CTNNA1 c.347G>A, p.(Cys116Tyr)

Reasoning

PM2 is met at supporting strength. The highest observed population allele frequency is 0.01349%, below the generic rare-variant threshold of 0.1%; the variant is absent from gnomAD-Canada v1.0. Its low but nonzero frequency precludes describing it as absent.

Rule

Generic population-frequency rule: PM2 supporting applies when the highest observed allele frequency is <0.1% in population reference data.

Evidence considered

• No CTNNA1 VCEP/CSPEC or local gene-specific population-frequency framework is available; generic ACMG/AMP population assessment was used.
• gnomAD v4.1 reports AF 0.00217% overall (35/1,614,002 alleles), a maximum subgroup AF of 0.01349% (4/29,654 remaining-XY alleles), and zero homozygotes.
• gnomAD v2.1 reports AF 0.00141% overall (4/282,876 alleles), a maximum population AF of 0.00310% in non-Finnish European individuals, and zero homozygotes.
• The variant was absent from gnomAD-Canada v1.0.

Still missing

• No CTNNA1-specific PM2 threshold or maximum credible allele-frequency calculation was available; the generic rare-frequency threshold was used.

Sources

Reasoning

PP1 is met at supporting strength. PMID:33435129 identifies the exact CTNNA1 c.347G>A (p.C116Y) variant and reports it across three affected members of Family 15 (I.2, II.1, and III.1), all characterized by comitant esotropia. This is compatible with segregation in a multigenerational family, but the available text does not provide a pedigree, genotypes for unaffected relatives, assay confirmation, or a quantitative segregation analysis; no higher strength is justified.

Rule

Generic ACMG/AMP 2015 PP1: co-segregation with disease in multiple affected family members; applied at supporting strength only because segregation detail is limited.

Evidence considered

• PMID:33435129 Table 1 lists the exact variant: CTNNA1 c.347G>A (p.C116Y).
• PMID:33435129 Results states: “Finally, a missense variant in CTNNA1 (Catenin α 1) was observed across multiple cases in Family 15, characterized by comitant esotropia (I.2, II.1, and III.1).”
• The observation supports only supporting-level PP1 because the report does not provide the pedigree-level information needed to count and independently verify informative meioses or calculate a segregation likelihood ratio.

Still missing

• Review the Family 15 pedigree and individual genotypes to confirm the number of informative meioses and affected carriers.
• Obtain genotype and phenotype data for unaffected relatives and confirm variant testing methodology.
• Establish the relevance and specificity of comitant esotropia to the applicable CTNNA1-associated disease context before relying heavily on this segregation evidence.

Sources

Reasoning

No VCEP/gene-specific framework exists for CTNNA1; generic ACMG applied. SpliceAI predicts no splice impact for NM_001903.5:c.347G>A (max delta score = 0.00 across all four donor/acceptor delta types), well below the 0.2 threshold established in the original SpliceAI publication for flagging splice-altering variants. This is unambiguous, well-below-threshold negative evidence supporting BP4 at supporting strength. REVEL (0.492) is indeterminate (between the calibrated BP4 supporting cutoff of <=0.290 and PP3 supporting cutoff of >=0.644 per PMID:36413997) and therefore does not independently contribute to BP4, avoiding double-counting of the missense predictor against the splice-based BP4 call. BayesDel is not usable (no verified published threshold available to this pipeline).

Rule

generic_acmg_2015

Evidence considered

• SpliceAI max delta score = 0.00 (DS_AG=0.00, DS_AL=0.00, DS_DG=0.00, DS_DL=0.00), well below the 0.2 default threshold from the original SpliceAI publication (Jaganathan et al. 2019, Cell, PMID 30661751), supports BP4 at supporting strength for absence of predicted splicing impact.
• REVEL score = 0.492 falls between the ClinGen SVI-calibrated BP4 supporting cutoff (<=0.290) and PP3 supporting cutoff (>=0.644) (Pejaver et al. 2022, PMID 36413997) -- indeterminate, not used to independently support BP4 to avoid double-counting against the splice-based call.
• BayesDel score = 0.132168 present but no verified published calibration threshold/PMID available to this pipeline, so treated as not_available for BP4 per pipeline policy.

