
silent=silent
)
logger.debug("Removing temp file")
os.remove(temp_file.name)
logger.debug("Temp file removed")
logger.info('Time for whole analyis: {0}'.format(
str(datetime.now() - start_time_analysis)))
def annotate(context, variant_file, annotate_regions, region_file, cadd_file,
thousand_g, exac, spidex, outfile, silent, cadd_raw, cosmic,
max_af, temp_dir, genome_build):
"""
Annotate vcf variants.
Annotate variants with a number of different sources.
Please use --help for more info.
"""
regions = annotate_regions
logger.info(
"Running genmod annotate_variant version {0}".format(__version__))
if not region_file:
if genome_build == '37':
region_file = ensembl_path_37
elif genome_build == '38':
region_file = ensembl_path_38
start_time_analysis = datetime.now()
annotation_arguments = {}
variants = get_file_handle(variant_file)
logger.info("Initializing a Header Parser")
head = HeaderParser()
line = None
for line in variants:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
#Add the first variant back to the iterator
# If the vcf has no variants the last line will be a header
if not line.startswith('#'):
variants = itertools.chain([line], variants)
else:
print_headers(head, outfile, silent)
sys.exit(0)
header_line = head.header
annotation_arguments['header_line'] = header_line
try:
if regions:
logger.info("Loading annotations")
logger.info("Use annotations file: {0}".format(region_file))
add_regions(head)
regions_handle = get_file_handle(region_file)
logger.debug("Adding region trees to arguments")
annotation_arguments['region_trees'] = build_region_trees(
regions_handle, padding=4000)
if exac:
logger.info("Annotating ExAC frequencies")
logger.debug("Using ExAC file: {0}".format(exac))
annotation_arguments['exac'] = get_tabixhandle(exac)
add_exac(head)
if thousand_g:
logger.info("Annotating 1000G frequencies")
logger.debug("Using 1000G file: {0}".format(thousand_g))
annotation_arguments['thousand_g'] = get_tabixhandle(thousand_g)
add_thousandg(head)
if spidex:
logger.info("Annotating Spidex z scores")
logger.debug("Using Spidex file: {0}".format(spidex))
annotation_arguments['spidex'] = get_tabixhandle(spidex)
add_spidex(head)
if cadd_file:
logger.info("Annotating CADD scores")
logger.debug("Using CADD file(s): {0}".format(
', '.join(cadd_file)))
annotation_arguments['cadd_files'] = [
get_tabixhandle(cadd) for cadd in cadd_file
]
add_cadd(head)
if cadd_raw:
annotation_arguments['cadd_raw'] = cadd_raw
add_cadd_raw(head)
if max_af:
annotation_arguments['max_af'] = max_af
if thousand_g:
add_thousandg_max(head)
if exac:
add_exac_max(head)
if cosmic:
logger.info("Annotating if variant is in COSMIC")
logger.debug("Using COSMIC file: {0}".format(cosmic))
annotation_arguments['cosmic'] = get_tabixhandle(cosmic)
add_cosmic(head)
except TabixError as err:
logger.warning(err)
context.abort()
print_headers(head, outfile, silent)
for variant in variants:
print_variant(variant_line=annotate_variant(variant,
annotation_arguments),
outfile=outfile,
silent=silent)
def compound(context, variant_file, silent, outfile, vep, processes, temp_dir):
"""
Score compound variants in a vcf file based on their rank score.
