Permit Het-Female-X variants in Biallelic disease (#432)

* allow for x-linked imprinting assessment

* Bump version: 5.3.1 → 5.4.0

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 5.3.1
+current_version = 5.4.0
 commit = True
 tag = False
 

.github/workflows/index_page_builder.yaml

@@ -16,7 +16,7 @@ jobs:
     runs-on: ubuntu-latest
 
     env:
-      VERSION: 5.3.1
+      VERSION: 5.4.0
 
     steps:
 

setup.py

@@ -27,7 +27,7 @@ def read_reqs(filename: str) -> list[str]:
     name='talos',
     description='Centre for Population Genomics Variant Prioritisation',
     long_description=readme,
-    version='5.3.1',
+    version='5.4.0',
     author='Matthew Welland, CPG',
     author_email='[email protected], [email protected]',
     package_data={'talos': ['templates/*.jinja', 'example_config.toml']},

src/talos/moi_tests.py

@@ -107,6 +107,7 @@ def __init__(self, pedigree: Pedigree, target_moi: str):
                 XRecessiveMale(pedigree=pedigree),
                 XRecessiveFemaleHom(pedigree=pedigree),
                 XRecessiveFemaleCH(pedigree=pedigree),
+                XPseudoDominantFemale(pedigree=pedigree),
             ]
 
         else:
@@ -202,7 +203,7 @@ def check_familial_inheritance(self, sample_id: str, called_variants: set[str],
 
         this_member = self.pedigree.by_id[sample_id]
 
-        # iterate through all family members, no interested in directionality of relationships at the moment
+        # check for valid inheritance within the immediate trio, if possible
         for member_id in [this_member.father, this_member.mother]:
             if member_id is None:
                 continue
@@ -673,6 +674,110 @@ def run(
         return classifications
 
 
+class XPseudoDominantFemale(BaseMoi):
+    """
+    X-Dominant method which only evaluates females, on the basis that a healthy allele could be inactivated
+    A special case of X-Dominant, where we consider Het. variants, used to assess genes which are on X and in genes
+    associated with a Biallelic MOI
+
+    Basically a Dominant MOI to be applied to Recessive genes, and results will be labelled as cautionary
+    """
+
+    def __init__(self, pedigree: Pedigree, applied_moi: str = 'X_Dominant'):
+        """
+        accept male hets and homs, and female hets without support
+
+        Args:
+            pedigree ():
+            applied_moi ():
+        """
+        self.ad_threshold = config_retrieve(['ValidateMOI', GNOMAD_RARE_THRESHOLD])
+        self.ac_threshold = config_retrieve(['ValidateMOI', GNOMAD_AD_AC_THRESHOLD])
+        self.hom_threshold = config_retrieve(['ValidateMOI', GNOMAD_DOM_HOM_THRESHOLD])
+
+        self.freq_tests = {
+            SmallVariant.__name__: {key: self.hom_threshold for key in INFO_HOMS}
+            | {'gnomad_ac': self.ac_threshold, 'gnomad_af': self.ad_threshold},
+            StructuralVariant.__name__: {key: self.hom_threshold for key in SV_HOMS},
+        }
+
+        super().__init__(pedigree=pedigree, applied_moi=applied_moi)
+
+    def run(
+        self,
+        principal: VARIANT_MODELS,
+        comp_het: CompHetDict | None = None,  # noqa: ARG002
+        partial_pen: bool = False,
+    ) -> list[ReportVariant]:
+        """
+        if variant is present and sufficiently rare, we take it
+        discarded if support
+
+        Args:
+            principal ():
+            comp_het ():
+            partial_pen ():
+        """
+
+        _unused = partial_pen
+
+        classifications = []
+
+        if principal.support_only:
+            return classifications
+
+        # never apply dominant MOI to support variants
+        # more stringent Pop.Freq checks for dominant - hemi restriction
+        if not self.check_frequency_passes(principal.info, self.freq_tests[principal.__class__.__name__]):
+            return classifications
+
+        # all females which have a variant call
+        females_under_consideration = {sam for sam in principal.het_samples if self.pedigree.by_id[sam].sex == '2'}
+        all_with_variant = principal.het_samples.union(principal.hom_samples)
+        for sample_id in females_under_consideration:
+            # skip primary analysis for unaffected members
+            # we require this specific sample to be categorised
+            # force minimum depth
+            if (
+                not (
+                    principal.sample_category_check(sample_id, allow_support=False)
+                    and self.pedigree.by_id[sample_id].affected == '2'
+                )
+            ) or principal.check_read_depth(sample_id, self.minimum_depth, principal.info.get('categoryboolean1')):
+                continue
+
+            # check if this is a candidate for dominant inheritance
+            # as we're allowing for flexible 'penetrance' in females, we send all het and hom variants, but allow for a
+            # partial penetrance check - the participants can have the variant, but not be affected. They may not be
+            # affected without the variant call.
+            # There's a slight breakdown here as the males should be interpreted under a full penetrance model, and
+            # females under partial penetrance, but that's not trivial without creating a second familial check method.
+            # Leaving that aside now as the current implementation is pretty central to the algorithm. Will revisit if
+            # this is noisy.
+            if not self.check_familial_inheritance(
+                sample_id=sample_id,
+                called_variants=all_with_variant,
+                partial_pen=True,
+            ):
+                continue
+
+            classifications.append(
+                ReportVariant(
+                    sample=sample_id,
+                    family=self.pedigree.by_id[sample_id].family,
+                    gene=principal.info.get('gene_id'),
+                    var_data=principal,
+                    categories=principal.category_values(sample_id),
+                    reasons={self.applied_moi},
+                    genotypes=self.get_family_genotypes(variant=principal, sample_id=sample_id),
+                    flags=principal.get_sample_flags(sample_id)
+                    | {'Affected female with heterozygous variant in XLR gene'},
+                    independent=True,
+                ),
+            )
+        return classifications
+
+
 class XRecessiveMale(BaseMoi):
     """
     accept all male non-ref GTs - male variants HOM because GATK