Still missing

• REVEL score is indeterminate under calibrated thresholds and could not independently corroborate BP4 from the missense-predictor path.
• BayesDel could not be used due to lack of a verified, citable calibration threshold in this pipeline.

Sources

Reasoning

No VCEP/gene-specific framework exists for CTNNA1 (generic ACMG fallback confirmed in vcep_materials: 'no cspec url available', files_found empty). NM_001903.5:c.347G>A (p.Cys116Tyr) is a missense variant not overlapping the canonical splice consensus. SpliceAI predicts no splice impact (max delta score = 0.00, well below the 0.2 predicted-impact threshold established in the original SpliceAI publication), which does not support PP3. REVEL score is 0.492, which falls in the indeterminate zone between the ClinGen SVI-calibrated BP4 supporting cutoff (<=0.290) and PP3 supporting cutoff (>=0.644) from Pejaver et al. 2022 (PMID:36413997), so REVEL does not support PP3 either. BayesDel score (0.132168) is present but per pipeline policy no verified published calibration threshold is available for BayesDel, so it cannot be used for PP3/BP4. aGVGD was not available for this variant. With no computational predictor meeting a calibrated pathogenic threshold, PP3 is not met.

Rule

generic_acmg_2015

Evidence considered

• SpliceAI max delta score = 0.00 (evidence_sentence: 'SpliceAI predicts no significant splice impact for this variant'), far below the 0.2 default variant-detection threshold described in the original SpliceAI publication (Jaganathan et al. 2019, Cell, PMID 30661751) -- does not support PP3.
• REVEL score = 0.492 for NM_001903.5:c.347G>A, which lies between the ClinGen SVI-calibrated BP4 supporting cutoff (<=0.290) and PP3 supporting cutoff (>=0.644) published in Pejaver et al. 2022 (PMID 36413997) -- indeterminate, does not support PP3.
• BayesDel score = 0.132168 present in local predictor lookup, but no verified published calibration threshold/PMID is available to this pipeline for BayesDel, so it is treated as not_available for PP3 per pipeline policy.
• Variant does not overlap the canonical +/-1,2 splice consensus per pvs1_variant_assessment (canonical_splice_consensus: false), consistent with a purely missense computational-evidence context.

Still missing

• No calibrated BayesDel threshold with a citable publication is available to this pipeline, so BayesDel could not be evaluated for PP3 despite a score being present.
• REVEL score (0.492) falls in the indeterminate zone of the calibrated ClinGen SVI thresholds and does not clearly support or refute pathogenicity.

Sources

Reasoning

BA1 is not met. The highest observed population allele frequency is 0.01349% (4/29,654 alleles in the gnomAD v4.1 remaining-XY subgroup), far below the generic BA1 threshold of 1%.

Rule

Generic population-frequency rule: BA1 requires allele frequency >=1% in an appropriate population dataset.

Evidence considered

• No CTNNA1 VCEP/CSPEC or local gene-specific population-frequency framework is available; generic ACMG/AMP population assessment was used.
• gnomAD v4.1 reports 35/1,614,002 alleles overall (AF 0.00217%), with the highest subgroup observation in remaining-XY: 4/29,654 alleles (AF 0.01349%); no homozygotes were reported.
• gnomAD v2.1 reports 4/282,876 alleles overall (AF 0.00141%), with a highest population AF of 0.00310% in non-Finnish European individuals; no homozygotes were reported.
• The variant was absent from gnomAD-Canada v1.0.

Sources

Reasoning

BS1 is not met. The highest observed population allele frequency is 0.01349%, below the generic BS1 threshold of 0.3%; therefore frequency is not greater than expected for a CTNNA1-associated disorder.

Rule

Generic population-frequency rule: BS1 requires allele frequency >0.3% when no disease-specific maximum credible allele frequency is available.

Evidence considered

• No CTNNA1 VCEP/CSPEC or local gene-specific population-frequency framework is available; generic ACMG/AMP population assessment was used.
• gnomAD v4.1 reports the highest subgroup AF as 0.01349% (4/29,654 alleles in remaining-XY), 35/1,614,002 alleles overall, and zero homozygotes.
• gnomAD v2.1 reports 4/282,876 alleles overall (AF 0.00141%), with zero homozygotes.