"""
logger.info(
'Running GENMOD score_compounds, version: {0}'.format(__version__))
variant_file = get_file_handle(variant_file)
start_time_analysis = datetime.now()
logger.info("Initializing a Header Parser")
head = HeaderParser()
line = None
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
logger.info("Headers parsed")
if not line.startswith('#'):
variant_file = itertools.chain([line], variant_file)
else:
print_headers(head=head, outfile=outfile, silent=silent)
sys.exit(0)
header_line = head.header
individuals = head.individuals
###################################################################
### The task queue is where all jobs(in this case batches that ###
### represents variants in a region) is put. The consumers will ###
### then pick their jobs from this queue. ###
###################################################################
logger.debug("Setting up a JoinableQueue for storing variant batches")
variant_queue = JoinableQueue(maxsize=1000)
logger.debug("Setting up a Queue for storing results from workers")
results = Manager().Queue()
num_scorers = processes
#Adapt the number of processes to the machine that run the analysis
logger.info('Number of CPU:s {}'.format(cpu_count()))
logger.info('Number of model checkers: {}'.format(num_scorers))
# These are the workers that do the heavy part of the analysis
logger.info('Seting up the workers')
compound_scorers = [
CompoundScorer(
task_queue=variant_queue,
results_queue=results,
individuals=individuals,
) for i in range(num_scorers)
]
try:
logger.info('Starting the workers')
for worker in compound_scorers:
logger.debug('Starting worker {0}'.format(worker))
worker.start()
# This process prints the variants to temporary files
logger.info('Seting up the variant printer')
# We use a temp file to store the processed variants
logger.debug("Build a tempfile for printing the variants")
if temp_dir:
temp_file = NamedTemporaryFile(delete=False, dir=temp_dir)
else:
temp_file = NamedTemporaryFile(delete=False)
temp_file.close()
variant_printer = VariantPrinter(task_queue=results,
head=head,
mode='chromosome',
outfile=temp_file.name)
logger.info('Starting the variant printer process')
variant_printer.start()
start_time_variant_parsing = datetime.now()
# This process parses the original vcf and create batches to put in the variant queue:
chromosome_list = get_batches(variants=variant_file,
batch_queue=variant_queue,
header=head,
vep=vep,
results_queue=results)
logger.debug("Put stop signs in the variant queue")
for i in range(num_scorers):
variant_queue.put(None)
variant_queue.join()
results.put(None)
variant_printer.join()
sort_variants(infile=temp_file.name, mode='chromosome')
print_headers(head=head, outfile=outfile, silent=silent)
with open(temp_file.name, 'r', encoding='utf-8') as f:
for line in f:
print_variant(variant_line=line,
outfile=outfile,
mode='modified',
silent=silent)
except Exception as e:
logger.warning(e)
for worker in compound_scorers:
worker.terminate()
variant_printer.terminate()
context.abort()
finally:
logger.info("Removing temp file")
os.remove(temp_file.name)
logger.debug("Temp file removed")
logger.info('Time for whole analyis: {0}'.format(
str(datetime.now() - start_time_analysis)))
def annotate(context, variant_file, annotate_regions, region_file, cadd_file,
thousand_g, exac, spidex, outfile, silent, cadd_raw, cosmic,
max_af, temp_dir, genome_build):
"""
Annotate vcf variants.
Annotate variants with a number of different sources.
Please use --help for more info.
