def addQuality(dtree):
dtree.addComment("0. Check sequencing quality")
dtree.addCondition(ConditionMaker.condNum("Proband_GQ", [None, 19]),
decision=False)
dtree.addCondition(ConditionMaker.condNum("FS", [31, None]),
decision=False)
dtree.addCondition(ConditionMaker.condNum("QD", [None, 3]), decision=False)
def defineBGMTree(cond_env):
dtree = DecisionTree(cond_env)
addQuality(dtree)
dtree.addComment("Exclude common variants")
dtree.addCondition(ConditionMaker.condNum("gnomAD_AF", [0.001, None]),
decision=False)
dtree.addComment("Exclude low impact variants")
dtree.addCondition(ConditionMaker.condNum("Severity", [None, 0]),
decision=False)
dtree.setFinalDecision(True)
return dtree
def __init__(self, xl_ds, descr):
XL_Unit.__init__(self, xl_ds, descr)
if descr.get("family") and self.getDS().getFamilyInfo() is None:
self.getDS()._setFamilyInfo(descr["family"])
self.mIsOK = (self.getDS().getFamilyInfo() is not None
and len(self.getDS().getFamilyInfo()) > 1)
self.mLabels = AnfisaConfig.configOption("zygosity.cases")
self.mConfig = descr.get("config", dict())
self.mXCondition = XL_Condition.parse(
self.mConfig.get("x_cond",
ConditionMaker.condEnum("Chromosome", ["chrX"])))
def setup(self):
self.mXCondition = self.getDS().getCondEnv().parse(
self.mConfig.get("x_cond",
ConditionMaker.condEnum("Chromosome", ["chrX"])))
def defineFilterSchema():
filters = FilterPrepareSetH()
with filters.viewGroup("Inheritance"):
filters.zygositySpecialUnit(
"Custom",
"/data/zygosity",
config={"x_cond": ConditionMaker.condEnum("Chromosome", ["chrX"])})
filters.presenceUnit("Compound_heterozygous",
[("True", "/_filters/nowhere")])
filters.multiStatusUnit("Callers",
"/view/bioinformatics/called_by[]",
title="Called by",
research_only=False)
filters.multiStatusUnit("Has_Variant", "/_filters/has_variant[]")
with filters.viewGroup("Variant"):
filters.statusUnit("Variant_Class", "/data/variant_class")
filters.statusUnit(
"Most_Severe_Consequence",
"/data/most_severe_consequence",
variants=[
"transcript_ablation", "splice_acceptor_variant",
"splice_donor_variant", "stop_gained", "frameshift_variant",
"stop_lost", "start_lost", "transcript_amplification",
"inframe_insertion", "inframe_deletion", "missense_variant",
"protein_altering_variant",
"incomplete_terminal_codon_variant", "stop_retained_variant",
"synonymous_variant", "splice_region_variant",
"coding_sequence_variant", "mature_miRNA_variant",
"5_prime_UTR_variant", "3_prime_UTR_variant",
"non_coding_transcript_exon_variant", "intron_variant",
"NMD_transcript_variant", "non_coding_transcript_variant",
"upstream_gene_variant", "downstream_gene_variant",
"TFBS_ablation", "TFBS_amplification",
"TF_binding_site_variant", "regulatory_region_ablation",
"regulatory_region_amplification", "feature_elongation",
"regulatory_region_variant", "feature_truncation",
"intergenic_variant", "undefined"
],
default_value="undefined")
filters.multiStatusUnit("Genes",
"/view/general/genes[]",
compact_mode=True)
with filters.viewGroup("Coordinates"):
filters.statusUnit("Chromosome",
"/_filters/chromosome",
variants=[
"chr1", "chr2", "chr3", "chr4", "chr5", "chr6",
"chr7", "chr8", "chr9", "chr10", "chr11",
"chr12", "chr13", "chr14", "chr15", "chr16",
"chr17", "chr18", "chr19", "chr20", "chr21",
"chr22", "chr23", "chrX", "chrY", "undefined"
],
research_only=False,
default_value="undefined")
filters.intValueUnit("Start_Pos",
"/data/start",
title="Start Position",
research_only=False,
render_mode="neighborhood",
default_value=sys.maxint)
filters.intValueUnit("End_Pos",
"/data/end",
title="End Position",
research_only=False,
default_value=0,
render_mode="neighborhood")
filters.intValueUnit(
"Dist_from_Exon",
"/_filters/dist_from_exon",
title="Distance From Intron/Exon Boundary (Canonical)",
research_only=False,
default_value=0,
render_mode="log,<")
filters.statusUnit("Region",
"/data/region_canonical",
title="Region (Canonical)",
research_only=False,
default_value="Other")
with filters.viewGroup("gnomAD"):
filters.floatValueUnit("gnomAD_AF",
"/_filters/gnomad_af_fam",
diap=(0., 1.),
default_value=0.,
title="gnomAD Allele Frequency (family)",
render_mode="log,<")
filters.floatValueUnit("gnomAD_AF_Exomes",
"/_filters/gnomad_db_exomes_af",
diap=(0., 1.),
default_value=0.,
title="gnomAD Exome Allele Frequency (family)",
render_mode="log,<")
filters.floatValueUnit("gnomAD_AF_Genomes",
"/_filters/gnomad_db_genomes_af",
diap=(0., 1.),
default_value=0.,
