def from_pileup_units(cls, raw_tumor_ref_units, raw_tumor_obs_units, raw_normal_ref_units, raw_normal_obs_units):
v = HeterozygousGermlineVariant()
params = settings.hetero_germline_variant
variant_key = raw_tumor_obs_units[0].key()
tumor_ref_units = pileup_unit.filter_by_base_quality(raw_tumor_ref_units, settings.base_quality_threshold)
tumor_obs_units = pileup_unit.filter_by_base_quality(raw_tumor_obs_units, settings.base_quality_threshold)
normal_ref_units = pileup_unit.filter_by_base_quality(raw_normal_ref_units, settings.base_quality_threshold)
normal_obs_units = pileup_unit.filter_by_base_quality(raw_normal_obs_units, settings.base_quality_threshold)
v.tumor_ref_count = pileup_unit.get_depth_without_indel(tumor_ref_units)
v.normal_ref_count = pileup_unit.get_depth_without_indel(normal_ref_units)
v.tumor_obs_count = len(tumor_obs_units)
v.normal_obs_count = len(normal_obs_units)
if not (v.tumor_obs_count >= settings.min_variant_supporting_reads):
raise common.TooFewVariantReadsError
if not (v.normal_obs_count >= settings.min_variant_supporting_reads):
raise common.TooFewVariantReadsError
tumor_depth = pileup_unit.get_depth_without_indel(tumor_ref_units + tumor_obs_units)
if tumor_depth < settings.min_depth:
raise common.LowDepthError
v.tumor_freq = float(v.tumor_obs_count) / tumor_depth
if not (params.min_freq <= v.tumor_freq <= params.max_freq):
raise common.AlleleFreqOutOfRangeError
v.tumor_strand = pileup_unit.calc_strand_freq(tumor_obs_units)
if not (params.min_strand_freq <= v.tumor_strand <= params.max_strand_freq):
raise common.StrandFreqOutOfRangeError
normal_depth = pileup_unit.get_depth_without_indel(normal_ref_units + normal_obs_units)
if normal_depth < settings.min_depth:
raise common.LowDepthError
v.normal_freq = float(v.normal_obs_count) / normal_depth
if not (params.min_freq <= v.normal_freq <= params.max_freq):
raise common.AlleleFreqOutOfRangeError
v.normal_strand = pileup_unit.calc_strand_freq(normal_obs_units)
if not (params.min_strand_freq <= v.normal_strand <= params.max_strand_freq):
raise common.StrandFreqOutOfRangeError
v.tumor_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_tumor_ref_units)
v.tumor_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_tumor_obs_units)
v.normal_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_normal_ref_units)
v.normal_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_normal_obs_units)
for avg_bq in [v.tumor_ref_avg_base_quality,
v.tumor_obs_avg_base_quality,
v.normal_ref_avg_base_quality,
v.normal_obs_avg_base_quality]:
if avg_bq is not None and not (params.min_avg_base_quality <= avg_bq):
raise common.LowBaseQualityError(u"low_avg_base_quality({0} < {1})".format(avg_bq, params.min_avg_base_quality))
return v
def from_pileup_units(cls, raw_tumor_ref_units, raw_tumor_obs_units, raw_normal_ref_units, raw_normal_obs_units):
v = SomaticVariant()
params = settings.somatic_variant
tumor_ref_units = pileup_unit.filter_by_base_quality(raw_tumor_ref_units, settings.base_quality_threshold)
tumor_obs_units = pileup_unit.filter_by_base_quality(raw_tumor_obs_units, settings.base_quality_threshold)
normal_ref_units = pileup_unit.filter_by_base_quality(raw_normal_ref_units, settings.base_quality_threshold)
normal_obs_units = pileup_unit.filter_by_base_quality(raw_normal_obs_units, settings.base_quality_threshold)
v.tumor_ref_count = pileup_unit.get_depth_without_indel(tumor_ref_units)
v.normal_ref_count = pileup_unit.get_depth_without_indel(normal_ref_units)
v.tumor_obs_count = len(tumor_obs_units)
v.normal_obs_count = len(normal_obs_units)