"""
regions = annotate_regions
logger.info("Running genmod annotate_variant version {0}".format(__version__))
if not region_file:
if genome_build == '37':
region_file = ensembl_path_37
elif genome_build == '38':
region_file = ensembl_path_38
start_time_analysis = datetime.now()
annotation_arguments = {}
variants = get_file_handle(variant_file)
logger.info("Initializing a Header Parser")
head = HeaderParser()
line = None
for line in variants:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
#Add the first variant back to the iterator
# If the vcf has no variants the last line will be a header
if not line.startswith('#'):
variants = itertools.chain([line], variants)
else:
print_headers(head, outfile, silent)
sys.exit(0)
header_line = head.header
annotation_arguments['header_line'] = header_line
try:
if regions:
logger.info("Loading annotations")
logger.info("Use annotations file: {0}".format(region_file))
add_regions(head)
regions_handle = get_file_handle(region_file)
logger.debug("Adding region trees to arguments")
annotation_arguments['region_trees'] = build_region_trees(regions_handle, padding=4000)
if exac:
logger.info("Annotating ExAC frequencies")
logger.debug("Using ExAC file: {0}".format(exac))
annotation_arguments['exac'] = get_tabixhandle(exac)
add_exac(head)
if thousand_g:
logger.info("Annotating 1000G frequencies")
logger.debug("Using 1000G file: {0}".format(thousand_g))
annotation_arguments['thousand_g'] = get_tabixhandle(thousand_g)
add_thousandg(head)
if spidex:
logger.info("Annotating Spidex z scores")
logger.debug("Using Spidex file: {0}".format(spidex))
annotation_arguments['spidex'] = get_tabixhandle(spidex)
add_spidex(head)
if cadd_file:
logger.info("Annotating CADD scores")
logger.debug("Using CADD file(s): {0}".format(', '.join(cadd_file)))
annotation_arguments['cadd_files'] = [get_tabixhandle(cadd) for cadd in cadd_file]
add_cadd(head)
if cadd_raw:
annotation_arguments['cadd_raw'] = cadd_raw
add_cadd_raw(head)
if max_af:
annotation_arguments['max_af'] = max_af
if thousand_g:
add_thousandg_max(head)
if exac:
add_exac_max(head)
if cosmic:
logger.info("Annotating if variant is in COSMIC")
logger.debug("Using COSMIC file: {0}".format(cosmic))
annotation_arguments['cosmic'] = get_tabixhandle(cosmic)
add_cosmic(head)
except TabixError as err:
logger.warning(err)
context.abort()
print_headers(head, outfile, silent)
for variant in variants:
print_variant(
variant_line = annotate_variant(variant, annotation_arguments),
outfile = outfile,
silent = silent
)
def filter(variant_file, annotation, threshold, discard, greater, silent,
outfile):
"""
Filter vcf variants.
Filter variants based on their annotation
"""
logger.info("Running genmod filter version {0}".format(__version__))
variant_file = get_file_handle(variant_file)
start_time_analysis = datetime.now()
logger.info("Initializing a Header Parser")
head = HeaderParser()
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
#Add the first variant to the iterator
variant_file = itertools.chain([line], variant_file)
header_line = head.header
if not annotation in head.info_dict:
logger.warning(
"Annotation {0} not specified in header".format(annotation))
logger.info("Please check VCF file")
logger.info("Exiting...")
sys.exit(1)
logger.info(
"Building a plugin from extract_vcf for {0}".format(annotation))
annotation_plugin = Plugin(name=annotation,
field='INFO',
info_key=annotation,
separators=[','],
record_rule='min',
data_type='float')
logger.debug("Plugin=(field={0},info_key={1},separators={2},record_rule={3}"\
",data_type={4})".format('INFO', annotation, "','", 'min', 'float'))
print_headers(head=head, outfile=outfile, silent=silent)
nr_of_variants = 0
nr_of_passed_variants = 0
for variant in variant_file:
nr_of_variants += 1
keep_variant = False
value = annotation_plugin.get_value(variant_line=variant)
logger.debug("Found value {0}".format(value))
if value:
if greater:
if value > threshold:
keep_variant = True
else:
if value < threshold:
keep_variant = True
else:
if not discard:
keep_variant = True
if keep_variant:
logger.debug("Keeping variant")
nr_of_passed_variants += 1
print_variant(variant_line=variant,
outfile=outfile,
mode='vcf',
silent=silent)
else:
logger.debug("Discarding variant")
logger.info("Number of variants in file {0}".format(nr_of_variants))
logger.info(
"Number of variants passing filter {0}".format(nr_of_passed_variants))
logger.info(
"Number of variants filtered {0}".format(nr_of_variants -
nr_of_passed_variants))
def score(context, variant_file, family_id, family_file, family_type,
score_config, silent, skip_plugin_check, rank_results, outfile):
"""
Score variants in a vcf file using a Weighted Sum Model.
The specific scores should be defined in a config file, see examples on
github.
