FLT3 c.2515G>A, p.(Asp839Asn)

Reasoning

PM2 is met at Supporting strength because the variant is absent from gnomAD v2.1 and gnomAD-Canada v1.0 and is exceptionally rare in gnomAD v4.1, with 1/1,614,010 alleles observed (AF 6.19575e-07), zero homozygotes, and a maximum reported population AF of 1.60051e-05. These observations are below the non-VCEP PM2 rare-frequency threshold of <0.1% and are compatible with a rare disease-associated allele.

Rule

Generic ACMG/AMP PM2 rule as operationalized for this non-VCEP case: apply Supporting PM2 when the variant is absent or below the rare-frequency threshold in reference populations; the non-VCEP operating threshold used here is <0.1%.

Evidence considered

• gnomAD v2.1: variant absent.
• gnomAD-Canada v1.0: variant absent.
• gnomAD v4.1: 1 alternate allele among 1,614,010 total alleles, AF 6.195748477394811e-07, zero homozygotes; highest observed population AF 1.6005121638924455e-05 in Remaining individuals, also with zero homozygotes.
• The total and maximum population frequencies are below the non-VCEP PM2 operating threshold of <0.1%; no ancestry-specific frequency or homozygote signal weakens PM2.

Sources

Reasoning

No variant-specific functional/wet-lab assay evidence is available for FLT3 NM_004119.2:c.2515G>A (p.Asp839Asn). This is a single-nucleotide missense substitution, not an ITD or in-frame activating length mutation, so it falls outside the type scope (inframe/ins/delins/duplication/itd/length mutation) declared for the local FLT3 ITD hotspot/OncoKB custom framework files, and the framework's PS3 rule text is explicitly written for 'an exact FLT3 ITD / activating length-mutation variant' with an OncoKB page reporting Oncogenic/Likely Oncogenic together with Gain-of-function/Likely Gain-of-function. OncoKB was queried at the exact protein-event page (D839N) and explicitly did not return variant-specific reviewed functional evidence: is_variant_specific=false, not_reviewed=true, and biological_effect is empty (no Gain-of-function/Likely Gain-of-function assertion for this exact substitution). Only gene-level curated context (FLT3 recurrently altered in AML) was returned. No PMIDs were identified by the literature pass for this variant (total_pmids=0), the publications directory for this case is empty, and no functional assay data (e.g., kinase autophosphorylation, BaF3 transformation, luciferase reporter) specific to D839N was located in any bundled source. Under both the local FLT3 framework's own 'Not met' branch (class-level literature only, no exact-variant OncoKB/functional support) and generic ACMG/AMP PS3 (which requires a well-established functional study demonstrating a damaging effect), PS3 is not met for this exact variant.

Rule

Local FLT3 custom framework PS3 rule (flt3_oncokb_guidance.md / flt3_itd_hotspot_and_function.md): PS3_Supporting requires exact-variant OncoKB Oncogenic/Likely Oncogenic + Gain-of-function/Likely Gain-of-function; otherwise not met. Criterion not met here because OncoKB support is not variant-specific and no functional assay data exists.

Evidence considered

• OncoKB variant page for FLT3 D839N (queried at https://www.oncokb.org/gene/FLT3/D839N) returned found=true but is_variant_specific=false and not_reviewed=true, with classification 'Likely Oncogenic' applied at the gene/summary level only ('FLT3, a receptor tyrosine kinase, is recurrently altered in acute myeloid leukemia and other hematologic malignancies'); no biological_effect (Gain-of-function/Loss-of-function) was returned for the exact substitution.
• Local FLT3 custom framework (flt3_oncokb_guidance.md, flt3_itd_hotspot_and_function.md) explicitly scopes its PS3_Supporting rule to ITDs and in-frame activating length mutations, and states PS3 should not be automatically triggered when exact-variant OncoKB support is absent -- only class-level literature is then available and PS3 remains not met.
• literature_pass.json reports total_pmids=0 and 'No PMIDs found across evidence sources' for this variant; the case publications/ directory is empty, confirming no full-text functional papers were available to review for PS3.

Still missing

• No variant-specific (D839N) functional assay publication identified (e.g., kinase activity, autophosphorylation, BaF3/32D transformation, luciferase reporter assay)
• OncoKB has no variant-specific Oncogenic/Likely Oncogenic + Gain-of-function/Likely Gain-of-function assertion for this exact substitution
• Literature search returned zero PMIDs for this case; no full-text papers were available to review for D839N-specific functional data
• Local FLT3 custom framework's PS3 rule is scoped to ITD/activating length-mutation variant types, not single-residue missense substitutions, so its supporting-strength pathway does not apply to this variant