if not (v.tumor_obs_count >= settings.min_variant_supporting_reads):
raise common.TooFewVariantReadsError
if not (v.normal_obs_count <= params.max_variant_supporting_reads_normal):
raise common.TooManyNormalVariantReadsError
tumor_depth = pileup_unit.get_depth_without_indel(tumor_ref_units + tumor_obs_units)
if tumor_depth < settings.min_depth:
raise common.LowDepthError
v.tumor_freq = float(v.tumor_obs_count) / tumor_depth
if not (params.min_tumor_freq <= v.tumor_freq):
if params.sufficient_num_variant_reads is not None\
and v.tumor_obs_count < params.sufficient_num_variant_reads:
raise common.AlleleFreqOutOfRangeError
v.tumor_strand = pileup_unit.calc_strand_freq(tumor_obs_units)
if not (params.min_strand_freq <= v.tumor_strand <= params.max_strand_freq):
raise common.StrandFreqOutOfRangeError
normal_depth = pileup_unit.get_depth_without_indel(normal_ref_units + normal_obs_units)
if normal_depth < settings.min_depth:
raise common.LowDepthError
v.normal_freq = float(v.normal_obs_count) / normal_depth
if not (v.normal_freq <= params.max_normal_freq):
raise common.AlleleFreqOutOfRangeError
v.normal_strand = pileup_unit.calc_strand_freq(normal_obs_units)
v.tumor_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_tumor_ref_units)
v.tumor_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_tumor_obs_units)
v.normal_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_normal_ref_units)
v.normal_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(raw_normal_obs_units)
for avg_bq in [v.tumor_ref_avg_base_quality,
v.tumor_obs_avg_base_quality,
v.normal_ref_avg_base_quality]:
if avg_bq is not None and not (params.min_avg_base_quality <= avg_bq):
raise common.LowBaseQualityError(u"low_avg_base_quality({0} < {1}".format(avg_bq, params.min_avg_base_quality))
return v
def from_pileup_units(cls, raw_tumor_ref_units, raw_tumor_obs_units,
raw_normal_ref_units, raw_normal_obs_units):
v = HeterozygousGermlineVariant()
params = settings.hetero_germline_variant
variant_key = raw_tumor_obs_units[0].key()
tumor_ref_units = pileup_unit.filter_by_base_quality(
raw_tumor_ref_units, settings.base_quality_threshold)
tumor_obs_units = pileup_unit.filter_by_base_quality(
raw_tumor_obs_units, settings.base_quality_threshold)
normal_ref_units = pileup_unit.filter_by_base_quality(
raw_normal_ref_units, settings.base_quality_threshold)
normal_obs_units = pileup_unit.filter_by_base_quality(
raw_normal_obs_units, settings.base_quality_threshold)
v.tumor_ref_count = pileup_unit.get_depth_without_indel(
tumor_ref_units)
v.normal_ref_count = pileup_unit.get_depth_without_indel(
normal_ref_units)
v.tumor_obs_count = len(tumor_obs_units)
v.normal_obs_count = len(normal_obs_units)
if not (v.tumor_obs_count >= settings.min_variant_supporting_reads):
raise common.TooFewVariantReadsError
if not (v.normal_obs_count >= settings.min_variant_supporting_reads):
raise common.TooFewVariantReadsError
tumor_depth = pileup_unit.get_depth_without_indel(tumor_ref_units +
tumor_obs_units)
if tumor_depth < settings.min_depth:
raise common.LowDepthError
v.tumor_freq = float(v.tumor_obs_count) / tumor_depth
if not (params.min_freq <= v.tumor_freq <= params.max_freq):
raise common.AlleleFreqOutOfRangeError
v.tumor_strand = pileup_unit.calc_strand_freq(tumor_obs_units)
if not (params.min_strand_freq <= v.tumor_strand <=
params.max_strand_freq):
raise common.StrandFreqOutOfRangeError
normal_depth = pileup_unit.get_depth_without_indel(normal_ref_units +
normal_obs_units)
if normal_depth < settings.min_depth:
raise common.LowDepthError
v.normal_freq = float(v.normal_obs_count) / normal_depth