"""
logger.info('Running GENMOD score, version: {0}'.format(__version__))
logger.info("Checking family id")
variant_file = get_file_handle(variant_file)
if family_file:
logger.info("Setting up a family parser")
family_parser = FamilyParser(family_file, family_type)
logger.debug("Family parser done")
family_id = list(family_parser.families.keys())[0]
logger.info("Family used in analysis: {0}".format(family_id))
## Check the score config:
if not score_config:
logger.warning("Please provide a score config file.")
context.abort()
logger.debug("Parsing config file")
try:
config_parser = ConfigParser(score_config)
except ValidateError as e:
logger.error(e.message)
context.abort()
score_categories = list(config_parser.categories.keys())
logger.debug("Config parsed succesfully")
logger.info("Initializing a Header Parser")
head = HeaderParser()
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
logger.info("Check if all score plugins exist in vcf ...")
if not check_plugins(config_parser, head):
if not skip_plugin_check:
logger.error("All score plugins has to be defined in vcf header")
context.abort()
else:
logger.info("All plugins are defined in vcf")
csq_format = head.vep_columns
#Add the first variant to the iterator
if not line.startswith('#'):
variant_file = itertools.chain([line], variant_file)
else:
print_headers(head=head, outfile=outfile, silent=silent)
sys.exit(0)
header_line = head.header
if "RankScore" in head.info_dict:
logger.warning("Variants already scored according to VCF header")
logger.info("Please check VCF file")
context.abort()
add_metadata(
head,
'info',
'RankScore',
annotation_number='.',
entry_type='String',
description=
"The rank score for this variant in this family. family_id:rank_score."
)
if rank_results:
add_metadata(head,
'info',
'RankResult',
annotation_number='.',
entry_type='String',
description='|'.join(score_categories))
print_headers(head=head, outfile=outfile, silent=silent)
start_scoring = datetime.now()
last_twenty = datetime.now()
nr_of_variants = 1
for line in variant_file:
if not line.startswith('#'):
variant = get_variant_dict(line, header_line)
variant['info_dict'] = get_info_dict(variant['INFO'])
rank_score = 0
# This is for printing results to vcf:
category_scores = []
for category in score_categories:
category_score = get_category_score(
variant=variant,
category=category,
config_parser=config_parser,
csq_format=csq_format)
logger.debug("Adding category score {0} to rank_score".format(
category_score))
rank_score += category_score
logger.debug("Updating rank score to {0}".format(rank_score))
category_scores.append(str(category_score))
variant = add_vcf_info(keyword='RankScore',
variant_dict=variant,
annotation="{0}:{1}".format(
family_id, rank_score))
if rank_results:
variant = add_vcf_info(keyword='RankResult',
variant_dict=variant,
annotation="|".join(category_scores))
print_variant(variant_dict=variant,
header_line=header_line,
outfile=outfile,
silent=silent)
nr_of_variants += 1
if nr_of_variants % 20000 == 0:
logger.info("{0} variants scored.".format(nr_of_variants))
logger.info(
"Last 20000 took {0} to score.".format(datetime.now() -
last_twenty))
last_twenty = datetime.now()
logger.info(
"Variants scored. Number of variants: {0}".format(nr_of_variants))
logger.info("Time to score variants: {0}".format(datetime.now() -
start_scoring))
def score(variant_file, family_id, family_file, family_type, score_config,
silent, skip_plugin_check, rank_results, outfile):
"""
Score variants in a vcf file using a Weighted Sum Model.
The specific scores should be defined in a config file, see examples on
github.