Still missing

• No CTNNA1 disease-specific prevalence, penetrance, inheritance, and maximum credible allele-frequency calculation was provided; a disease-specific BS1 threshold cannot be derived from this case evidence.

Sources

Reasoning

PS1 requires a prior established pathogenic variant producing the identical amino acid change (p.Cys116Tyr) via a different nucleotide substitution. No CSPEC/VCEP exists for CTNNA1 (gene_vcep_dir null, framework_mode generic_acmg), so generic ACMG applies. The case bundle's automated same-residue comparator search (pm5_candidates.json) explicitly declined to harvest candidates ('Unable to confirm classic same-residue PM5 semantics safely; skipping automatic candidate harvesting'), so no comparator variant data of any kind (same-AA or alternate-AA at codon 116) is available in this bundle to evaluate PS1.

Evidence considered

• pm5_candidates.json reports found=false, eligible_for_classic_pm5_search=false, and reason='Unable to confirm classic same-residue PM5 semantics safely; skipping automatic candidate harvesting', meaning no same-residue comparator search was executed for this case.

Still missing

• No ClinVar/literature search for other pathogenic variants encoding p.Cys116Tyr via a different codon change was performed by the pipeline; pm5_candidates.json search was explicitly skipped.
• No functional/structural evidence comparing this residue to established pathogenic variants is present in the bundle.

Reasoning

No proband-level parental genotypes, parental identity testing, or phenotype information supporting a confirmed de novo occurrence of CTNNA1 c.347G>A were provided. Therefore, PS2 cannot be assessed under the generic ACMG/AMP framework.

Rule

Generic ACMG/AMP 2015 PS2: confirmed de novo occurrence with biological parentage established and phenotype consistent with the relevant gene-disease association.

Evidence considered

• The generic ACMG/AMP framework requires a de novo observation with both parental samples negative and identity testing for PS2; no such family-level evidence is present in the consolidated case bundle.
• The available exact-variant paper reports familial observation rather than a de novo event.

Still missing

• Obtain testing of both biological parents showing absence of c.347G>A, with parentage confirmation.
• Document the proband phenotype and its specificity for an established CTNNA1-associated condition.

Sources

Reasoning

No CTNNA1 VCEP/ClinGen CSPEC exists for this gene (case_summary.cspec.found=false, vcep_materials note='no cspec url available'), so generic ACMG/AMP applies. The consolidated evidence bundle (prefetch, evidence.json, literature_pass, source_registry) contains no validated functional/biochemical assay data for NM_001903.5:c.347G>A (p.Cys116Tyr). The only literature identified are the ACMG/AMP 2015 standards paper, a cancer predisposition referral guideline, the Sherloc refinement paper, and a familial strabismus exome-sequencing/segregation study (PMID:33435129) -- none report a validated functional assay (e.g., cell-based alpha-catenin binding/adhesion assay, splicing minigene, or biochemical characterization) for this specific variant. No assay, controls, or biological relevance data are available to assess.

Still missing

• No published functional/biochemical assay (e.g., cadherin-catenin complex binding, cell adhesion, actin-binding) data identified for p.Cys116Tyr in the bundled evidence or literature triage.
• No CTNNA1-specific VCEP functional assay specification available to define validated assay types or PS3 strength calibration.

Reasoning

No statistically robust case-control enrichment or other qualifying prevalence analysis for NM_001903.5:c.347G>A (p.Cys116Tyr) was identified.

Rule

Generic ACMG/AMP PS4 requires increased prevalence in affected individuals compared with controls, supported by appropriate statistical evidence.

Evidence considered

• PMID:33435129 studied 18 familial strabismus families (53 affected and 34 unaffected individuals) but explicitly states that its sample was insufficient for a robust gene-discovery association test; it reports an exploratory CTNNA1 family observation rather than an exact-variant case-control enrichment analysis.

Still missing

• A phenotype-defined case series or case-control dataset reporting exact-variant counts in affected individuals and appropriate controls, with a valid statistical comparison.

