def check_format_ad_dp_tags(vcf, pcgr_directory, config_options, logger):
found_taf_tag = 0
found_tdp_tag = 0
found_naf_tag = 0
found_ndp_tag = 0
found_call_conf_tag = 0
tumor_dp_tag = config_options['allelic_support']['tumor_dp_tag']
tumor_af_tag = config_options['allelic_support']['tumor_af_tag']
normal_dp_tag = config_options['allelic_support']['normal_dp_tag']
normal_af_tag = config_options['allelic_support']['normal_af_tag']
call_conf_tag = config_options['allelic_support']['call_conf_tag']
annoutils.detect_reserved_info_tag(tumor_dp_tag, 'tumor_dp_tag', logger)
annoutils.detect_reserved_info_tag(normal_dp_tag, 'normal_dp_tag', logger)
annoutils.detect_reserved_info_tag(tumor_af_tag, 'tumor_af_tag', logger)
annoutils.detect_reserved_info_tag(normal_af_tag, 'normal_af_tag', logger)
annoutils.detect_reserved_info_tag(call_conf_tag, 'call_conf_tag', logger)
for e in vcf.header_iter():
header_element = e.info()
if 'ID' in header_element.keys(
) and 'HeaderType' in header_element.keys():
if header_element['HeaderType'] == 'INFO':
if header_element['ID'] == tumor_dp_tag:
if header_element['Type'] == 'Integer':
logger.info(
'Found INFO tag for tumor variant sequencing depth (tumor_dp_tag '
+ str(tumor_dp_tag) + ') in input VCF')
found_tdp_tag = 1
else:
err_msg = 'INFO tag for tumor variant sequencing depth (tumor_dp_tag ' + str(
tumor_dp_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']
) + '), should be Type=Integer'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == tumor_af_tag:
if header_element['Type'] == 'Float':
logger.info(
'Found INFO tag for tumor variant allelic fraction (tumor_af_tag '
+ str(tumor_af_tag) + ') in input VCF')
found_taf_tag = 1
else:
err_msg = 'INFO tag for tumor variant allelic fraction (tumor_af_tag ' + str(
tumor_af_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']) + '), should be Type=Float'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == normal_dp_tag:
if header_element['Type'] == 'Integer':
logger.info(
'Found INFO tag for normal/control variant sequencing depth (normal_dp_tag '
+ str(normal_dp_tag) + ') in input VCF')
found_ndp_tag = 1
else:
err_msg = 'INFO tag for normal/control variant sequencing depth (normal_dp_tag ' + str(
normal_dp_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']
) + '), should be Type=Integer'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == normal_af_tag:
if header_element['Type'] == 'Float':
logger.info(
'Found INFO tag for normal/control allelic fraction (normal_af_tag '
+ str(normal_af_tag) + ') in input VCF')
found_naf_tag = 1
else:
err_msg = 'INFO tag for for normal/control allelic fraction (normal_af_tag ' + str(
normal_af_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']) + ') should be Type=Float'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == call_conf_tag:
if header_element['Type'] == 'String':
logger.info(
'Found INFO tag for variant call confidence (call_conf_tag '
+ str(call_conf_tag) + ') in input VCF')
found_call_conf_tag = 1
else:
err_msg = 'INFO tag for variant call confidence (call_conf_tag) is not correctly specified in input VCF (Type=' + str(
header_element['Type']
) + '), should be Type=String'
return pcgr_error_message(err_msg, logger)
if call_conf_tag != '' and found_call_conf_tag == 0:
logger.warn('Could not find the specified call_conf_tag (' +
str(call_conf_tag) + ') in INFO column of input VCF')
if tumor_dp_tag != '' and found_tdp_tag == 0:
logger.warn('Could not find the specified tumor_dp_tag (' +
str(tumor_dp_tag) + ') in INFO column of input VCF')
if tumor_af_tag != '' and found_taf_tag == 0:
logger.warn('Could not find the specified tumor_af_tag (' +
str(tumor_af_tag) + ') in INFO column of input VCF')
if normal_dp_tag != '' and found_ndp_tag == 0:
logger.warn('Could not find the specified normal_dp_tag (' +
str(normal_dp_tag) + ') in INFO column of input VCF')
if normal_af_tag != '' and found_naf_tag == 0:
logger.warn('Could not find the specified normal_af_tag (' +
str(normal_af_tag) + ') in INFO column of input VCF')
if found_tdp_tag == 1 and found_taf_tag == 0:
logger.warn(
'BOTH \' tumor_dp_tag\' AND \' tumor_af_tag\' need to be specified for use in tumor report (\'tumor_af_tag\' is missing)'
)
if found_tdp_tag == 0 and found_taf_tag == 1:
logger.warn(
'BOTH \'tumor_dp_tag\' AND \'tumor_af_tag\' need to be specified for use in tumor report (\'tumor_dp_tag\' is missing)'
)
if found_ndp_tag == 1 and found_naf_tag == 0:
logger.warn(
'BOTH \'normal_dp_tag\' AND \'normal_af_tag\' need to be specified for use in tumor report (\'normal_af_tag\' is missing)'
)
if found_ndp_tag == 0 and found_naf_tag == 1:
logger.warn(
'BOTH \'normal_dp_tag\' AND \'normal_af_tag\' need to be specified for use in tumor report (\'normal_dp_tag\' is missing)'
)
return 0
def check_format_ad_dp_tags(vcf, pcgr_directory, config_options, logger):
found_taf_tag = 0
found_tdp_tag = 0
found_naf_tag = 0
found_ndp_tag = 0
found_call_conf_tag = 0
tumor_dp_tag = config_options['allelic_support']['tumor_dp_tag']
tumor_af_tag = config_options['allelic_support']['tumor_af_tag']
control_dp_tag = config_options['allelic_support']['control_dp_tag']
control_af_tag = config_options['allelic_support']['control_af_tag']
call_conf_tag = config_options['allelic_support']['call_conf_tag']
annoutils.detect_reserved_info_tag(tumor_dp_tag, 'tumor_dp_tag', logger)
annoutils.detect_reserved_info_tag(control_dp_tag, 'control_dp_tag',
logger)
annoutils.detect_reserved_info_tag(tumor_af_tag, 'tumor_af_tag', logger)
annoutils.detect_reserved_info_tag(control_af_tag, 'control_af_tag',
logger)
annoutils.detect_reserved_info_tag(call_conf_tag, 'call_conf_tag', logger)
for e in vcf.header_iter():
header_element = e.info()
if 'ID' in header_element.keys(
) and 'HeaderType' in header_element.keys():
if header_element['HeaderType'] == 'INFO':
if header_element['ID'] == tumor_dp_tag:
if header_element['Type'] == 'Integer':
logger.info(
'Found INFO tag for tumor variant sequencing depth (tumor_dp_tag '
+ str(tumor_dp_tag) + ') in input VCF')
found_tdp_tag = 1
else:
err_msg = 'INFO tag for tumor variant sequencing depth (tumor_dp_tag ' + str(
tumor_dp_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']
) + '), should be Type=Integer'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == tumor_af_tag:
if header_element['Type'] == 'Float':
logger.info(
'Found INFO tag for tumor variant allelic fraction (tumor_af_tag '
+ str(tumor_af_tag) + ') in input VCF')
found_taf_tag = 1
else:
err_msg = 'INFO tag for tumor variant allelic fraction (tumor_af_tag ' + str(
tumor_af_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']) + '), should be Type=Float'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == control_dp_tag:
if header_element['Type'] == 'Integer':
logger.info(
'Found INFO tag for normal/control variant sequencing depth (control_dp_tag '
+ str(control_dp_tag) + ') in input VCF')
found_ndp_tag = 1
else:
err_msg = 'INFO tag for normal/control variant sequencing depth (control_dp_tag ' + str(
control_dp_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']
) + '), should be Type=Integer'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == control_af_tag:
if header_element['Type'] == 'Float':
logger.info(
'Found INFO tag for normal/control allelic fraction (control_af_tag '
+ str(control_af_tag) + ') in input VCF')
found_naf_tag = 1
else:
err_msg = 'INFO tag for for normal/control allelic fraction (control_af_tag ' + str(
control_af_tag
) + ') is not correctly specified in input VCF (Type=' + str(
header_element['Type']) + ') should be Type=Float'
return pcgr_error_message(err_msg, logger)
if header_element['ID'] == call_conf_tag:
if header_element['Type'] == 'String':
logger.info(
'Found INFO tag for variant call confidence (call_conf_tag '
+ str(call_conf_tag) + ') in input VCF')
found_call_conf_tag = 1
else:
err_msg = 'INFO tag for variant call confidence (call_conf_tag) is not correctly specified in input VCF (Type=' + str(
header_element['Type']
) + '), should be Type=String'
return pcgr_error_message(err_msg, logger)
if call_conf_tag != '' and found_call_conf_tag == 0:
logger.warn('Could not find the specified call_conf_tag (' +
str(call_conf_tag) + ') in INFO column of input VCF')
if tumor_dp_tag != '' and found_tdp_tag == 0:
logger.warn('Could not find the specified tumor_dp_tag (' +
str(tumor_dp_tag) + ') in INFO column of input VCF')
if tumor_af_tag != '' and found_taf_tag == 0:
logger.warn('Could not find the specified tumor_af_tag (' +
str(tumor_af_tag) + ') in INFO column of input VCF')
if control_dp_tag != '' and found_ndp_tag == 0:
logger.warn('Could not find the specified control_dp_tag (' +
str(control_dp_tag) + ') in INFO column of input VCF')
if control_af_tag != '' and found_naf_tag == 0:
logger.warn('Could not find the specified control_af_tag (' +
str(control_af_tag) + ') in INFO column of input VCF')
if config_options['tumor_only'][
'exclude_likely_hom_germline'] is True and config_options[
'tumor_only']['vcf_tumor_only'] is True and found_taf_tag == 0:
logger.warn(
'Could not find the specified tumor_af_tag (' + str(tumor_af_tag) +
') in INFO column of input VCF - filtering of homozygous germline variants in tumor-only mode will be ignored'
)
if config_options['tumor_only'][
'exclude_likely_het_germline'] is True and config_options[
'tumor_only']['vcf_tumor_only'] is True and found_taf_tag == 0:
logger.warn(
'Could not find the specified tumor_af_tag (' + str(tumor_af_tag) +
') in INFO column of input VCF - filtering of heterozygous germline variants in tumor-only mode will be ignored'
)
if found_tdp_tag == 1 and found_taf_tag == 0:
logger.warn(
'BOTH \' tumor_dp_tag\' AND \' tumor_af_tag\' need to be specified for use in tumor report (\'tumor_af_tag\' is missing)'
)
if found_tdp_tag == 0 and found_taf_tag == 1:
logger.warn(
'BOTH \'tumor_dp_tag\' AND \'tumor_af_tag\' need to be specified for use in tumor report (\'tumor_dp_tag\' is missing)'
)
if found_ndp_tag == 1 and found_naf_tag == 0:
logger.warn(
'BOTH \'control_dp_tag\' AND \'control_af_tag\' need to be specified for use in tumor report (\'control_af_tag\' is missing)'
)
if found_ndp_tag == 0 and found_naf_tag == 1:
logger.warn(
'BOTH \'control_dp_tag\' AND \'control_af_tag\' need to be specified for use in tumor report (\'control_dp_tag\' is missing)'
)
## if filtering turned on for AF-based tumor-only filtering, return error if TVAF not defined
return 0
def check_format_ad_dp_tags(vcf,
tumor_dp_tag,
tumor_af_tag,
control_dp_tag,
control_af_tag,
call_conf_tag,
exclude_hom_germline,
exclude_het_germline,
tumor_only,
logger):
"""
Function that checks whether the INFO tags specified for depth/allelic fraction are correctly formatted in the VCF header (i.e. Type)
"""
found_taf_tag = 0
found_tdp_tag = 0
found_naf_tag = 0
found_ndp_tag = 0
found_call_conf_tag = 0
annoutils.detect_reserved_info_tag(tumor_dp_tag,'tumor_dp_tag', logger)
annoutils.detect_reserved_info_tag(control_dp_tag,'control_dp_tag', logger)
annoutils.detect_reserved_info_tag(tumor_af_tag,'tumor_af_tag', logger)
annoutils.detect_reserved_info_tag(control_af_tag,'control_af_tag', logger)
annoutils.detect_reserved_info_tag(call_conf_tag,'call_conf_tag', logger)
for e in vcf.header_iter():
header_element = e.info()
if 'ID' in header_element.keys() and 'HeaderType' in header_element.keys():
if header_element['HeaderType'] == 'INFO':
if header_element['ID'] == tumor_dp_tag:
if header_element['Type'] == 'Integer':
logger.info(f'Found INFO tag for tumor variant sequencing depth (tumor_dp_tag {tumor_dp_tag}) in input VCF')
found_tdp_tag = 1
else:
err_msg = f'INFO tag for tumor variant sequencing depth (tumor_dp_tag {tumor_dp_tag}) is not correctly specified in input VCF (Type={header_element["Type"]}), should be Type=Integer'
return error_message(err_msg, logger)