"""
logger = logging.getLogger(__name__)
logger.info('Running GENMOD score, version: {0}'.format(__version__))
logger.info("Checking family id")
if family_file:
logger.info("Setting up a family parser")
family_parser = FamilyParser(family_file, family_type)
logger.debug("Family parser done")
family_id = list(family_parser.families.keys())[0]
logger.info("Family used in analysis: {0}".format(family_id))
## Check the score config:
if not score_config:
logger.warning("Please provide a score config file.")
logger.info("Exiting")
sys.exit(1)
logger.debug("Parsing config file")
try:
config_parser = ConfigParser(score_config)
except ValidateError as e:
logger.error(e.message)
logger.info("Exiting")
sys.exit(1)
score_categories = list(config_parser.categories.keys())
logger.debug("Config parsed succesfully")
logger.info("Initializing a Header Parser")
head = HeaderParser()
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
logger.info("Check if all score plugins exist in vcf ...")
if not check_plugins(config_parser, head):
if not skip_plugin_check:
logger.error("All score plugins has to be defined in vcf header")
logger.info("Exiting")
sys.exit(1)
else:
logger.info("All plugins are defined in vcf")
#Add the first variant to the iterator
variant_file = itertools.chain([line], variant_file)
header_line = head.header
if "RankScore" in head.info_dict:
logger.warning("Variants already scored according to VCF header")
logger.info("Please check VCF file")
logger.info("Exiting...")
sys.exit(1)
add_metadata(
head,
'info',
'RankScore',
annotation_number='.',
entry_type='String',
description="The rank score for this variant in this family. family_id:rank_score."
)
if rank_results:
add_metadata(
head,
'info',
'RankResult',
annotation_number='.',
entry_type='String',
description= '|'.join(score_categories)
)
print_headers(
head=head,
outfile=outfile,
silent=silent
)
start_scoring = datetime.now()
last_twenty = datetime.now()
nr_of_variants = 1
for line in variant_file:
if not line.startswith('#'):
variant = get_variant_dict(line, header_line)
variant['info_dict'] = get_info_dict(variant['INFO'])
rank_score = 0
# This is for printing results to vcf:
category_scores = []
for category in score_categories:
category_score = get_category_score(variant, category, config_parser)
logger.debug("Adding category score {0} to rank_score".format(category_score))
rank_score += category_score
logger.debug("Updating rank score to {0}".format(rank_score))
category_scores.append(str(category_score))
variant = add_vcf_info(
keyword = 'RankScore',
variant_dict=variant,
annotation="{0}:{1}".format(family_id, rank_score)
)
if rank_results:
variant = add_vcf_info(
keyword = 'RankResult',
variant_dict=variant,
annotation="|".join(category_scores)
)
print_variant(
variant_dict=variant,
header_line=header_line,
outfile=outfile,
silent=silent
)
nr_of_variants += 1
if nr_of_variants % 20000 == 0:
logger.info("{0} variants scored.".format(nr_of_variants))
logger.info("Last 20000 took {0} to score.".format(datetime.now()-last_twenty))
last_twenty = datetime.now()
logger.info("Variants scored. Number of variants: {0}".format(nr_of_variants))
logger.info("Time to score variants: {0}".format(datetime.now()-start_scoring))
def sort(variant_file, outfile, family_id, silent, position, temp_dir):
"""
Sort a VCF file based on rank score.