src/talos/version.py

@@ -3,4 +3,4 @@
 """
 
 # Do not edit this file manually
-__version__ = '5.3.1'
+__version__ = '5.4.0'

test/test_moi_tests.py

@@ -15,6 +15,7 @@
     RecessiveAutosomalCH,
     RecessiveAutosomalHomo,
     XDominant,
+    XPseudoDominantFemale,
     XRecessiveFemaleCH,
     XRecessiveFemaleHom,
     XRecessiveMale,
@@ -434,6 +435,34 @@ def test_x_dominant_info_fails(info, pedigree_path):
     assert len(x_dom.run(passing_variant)) == 0
 
 
+def test_x_dominant_female_inactivation_passes(pedigree_path):
+    """
+    check that a het female, but not a het male, is accepted as dominant
+    current test implementation doesn't include family consideration
+    """
+    info_dict = {
+        'gnomad_af': 0.0001,
+        'gnomad_ac': 0,
+        'gnomad_hom': 0,
+        'gene_id': 'TEST1',
+        'categoryboolean1': True,
+    }
+    passing_variant = SmallVariant(
+        depths={'female': 100, 'male': 100},
+        info=info_dict,
+        het_samples={'female', 'male'},
+        hom_samples={'male'},
+        boolean_categories=['categoryboolean1'],
+        coordinates=TEST_COORDS_X_1,
+        transcript_consequences=[],
+    )
+    x_dom = XPseudoDominantFemale(pedigree=make_flexible_pedigree(pedigree_path))
+    results = x_dom.run(passing_variant)
+    assert len(results) == 1
+    assert results[0].reasons == {'X_Dominant'}
+    assert 'Affected female with heterozygous variant in XLR gene' in results[0].flags
+
+
 def test_x_recessive_male_hom_passes(pedigree_path):
     passing_variant = SmallVariant(
         hom_samples={'female', 'male'},