if header_element['ID'] == tumor_af_tag:
if header_element['Type'] == 'Float':
logger.info(f'Found INFO tag for tumor variant allelic fraction (tumor_af_tag {tumor_af_tag}) in input VCF')
found_taf_tag = 1
else:
err_msg = f'INFO tag for tumor variant allelic fraction (tumor_af_tag {tumor_af_tag}) is not correctly specified in input VCF (Type={header_element["Type"]}), should be Type=Float'
return error_message(err_msg, logger)
if header_element['ID'] == control_dp_tag:
if header_element['Type'] == 'Integer':
logger.info(f'Found INFO tag for normal/control variant sequencing depth (control_dp_tag {control_dp_tag}) in input VCF')
found_ndp_tag = 1
else:
err_msg = f'INFO tag for normal/control variant sequencing depth (control_dp_tag {control_dp_tag}) is not correctly specified in input VCF (Type={header_element["Type"]}), should be Type=Integer'
return error_message(err_msg, logger)
if header_element['ID'] == control_af_tag:
if header_element['Type'] == 'Float':
logger.info(f'Found INFO tag for normal/control allelic fraction (control_af_tag {control_af_tag}) in input VCF')
found_naf_tag = 1
else:
err_msg = f'INFO tag for for normal/control allelic fraction (control_af_tag {control_af_tag}) is not correctly specified in input VCF (Type={header_element["Type"]}) should be Type=Float'
return error_message(err_msg, logger)
if header_element['ID'] == call_conf_tag:
if header_element['Type'] == 'String':
logger.info(f'Found INFO tag for variant call confidence (call_conf_tag {call_conf_tag}) in input VCF')
found_call_conf_tag = 1
else:
err_msg = f'INFO tag for variant call confidence (call_conf_tag) is not correctly specified in input VCF (Type={header_element["Type"]}), should be Type=String'
return error_message(err_msg, logger)
if call_conf_tag != '_NA_' and found_call_conf_tag == 0:
logger.warning(f"Could not find the specified call_conf_tag ('{call_conf_tag}') in INFO column of input VCF")
if tumor_dp_tag != '_NA_' and found_tdp_tag == 0:
logger.warning(f"Could not find the specified tumor_dp_tag ('{tumor_dp_tag}') in INFO column of input VCF")
if tumor_af_tag != '_NA_' and found_taf_tag == 0:
logger.warning(f"Could not find the specified tumor_af_tag ('{tumor_af_tag}') in INFO column of input VCF")
if control_dp_tag != '_NA_' and found_ndp_tag == 0:
logger.warning(f"Could not find the specified control_dp_tag ('{control_dp_tag}') in INFO column of input VCF")
if control_af_tag != '_NA_' and found_naf_tag == 0:
logger.warning(f"Could not find the specified control_af_tag ('{control_af_tag}') in INFO column of input VCF")
if exclude_hom_germline is True and tumor_only == 1 and found_taf_tag == 0:
logger.warning(f"Could not find the specified tumor_af_tag ('{tumor_af_tag}') in INFO column of input VCF - filtering of homozygous germline variants in tumor-only mode will be ignored")
if exclude_het_germline is True and tumor_only == 1 and found_taf_tag == 0:
logger.warning(f"Could not find the specified tumor_af_tag ('{tumor_af_tag}') in INFO column of input VCF - filtering of heterozygous germline variants in tumor-only mode will be ignored")
if found_tdp_tag == 1 and found_taf_tag == 0:
logger.warning('BOTH \' tumor_dp_tag\' AND \' tumor_af_tag\' need to be specified for use in tumor report (\'tumor_af_tag\' is missing)')
if found_tdp_tag == 0 and found_taf_tag == 1:
logger.warning('BOTH \'tumor_dp_tag\' AND \'tumor_af_tag\' need to be specified for use in tumor report (\'tumor_dp_tag\' is missing)')
if found_ndp_tag == 1 and found_naf_tag == 0:
logger.warning('BOTH \'control_dp_tag\' AND \'control_af_tag\' need to be specified for use in tumor report (\'control_af_tag\' is missing)')
if found_ndp_tag == 0 and found_naf_tag == 1:
logger.warning('BOTH \'control_dp_tag\' AND \'control_af_tag\' need to be specified for use in tumor report (\'control_dp_tag\' is missing)')
## if filtering turned on for AF-based tumor-only filtering, return error if TVAF not defined
return 0