title="gnomAD Genome Allele Frequency (family)",
render_mode="log,<")
filters.floatValueUnit("gnomAD_AF_Proband",
"/_filters/gnomad_af_pb",
diap=(0., 1.),
default_value=0.,
title="gnomAD Allele Frequency (proband)",
render_mode="log,<")
filters.floatValueUnit("gnomAD_PopMax_AF",
"/_filters/gnomad_popmax_af",
diap=(0., 1.),
default_value=0.,
title="gnomAD PopMax Allele Frequency",
render_mode="log,<")
filters.statusUnit("gnomAD_PopMax",
"/_filters/gnomad_popmax",
default_value="None",
title="gnomAD PopMax Ancestry")
filters.intValueUnit(
"gnomAD_PopMax_AN",
"/_filters/gnomad_popmax_an",
default_value=0,
title="gnomAD: Number of alleles in PopMax Ancestry",
render_mode="log,>")
filters.intValueUnit("gnomAD_Hom",
"/_filters/gnomad_hom",
default_value=0,
title="gnomAD: Number of homozygous",
render_mode="log,>")
filters.intValueUnit("gnomAD_Hem",
"/_filters/gnomad_hem",
default_value=0,
title="gnomAD: Number of hemizygous",
render_mode="log,>")
with filters.viewGroup("Databases"):
filters.presenceUnit(
"Presence_in_Databases",
[("ClinVar", "/view/databases/clinVar"),
("LMM", "/view/databases/lmm_significance"),
("GeneDx", "/view/databases/gene_dx_significance"),
("GnomAD", "/_filters/gnomad_af_fam"),
("HGMD", "/view/databases/hgmd_pmids[]"),
("OMIM", "/view/databases/omim")],
title="Presence in Databases")
filters.multiStatusUnit("ClinVar_Submitters",
"/view/databases/clinVar_submitters[]",
title="ClinVar Submitters",
compact_mode=True)
# filters.multiStatusUnit("beacons",
# "/data/beacon_names",
# title = "Observed at")
with filters.viewGroup("Call_Quality"):
filters.floatValueUnit("Proband_GQ",
"/_filters/proband_gq",
title="Genotype Quality (GQ) for Proband",
render_mode="linear,>",
default_value=float('inf'))
filters.floatValueUnit("Min_GQ",
"/_filters/min_gq",
title="Minimum GQ for the family)",
render_mode="linear,>",
default_value=float('inf'))
filters.floatValueUnit("QD",
"/_filters/qd",
title="Quality by Depth",
render_mode="linear,>",
default_value=float('inf'))
filters.floatValueUnit("FS",
"/_filters/fs",
"Fisher Strand Bias",
render_mode="linear,<",
default_value=0.)
filters.multiStatusUnit("FT", "/_filters/filters[]", title="FILTER")
with filters.viewGroup("Predictions"):
filters.statusUnit(
"Clinvar_Benign",
"/_filters/clinvar_benign",
default_value="Not in ClinVar",
title="Categorized Benign in ClinVar by all submitters")
filters.statusUnit(
"Clinvar_Trusted_Benign",
"/_filters/clinvar_trusted_benign",
default_value="No data",
title="Categorized Benign by Clinvar Trusted Submitters")
filters.statusUnit("HGMD_Benign",
"/_filters/hgmd_benign",
title="Categorized Benign in HGMD",
default_value="Not in HGMD")
filters.multiStatusUnit("HGMD_Tags",
"/view/databases/hgmd_tags[]",
default_value="None")
filters.multiStatusUnit("ClinVar_Significance",
"/data/clinvar_significance[]",
title="Clinical Significance in CLinVar")
filters.multiStatusUnit("LMM_Significance",
"/data/lmm",
title="Clinical Significance by LMM")
filters.multiStatusUnit("GeneDx_Significance",
"/data/gene_dx",
title="Clinical Significance by GeneDx")
filters.multiStatusUnit("Polyphen", "/view/predictions/polyphen[]")
filters.multiStatusUnit("SIFT", "/view/predictions/sift[]")
filters.multiStatusUnit("Polyphen_2_HVAR",
"/view/predictions/polyphen2_hvar[]",
separators="[\s\,]",
default_value="undef")
filters.multiStatusUnit("Polyphen_2_HDIV",
"/view/predictions/polyphen2_hdiv[]",
separators="[\s\,]",
default_value="undef")
with filters.viewGroup("Debug_Info"):
filters.intValueUnit("Severity",
"/_filters/severity",
research_only=True,
default_value=-1)
return filters
def defineDefaultDecisionTree():
dtree = DecisionTree()
dtree.addComment("0. Quality check")
dtree.addCondition(ConditionMaker.condNum("Proband_GQ", [None, 19]),
decision=False)
dtree.addCondition(ConditionMaker.condNum("FS", [31, None]),
decision=False)
dtree.addCondition(ConditionMaker.condNum("QD", [None, 3]), decision=False)
dtree.addComment('1. Present in HGMD as "DM"')
dtree.addCondition(ConditionMaker.condEnum("HGMD_Tags", ["DM"]),
decision=True)
dtree.addComment(
'2. AF < 5% AND +/- bases from intronic/exonic border')
dtree.addCondition(ConditionMaker.condNum("gnomAD_AF", [0.05, None]),
decision=False)
dtree.addCondition([
"and",
ConditionMaker.condNum("Dist_from_Exon", [6, None]),
ConditionMaker.condEnum("Region", ["exon"], "NOT")
],
decision=False)
dtree.addComment(
'2.a. Present in ClinVar Path, Likely Path, VUS (worst annotation).')