Sources

Reasoning

PM1 requires the variant to fall in a mutational hotspot or well-established functional/structural domain without benign variation. No CSPEC/VCEP exists for CTNNA1, so no gene-specific domain map or hotspot region definition is available. The bundle includes a cancerhotspots.org lookup (evidence.json 'hotspots' result) that returned search_status='no_result' with evidence_sentence 'This variant does not lie in a statistically significant hotspot,' but this is a somatic cancer-hotspot database, not a validated critical functional domain map for germline CTNNA1 missense mechanism, and it is not present as a citable key in source_registry.json. No other structural/domain annotation (e.g., Pfam/InterPro domain boundaries, vinculin-homology domain mapping) is present in the bundle to establish or rule out a critical domain for residue 116.

Evidence considered

• evidence.json hotspots result (cancerhotspots.org query 'CTNNA1 C116'): search_status=no_result, evidence_sentence='This variant does not lie in a statistically significant hotspot.' This somatic-cancer-hotspot source is not present in source_registry.json as a citable key and does not by itself establish or exclude a critical germline functional domain.

Still missing

• No CSPEC/VCEP-defined critical domain or hotspot region for CTNNA1 is available.
• cancerhotspots.org lookup found no significant somatic hotspot at C116, but this does not address germline structural/functional domain criticality and is not a citable source_registry entry.
• No structural domain annotation (e.g., vinculin-homology domain boundaries) for residue 116 was retrieved in this case bundle.

Reasoning

Not assessed. No affected-proband observation of NM_001903.5:c.347G>A in trans with a pathogenic or likely pathogenic CTNNA1 variant is documented in the available evidence.

Rule

Generic ACMG/AMP PM3: for a recessive disorder, the variant must be observed in trans with a pathogenic variant in an affected individual.

Evidence considered

• No CTNNA1 VCEP or other gene-specific framework was retrieved; the generic ACMG/AMP framework therefore governs this assessment.
• The exact ClinVar record is classified as uncertain significance by six single-submitters. Its available submission audit reports only PP1, PM2, and PP3 evidence signals; it documents neither a second CTNNA1 pathogenic/likely pathogenic variant nor trans phase.
• PMID:33435129 was searched in the available full text for c.347G>A, p.Cys116Tyr/C116Y, rs761084210, and CTNNA1; no exact-variant or allelic-phase evidence was located.

Still missing

• Affected-proband genotype data identifying a second CTNNA1 variant and its pathogenicity classification.
• Phase evidence demonstrating that c.347G>A and the second variant are in trans.
• A disease-specific recessive CTNNA1 context, if the proposed indication is recessive.

Sources

Reasoning

PM5 requires a previously established pathogenic missense variant at the same residue (codon 116) but with a different amino acid substitution than Tyr. The pipeline's dedicated PM5 comparator harvesting artifact (pm5_candidates.json) reports zero candidates and explicitly states classic same-residue PM5 semantics could not be safely confirmed, so automatic candidate harvesting was skipped entirely ('recommendation.level': 'not_applicable', hermes_instruction to not apply PM5 from this artifact alone). No other source in the bundle (ClinVar record, literature_pass.json, vcep_materials.json) provides data on alternate pathogenic substitutions at Cys116.

Evidence considered

• pm5_candidates.json: found=false, same_residue_candidates=0, recommendation.level='not_applicable', reason='Unable to confirm classic same-residue PM5 semantics safely; skipping automatic candidate harvesting.' No alternate amino-acid substitutions at residue 116 were identified or excluded.

Still missing

• PM5 comparator candidate search (pm5_candidates.json) was skipped by the pipeline rather than executed, so codon-116 alternate-substitution pathogenicity data is entirely absent from this bundle.
• A manual ClinVar/literature search for other Cys116 substitutions (e.g., p.Cys116Ser, p.Cys116Arg, p.Cys116Phe) has not been performed and is needed before PM5 can be assessed.

Sources

Reasoning

No proband-level parental testing is reported that would establish an assumed de novo occurrence of CTNNA1 c.347G>A. PM6 is therefore not assessable.

Rule

Generic ACMG/AMP 2015 PM6: de novo occurrence with parental absence but without confirmed biological parentage; phenotype consistency remains required.