"""
head = HeaderParser()
variant_file = get_file_handle(variant_file)
logger.info("Running GENMOD sort version {0}".format(__version__))
start = datetime.now()
# Create a temporary variant file for sorting
logger.debug("Creating temporary file for sorting")
if temp_dir:
temp_file = NamedTemporaryFile(delete=False, dir=temp_dir)
else:
temp_file = NamedTemporaryFile(delete=False)
temp_file.close()
# Open the temp file with codecs
temp_file_handle = open(temp_file.name,
mode='w',
encoding='utf-8',
errors='replace')
logger.debug("Temp file created")
logger.info("Printing variants to temp file")
nr_variants = 0
# Print the variants with rank score in first column
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
nr_variants += 1
priority = '0'
if position:
chrom = line.split()[0]
priority = get_chromosome_priority(chrom)
else:
priority = get_rank_score(line)
print_variant(variant_line=line,
priority=priority,
outfile=temp_file_handle)
temp_file_handle.close()
logger.info("Variants printed to temp file")
logger.info("Nr or variants in VCF file: {0}".format(nr_variants))
sort_mode = 'rank'
if nr_variants == 0:
logger.debug("Printing headers")
print_headers(head=head, outfile=outfile, silent=silent)
sys.exit(0)
if position:
sort_mode = 'chromosome'
logger.info("Sorting variants")
sort_variants(infile=temp_file.name, mode=sort_mode)
logger.info("Variants sorted")
logger.debug("Printing headers")
print_headers(head=head, outfile=outfile, silent=silent)
logger.debug("Headers printed")
logger.info("Printing variants")
with open(temp_file.name, mode='r', encoding='utf-8',
errors='replace') as f:
for variant_line in f:
print_variant(variant_line=variant_line,
outfile=outfile,
mode='modified',
silent=False)
logger.debug("Variants printed")
logger.info("Removing temp file")
os.remove(temp_file.name)
logger.debug("Temp file removed")
logger.info("Sorting done, time for sorting: {0}".format(datetime.now() -
start))
def sort(variant_file, outfile, family_id, silent, position, temp_dir):
"""
Sort a VCF file based on rank score.
"""
logger = logging.getLogger(__name__)
head = HeaderParser()
logger.info("Running GENMOD sort version {0}".format(__version__))
start = datetime.now()
# Create a temporary variant file for sorting
logger.debug("Creating temporary file for sorting")
if temp_dir:
temp_file = NamedTemporaryFile(delete=False, dir=temp_dir)
else:
temp_file = NamedTemporaryFile(delete=False)
temp_file.close()
# Open the temp file with codecs
temp_file_handle = open(
temp_file.name,
mode='w',
encoding='utf-8',
errors='replace'
)
logger.debug("Temp file created")
logger.info("Printing variants to temp file")
nr_variants = 0
# Print the variants with rank score in first column
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
nr_variants += 1
priority = '0'
if position:
chrom = line.split()[0]
priority = get_chromosome_priority(chrom)
else:
priority = get_rank_score(line)
print_variant(
variant_line=line,
priority=priority,
outfile=temp_file_handle
)
temp_file_handle.close()
logger.info("Variants printed to temp file")
logger.info("Nr or variants in VCF file: {0}".format(nr_variants))
sort_mode = 'rank'
if position:
sort_mode = 'chromosome'
logger.info("Sorting variants")
sort_variants(
infile = temp_file.name,
mode=sort_mode
)
logger.info("Variants sorted")
logger.debug("Printing headers")
print_headers(
head = head,
outfile = outfile,
silent=silent
)
logger.debug("Headers printed")
logger.info("Printing variants")
with open(temp_file.name, mode='r', encoding='utf-8', errors='replace') as f:
for variant_line in f:
print_variant(
variant_line = variant_line,
outfile = outfile,
mode = 'modified',
silent=False
)
logger.debug("Variants printed")
logger.info("Removing temp file")
os.remove(temp_file.name)
logger.debug("Temp file removed")
logger.info("Sorting done, time for sorting: {0}".format(datetime.now()-start))
def summarize(variant_file, family_file, frequency_treshold, frequency_keyword,
cadd_treshold, cadd_keyword, gq_treshold, read_depth_treshold):
"""
Summarize the the variants in a vcf.
There will be one result line per individual.
- How many variants found\n
- How many variants did not satisfy the base call
quality treshold. (Default 20)\n
- How many variants where not covered in all individuals.