if not (params.min_freq <= v.normal_freq <= params.max_freq):
raise common.AlleleFreqOutOfRangeError
v.normal_strand = pileup_unit.calc_strand_freq(normal_obs_units)
if not (params.min_strand_freq <= v.normal_strand <=
params.max_strand_freq):
raise common.StrandFreqOutOfRangeError
v.tumor_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_tumor_ref_units)
v.tumor_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_tumor_obs_units)
v.normal_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_normal_ref_units)
v.normal_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_normal_obs_units)
for avg_bq in [
v.tumor_ref_avg_base_quality, v.tumor_obs_avg_base_quality,
v.normal_ref_avg_base_quality, v.normal_obs_avg_base_quality
]:
if avg_bq is not None and not (params.min_avg_base_quality <=
avg_bq):
raise common.LowBaseQualityError(
u"low_avg_base_quality({0} < {1})".format(
avg_bq, params.min_avg_base_quality))
return v
def from_pileup_units(cls, raw_tumor_ref_units, raw_tumor_obs_units,
raw_normal_ref_units, raw_normal_obs_units):
v = SomaticVariant()
params = settings.somatic_variant
tumor_ref_units = pileup_unit.filter_by_base_quality(
raw_tumor_ref_units, settings.base_quality_threshold)
tumor_obs_units = pileup_unit.filter_by_base_quality(
raw_tumor_obs_units, settings.base_quality_threshold)
normal_ref_units = pileup_unit.filter_by_base_quality(
raw_normal_ref_units, settings.base_quality_threshold)
normal_obs_units = pileup_unit.filter_by_base_quality(
raw_normal_obs_units, settings.base_quality_threshold)
v.tumor_ref_count = pileup_unit.get_depth_without_indel(
tumor_ref_units)
v.normal_ref_count = pileup_unit.get_depth_without_indel(
normal_ref_units)
v.tumor_obs_count = len(tumor_obs_units)
v.normal_obs_count = len(normal_obs_units)
if not (v.tumor_obs_count >= settings.min_variant_supporting_reads):
raise common.TooFewVariantReadsError
if not (v.normal_obs_count <=
params.max_variant_supporting_reads_normal):
raise common.TooManyNormalVariantReadsError
tumor_depth = pileup_unit.get_depth_without_indel(tumor_ref_units +
tumor_obs_units)
if tumor_depth < settings.min_depth:
raise common.LowDepthError
v.tumor_freq = float(v.tumor_obs_count) / tumor_depth
if not (params.min_tumor_freq <= v.tumor_freq):
if params.sufficient_num_variant_reads is not None\
and v.tumor_obs_count < params.sufficient_num_variant_reads:
raise common.AlleleFreqOutOfRangeError
v.tumor_strand = pileup_unit.calc_strand_freq(tumor_obs_units)
if not (params.min_strand_freq <= v.tumor_strand <=
params.max_strand_freq):
raise common.StrandFreqOutOfRangeError
normal_depth = pileup_unit.get_depth_without_indel(normal_ref_units +
normal_obs_units)
if normal_depth < settings.min_depth:
raise common.LowDepthError
v.normal_freq = float(v.normal_obs_count) / normal_depth
if not (v.normal_freq <= params.max_normal_freq):
raise common.AlleleFreqOutOfRangeError
v.normal_strand = pileup_unit.calc_strand_freq(normal_obs_units)
v.tumor_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_tumor_ref_units)
v.tumor_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_tumor_obs_units)
v.normal_ref_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_normal_ref_units)
v.normal_obs_avg_base_quality = pileup_unit.calc_avg_base_quality(
raw_normal_obs_units)
for avg_bq in [
v.tumor_ref_avg_base_quality, v.tumor_obs_avg_base_quality,
v.normal_ref_avg_base_quality
]:
if avg_bq is not None and not (params.min_avg_base_quality <=
avg_bq):
raise common.LowBaseQualityError(
u"low_avg_base_quality({0} < {1}".format(
avg_bq, params.min_avg_base_quality))
return v