Evidence considered

• Under generic ACMG/AMP guidance, PM6 is the de novo evidence category when parentage is assumed but not confirmed; the bundle provides no parental genotypes or de novo observation.
• The exact-variant report describes familial occurrence, not an apparently de novo proband.

Still missing

• Obtain parental genotypes showing the variant is absent in both parents when biological parentage cannot be confirmed.
• Document a phenotype in the proband consistent with an established CTNNA1-associated condition.

Sources

Reasoning

PP2 requires the variant to occur in a gene where missense variants are a common disease mechanism and where benign missense variation is rare. No CSPEC/VCEP exists for CTNNA1 (framework_mode: generic_acmg), and no gene-level missense constraint metric (e.g., gnomAD missense Z-score/observed:expected ratio, ClinGen gene-disease validity/mechanism statement) is present anywhere in this case bundle (case_summary.json, evidence.json, vcep_materials.json all lack such content). CTNNA1 is associated with more than one distinct phenotype (per ClinVar conditions: hereditary cancer-predisposing syndrome and patterned macular dystrophy 2) which may have different underlying mechanisms (loss-of-function vs. missense), but the bundle does not contain data establishing which mechanism predominates for either condition.

Evidence considered

• case_summary.json ClinVar api_conditions list associates CTNNA1 with 'Hereditary cancer-predisposing syndrome' and 'Patterned macular dystrophy 2,' but no gene-mechanism or missense-constraint data accompanies these condition labels in this bundle.

Still missing

• No gene-level missense constraint statistic (e.g., gnomAD missense Z-score, ExAC/gnomAD constraint metrics) is present in the bundle.
• No CSPEC/VCEP statement on CTNNA1 missense disease mechanism is available.
• ClinVar lists multiple associated conditions (hereditary cancer-predisposing syndrome; patterned macular dystrophy 2) with potentially different missense-vs-LOF mechanisms, and the bundle does not disambiguate which applies to this variant's clinical context.

Reasoning

The available observation of familial comitant esotropia is not sufficiently specific for a CTNNA1-associated disorder to support PP4.

Rule

Generic ACMG/AMP PP4 requires a patient's phenotype or family history to be highly specific for a disease with a single genetic etiology.

Evidence considered

• The exact matched ClinVar record contains only uncertain-significance submissions and does not provide a phenotype-specific, expert-panel assertion for this variant.
• PMID:33435129 describes a CTNNA1 missense variant in three affected members of one family with comitant esotropia, but the authors characterize the study as small and describe genetic heterogeneity and weak/partial evidence; this does not establish a highly specific CTNNA1 phenotype.

Still missing

• A well-characterized phenotype in the tested individual and evidence that this phenotype is highly specific for CTNNA1-related disease, with relevant differential diagnoses excluded.

Sources

Reasoning

PP5 cannot be applied because no exact-variant ClinVar Pathogenic/Likely pathogenic expert-panel assertion is present.

Rule

Apply PP5 only for an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification.

Evidence considered

• The ClinVar submission audit for the exact matched variant reports zero expert-panel submissions; all six listed laboratory submissions are uncertain significance and are not eligible for PP5.

Still missing

• An exact-variant ClinVar expert-panel submission classifying the variant as Pathogenic or Likely pathogenic.

Sources

Reasoning

BS2 cannot be assessed. Although the variant is observed in population datasets, gnomAD v2.1 and v4.1 report zero homozygotes, and the supplied population records do not establish well-phenotyped, unaffected adult carriers at an age informative for CTNNA1-associated disease or its penetrance.

Rule

Generic ACMG/AMP BS2 requires observation in healthy adult individual(s) for a disorder expected to be fully penetrant at an early age; this condition is not documented here.

Evidence considered

• gnomAD v4.1 contains 35 alternate alleles and zero homozygotes; gnomAD v2.1 contains 4 alternate alleles and zero homozygotes.
• Population database inclusion alone does not document unaffected status or age at evaluation and is insufficient for BS2 in this case.

Still missing

• Need age- and phenotype-ascertained unaffected carrier data, with disease-specific penetrance and age-of-onset context, to determine whether observations in healthy adults support BS2.