(Default depth 10)\n
- How many variants followed each model in each family:\n
- AR_hom\n
- AR_comp\n
- AR_hom_dn\n
- AR_comp_dn\n
- AD\n
- AD_dn\n
- XD\n
- XD_dn\n
- XR\n
- XR_dn\n
- How many rare variants (Default maf < 0.02)\n
- How many high scored cadd. (Default cadd = 0)\n
- How many rare + high score cadd\n
- How many no cadd score\n
- How many indels\n
- How many indels without cadd score\n
"""
logger = logging.getLogger(__name__)
logger = logging.getLogger("genmod.commands.summarize_variants")
head = HeaderParser()
nr_of_variants = 0
header = ['sample_id', 'nr_of_variants']
samples = {}
logger.debug("Setting up a variant parser")
if variant_file == '-':
variant_parser = VCFParser(fsock=sys.stdin, check_info=False)
else:
variant_parser = VCFParser(infile=variant_file, check_info=False)
logger.debug("Variant parser setup")
head = variant_parser.metadata
for sample_id in head.individuals:
samples[sample_id] = {}
samples[sample_id]["nr_of_variants"] = 0
for variant in variant_parser:
for sample_id in samples:
samples[sample_id]["nr_of_variants"] += 1
print(variant['genotypes'][sample_id].depth_of_coverage)
print(json.dumps(samples))
def annotate(variant_file, annotate_regions, cadd_file, thousand_g, exac,
spidex,annotation_dir, outfile, silent, cadd_raw, cosmic, max_af, processes,
temp_dir):
"""
Annotate vcf variants.
Annotate variants with a number of different sources.
Please use --help for more info.
"""
logger.info("Running genmod annotate_variant version {0}".format(__version__))
start_time_analysis = datetime.now()
annotator_arguments = {}
logger.info("Initializing a Header Parser")
head = HeaderParser()
line = None
for line in variant_file:
line = line.rstrip()
if line.startswith('#'):
if line.startswith('##'):
head.parse_meta_data(line)
else:
head.parse_header_line(line)
else:
break
#Add the first variant to the iterator
if line:
variant_file = itertools.chain([line], variant_file)
header_line = head.header
annotator_arguments['header_line'] = header_line
if annotate_regions:
logger.info("Loading annotations")
gene_trees, exon_trees = load_annotations(annotation_dir)
annotator_arguments['gene_trees'] = gene_trees
annotator_arguments['exon_trees'] = exon_trees
add_metadata(
head,
'info',
'Annotation',
annotation_number='.',
entry_type='String',
description='Annotates what feature(s) this variant belongs to.'
)
add_metadata(
head,
'info',
'Exonic',
annotation_number='0',
entry_type='Flag',
description='Indicates if the variant is exonic.'
)
if exac:
logger.info("Annotating ExAC frequencies")
logger.debug("Using ExAC file: {0}".format(exac))
annotator_arguments['exac'] = exac
add_metadata(
head,
'info',
'ExACAF',
annotation_number='1',
entry_type='Float',
description="Frequency in the ExAC database."
)
if thousand_g:
logger.info("Annotating 1000G frequencies")
logger.debug("Using 1000G file: {0}".format(thousand_g))
annotator_arguments['thousand_g'] = thousand_g
add_metadata(
head,
'info',
'1000GAF',
annotation_number='1',
entry_type='Float',
description="Frequency in the 1000G database."
)
if spidex:
logger.info("Annotating Spidex z scores")
logger.debug("Using Spidex file: {0}".format(spidex))
annotator_arguments['spidex'] = spidex
add_metadata(
head,
'info',
'SPIDEX',
annotation_number='1',
entry_type='Float',
description="Z score from the spidex database."
)
if cadd_file:
logger.info("Annotating CADD scores")
logger.debug("Using CADD file(s): {0}".format(', '.join(cadd_file)))
annotator_arguments['cadd_files'] = cadd_file
any_cadd_file = True
add_metadata(
head,
'info',
'CADD',
annotation_number='1',
entry_type='Integer',
description="The CADD relative score for this alternative."
)
if cadd_raw:
annotator_arguments['cadd_raw'] = cadd_raw
logger.debug("Adding vcf metadata for CADD raw score")
add_metadata(
head,
'info',
'CADD_raw',
annotation_number='1',
entry_type='Float',
description="The CADD raw score(s) for this alternative(s)."