dtree.addCondition([
"and",
ConditionMaker.condEnum("Clinvar_Benign", ["True"]),
ConditionMaker.condEnum("Clinvar_Trusted_Benign", ["False", "No data"])
],
decision=True)
dtree.addComment('2.b. All de novo variants')
dtree.addCondition(ConditionMaker.condEnum("Callers",
["BGM_BAYES_DE_NOVO"]),
decision=True)
dtree.addComment('2.c. All potential LOF variants '
'(stop-codon, frameshift, canonical splice site).')
dtree.addCondition(ConditionMaker.condNum("Severity", [3, None]),
decision=True)
dtree.addComment('3.a. annotated as "Missense", "synonymous" '
'and "splice region" variants')
dtree.addCondition(ConditionMaker.condNum("Severity", [None, 0]),
decision=False)
dtree.addComment('3. AF < 0.0007 (GnomAD Overall) AND +/- 5 bases')
dtree.addComment(
'PopMax < 0.01 (minimum 2000 alleles total in ancestral group)')
dtree.addCondition([
"and",
ConditionMaker.condNum("gnomAD_AF", [None, 0.0007]),
ConditionMaker.condNum("gnomAD_PopMax_AN", [2001, None]),
ConditionMaker.condNum("gnomAD_PopMax_AF", [None, 0.01])
],
decision=True)
dtree.setFinalDecision(False)
return dtree
def defineHearingLossTree(cond_env):
dtree = DecisionTree(cond_env)
addQuality(dtree)
dtree.addComment('1. Include if present in HGMD as "DM"')
dtree.addCondition(ConditionMaker.condEnum("HGMD_Tags", ["DM"]),
decision=True)
dtree.addComment(
'2. Exclude unless AF < 5% AND +/- bases from intronic/exonic border'
)
dtree.addCondition(ConditionMaker.condNum("gnomAD_AF", [0.05, None]),
decision=False)
dtree.addCondition([
"and",
ConditionMaker.condNum("Dist_from_Exon", [6, None]),
ConditionMaker.condEnum("Region", ["exon"], "NOT")
],
decision=False)
dtree.addComment(
'2.a. Include if present in ClinVar Path, Likely Path, VUS (worst annotation), unless annotated benign by trusted submitter'
)
dtree.addCondition([
"and",
ConditionMaker.condEnum("Clinvar_Benign", ["True"]),
ConditionMaker.condEnum("Clinvar_Trusted_Benign", ["False", "No data"])
],
decision=True)
dtree.addComment('2.b. Include all de-novo variants')
dtree.addCondition(ConditionMaker.condEnum("Callers",
["BGM_BAYES_DE_NOVO"]),
decision=True)
dtree.addComment('2.c. Include all potential LOF variants '
'(stop-codon, frameshift, canonical splice site).')
dtree.addCondition(ConditionMaker.condNum("Severity", [3, None]),
decision=True)
dtree.addComment('3.a. For the rest: exclude "Missense", "synonymous" '
'and "splice region" variants')
dtree.addCondition(ConditionMaker.condNum("Severity", [None, 0]),
decision=False)
dtree.addComment(
'3. Include: AF < 0.0007 (GnomAD Overall) AND +/- 5 bases')
dtree.addComment(
'And: PopMax < 0.01 (minimum 2000 alleles total in ancestral group)')
dtree.addCondition([
"and",
ConditionMaker.condNum("gnomAD_AF", [None, 0.0007]),
ConditionMaker.condNum("gnomAD_PopMax_AN", [2001, None]),
ConditionMaker.condNum("gnomAD_PopMax_AF", [None, 0.01])
],
decision=True)
dtree.setFinalDecision(False)
return dtree