Sources

Reasoning

No CTNNA1 VCEP/ClinGen CSPEC exists for this gene (case_summary.cspec.found=false), so generic ACMG/AMP applies. No published functional assay data (validated or otherwise) demonstrating normal/wild-type-like activity for NM_001903.5:c.347G>A (p.Cys116Tyr) were found in the consolidated evidence bundle or the triaged literature (PMID:25741868, PMID:25394175, PMID:28492532 are guideline/methodology papers; PMID:33435129 is a segregation-only exome study with no functional assay). No assay evidence is available to support a benign functional finding.

Still missing

• No functional assay data (of any result direction) identified for this variant to evaluate for BS3.

Reasoning

No informative non-segregation is reported. The available family report describes the exact variant in multiple affected relatives and does not document affected non-carriers or unaffected carriers that could support BS4.

Rule

Generic ACMG/AMP 2015 BS4: lack of segregation in affected members of a family; requires informative discordance rather than absence of reported segregation data.

Evidence considered

• PMID:33435129 reports the variant across affected Family 15 members but supplies no documented genotype-phenotype discordance.
• No other source in the consolidated bundle provides informative family non-segregation data for this exact variant.

Still missing

• Obtain pedigree-level genotype and phenotype data to identify informative affected non-carriers or unaffected carriers, accounting for age-dependent penetrance and phenotype ascertainment.

Sources

Reasoning

BP1 requires the variant to occur in a gene where truncating/loss-of-function variants are the primary established disease mechanism and missense variants are not commonly pathogenic. No CSPEC/VCEP exists for CTNNA1, and the bundle contains no gene-mechanism statement or missense-constraint data to confirm or refute this. Given the same gap as PP2 (no gene-level missense-vs-LOF mechanism evidence in the bundle), BP1 cannot be assessed without additional data.

Still missing

• No CSPEC/VCEP or gene-level mechanism statement establishing that CTNNA1 disease is driven primarily by truncating/LOF variants (as opposed to missense) is present in this bundle.
• No missense constraint metrics were retrieved to support or refute BP1.

Reasoning

Not assessed. The available evidence does not document c.347G>A in cis with a pathogenic/likely pathogenic CTNNA1 variant, nor in trans with such a variant in an individual with a fully penetrant dominant CTNNA1 disorder.

Rule

Generic ACMG/AMP BP2: observe the variant in trans with a pathogenic variant for a fully penetrant dominant disorder, or in cis with a pathogenic variant in any inheritance pattern.

Evidence considered

• No CTNNA1 VCEP or other gene-specific framework was retrieved; the generic ACMG/AMP framework therefore governs this assessment.
• The exact ClinVar record is classified as uncertain significance by six single-submitters. Its available submission audit contains no cis/trans observation with a CTNNA1 pathogenic or likely pathogenic variant.
• PMID:33435129 was searched in the available full text for c.347G>A, p.Cys116Tyr/C116Y, rs761084210, and CTNNA1; no exact-variant or allelic-phase evidence was located.

Still missing

• Genotype and phase data for any co-occurring CTNNA1 pathogenic/likely pathogenic variant.
• For a trans observation, phenotype and penetrance information sufficient to establish a fully penetrant dominant CTNNA1 disorder context.

Sources

Reasoning

No case-level evidence establishes that the tested phenotype is fully explained by an independent molecular diagnosis.

Rule

Generic ACMG/AMP BP5 requires a variant to be found in a case with an alternate molecular basis for disease.

Still missing

• Case-level phenotype and molecular work-up documenting an alternative pathogenic molecular diagnosis that fully explains the observed phenotype.

Reasoning

BP6 cannot be applied because no exact-variant ClinVar Benign/Likely benign expert-panel assertion is present.

Rule

Apply BP6 only for an exact-variant ClinVar expert-panel Benign or Likely benign classification.

Evidence considered

• The ClinVar submission audit for the exact matched variant reports zero expert-panel submissions; ordinary laboratory submissions and aggregate labels are not eligible for BP6.

Still missing

• An exact-variant ClinVar expert-panel submission classifying the variant as Benign or Likely benign.

Sources