)
if max_af:
annotator_arguments['max_af'] = max_af
if thousand_g:
add_metadata(
head,
'info',
'1000G_MAX_AF',
annotation_number='1',
entry_type='Float',
description="The max af for thousand genomes populations."
)
if exac:
add_metadata(
head,
'info',
'ExAC_MAX_AF',
annotation_number='1',
entry_type='Float',
description="The max af for ExAC populations."
)
if cosmic:
logger.info("Annotating if variant is in COSMIC")
logger.debug("Using COSMOC file: {0}".format(cosmic))
annotator_arguments['cosmic'] = cosmic
add_metadata(
head,
'info',
'COSMIC',
annotation_number='0',
entry_type='Flag',
description="If variant is in COSMIC database."
)
###################################################################
### The task queue is where all jobs(in this case batches that ###
### represents variants in a region) is put. The consumers will ###
### then pick their jobs from this queue. ###
###################################################################
logger.debug("Setting up a JoinableQueue for storing variant batches")
variant_queue = JoinableQueue(maxsize=1000)
logger.debug("Setting up a Queue for storing results from workers")
results = Manager().Queue()
num_annotators = processes
#Adapt the number of processes to the machine that run the analysis
if cadd_file or spidex:
# We need more power when annotating cadd scores:
# But if flag is used that overrides
if num_annotators == min(4, cpu_count()):
num_annotators = min(8, cpu_count())
logger.info('Number of CPU:s {}'.format(cpu_count()))
logger.info('Number of model checkers: {}'.format(num_annotators))
# These are the workers that do the heavy part of the analysis
logger.info('Setting up the workers')
annotators = [
VariantAnnotator(
variant_queue,
results,
**annotator_arguments
)
for i in range(num_annotators)
]
logger.info('Starting the workers')
for worker in annotators:
logger.debug('Starting worker {0}'.format(worker))
worker.start()
# This process prints the variants to temporary files
# If there is only one annotation process we can print the results as soon
# as they are done
logger.info('Setting up the variant printer')
if len(annotators) == 1:
print_headers(head, outfile, silent)
var_printer = VariantPrinter(
task_queue = results,
head = head,
mode='normal',
outfile = outfile
)
else:
# We use a temp file to store the processed variants
logger.debug("Build a tempfile for printing the variants")
if temp_dir:
temp_file = NamedTemporaryFile(delete=False, dir=temp_dir)
else:
temp_file = NamedTemporaryFile(delete=False)
temp_file.close()
var_printer = VariantPrinter(
task_queue = results,
head = head,
mode='chromosome',
outfile = temp_file.name
)
logger.info('Starting the variant printer process')
var_printer.start()
start_time_variant_parsing = datetime.now()
start_time_twenty = datetime.now()
nr_of_lines = 0
# This process parses the original vcf and create batches to put in the variant queue:
logger.info('Start parsing the variants')
for line in variant_file:
line = line.rstrip()
if not line.startswith('#'):
variant_queue.put(line)
nr_of_lines += 1
if nr_of_lines % 20000 == 0:
logger.info('{0} variants parsed'.format(nr_of_lines))
logger.info('Last 20000 took {0} to parse'.format(
datetime.now()-start_time_twenty))
start_time_twenty = datetime.now()
logger.info('Put stop signs in the variant queue')
for i in range(num_annotators):
variant_queue.put(None)
variant_queue.join()
results.put(None)
var_printer.join()
if len(annotators) > 1:
logger.info("Start sorting the variants")
sort_variants(temp_file.name, mode='chromosome')
logger.info("Print the headers")
print_headers(head, outfile, silent)
with open(temp_file.name, 'r', encoding='utf-8') as f:
for line in f:
print_variant(
variant_line=line,
outfile=outfile,
mode='modified',
silent=silent
)
logger.info("Removing temp file")
os.remove(temp_file.name)
logger.debug("Temp file removed")
logger.info('Time for whole analyis: {0}'.format(
str(datetime.now() - start_time_analysis)))