Sources

Reasoning

NM_004119.2:c.2515G>A / p.(Asp839Asn) is a missense substitution; no ClinGen FLT3 VCEP CSPEC exists and the local custom FLT3 framework in this case (flt3_itd_hotspot_and_function.md, flt3_oncokb_guidance.md) only customizes PM1/PM5/PS3/PM4/BP3 for in-frame ITD/activating length-mutation events and does not provide a PP3/BP4 missense in-silico lookup table, so generic calibrated thresholds were applied. REVEL score is 0.481, which falls in the indeterminate zone below the ~0.644 supporting-pathogenic threshold from the ClinGen SVI-endorsed REVEL calibration (Pejaver et al., PMID 36413997), so it does not reach PP3. SpliceAI predicts a possible donor-gain effect (max delta = 0.46, DS_DG), which is below the 0.5 'recommended' high-confidence cutoff described by Jaganathan et al. for actionable splice-altering predictions and is not accompanied by any VCEP-specific splicing threshold for FLT3, so it is treated as inconclusive rather than supportive. BayesDel score (-0.214309) was retrieved but is not used: this pipeline has no verified published calibration/threshold publication for BayesDel, so it is treated as not_available for PP3/BP4 regardless of the raw value.

Rule

Generic ACMG/AMP PP3 (calibrated in-silico evidence); no gene-specific VCEP PP3/BP4 rule exists for FLT3 missense variants in this case.

Evidence considered

• REVEL score 0.481 (source_registry key 'revel'), below the ~0.644 PP3-supporting threshold from the ClinGen SVI-endorsed REVEL calibration (Pejaver et al., PMID 36413997).
• SpliceAI max delta score 0.46 (DS_DG donor-gain=0.46, DS_DL donor-loss=0.242, DS_AG/DS_AL near zero), below the ~0.5 high-confidence splice-impact threshold widely cited from SpliceAI's original calibration; treated as inconclusive, not supportive.
• BayesDel score -0.214309 retrieved but excluded from the decision because no verified published BayesDel threshold/citation is available to this pipeline.

Still missing

• No FLT3-specific VCEP PP3/BP4 missense lookup spreadsheet is available for this case; only ITD/length-mutation guidance exists.
• SpliceAI max delta (0.46) sits between commonly cited 0.2 and 0.5 cutoffs; no gene-specific splicing threshold is defined for FLT3, so this remains borderline/inconclusive rather than clearly supportive.
• BayesDel score exists but cannot be used because no verified published threshold/citation is available to this pipeline.

Sources

Reasoning

PP5 is restricted here to an exact-variant ClinVar expert-panel classification of Pathogenic or Likely pathogenic. The exact variant NM_004119.2:c.2515G>A is absent from ClinVar: no variation ID, submissions, review status, or expert-panel assertion is present. OncoKB's Likely Oncogenic label is gene/variant-context evidence without variant-specific reviewed functional evidence and is not a ClinVar expert-panel classification. PP5 therefore does not trigger.

Rule

PP5 only when the exact variant has a ClinVar expert-panel Pathogenic/Likely pathogenic classification; non-expert and aggregate assertions are excluded.

Evidence considered

• ClinVar: 'This variant is absent from ClinVar.'; variation_ids = [], submissions = [], review_status = ''.
• OncoKB reports Likely Oncogenic but explicitly states that no variant-specific reviewed functional evidence was identified; this is not an expert-panel ClinVar assertion.
• Rule applied: ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions cannot trigger PP5.

Still missing

• No exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification exists in the case evidence.

Sources

Reasoning

The variant does not meet the benign stand-alone population-frequency threshold. gnomAD v4.1 reports 1/1,614,010 alleles (AF 6.19575e-07; MAF 0.00006%), with the highest reported population frequency in Remaining individuals at 1/62,480 alleles (AF 1.60051e-05; MAF 0.00160%); these frequencies are far below the operating BA1 threshold of >1%.

Rule

Generic ACMG/AMP benign stand-alone frequency rule as operationalized for this non-VCEP case: BA1 requires population frequency >1%.

Evidence considered

• gnomAD v4.1: total allele count 1 of 1,614,010, total AF 6.195748477394811e-07, and maximum observed population AF 1.6005121638924455e-05 in Remaining individuals.
• The variant was absent from gnomAD v2.1 and gnomAD-Canada v1.0.
• The observed frequency is far below the non-VCEP BA1 operating threshold of >1%; no threshold artifact supports BA1.

Sources

Reasoning

The variant does not meet the benign-supporting frequency threshold. gnomAD v4.1 reports a total AF of 6.19575e-07 and a highest observed population AF of 1.60051e-05, both far below the non-VCEP BS1 operating threshold of >0.3%. There is no evidence of a population frequency incompatible with a rare disease allele.

Rule

Generic ACMG/AMP BS1 rule as operationalized for this non-VCEP case: apply only when population frequency is above the disease-compatible maximum; the non-VCEP operating threshold used here is >0.3%.

Evidence considered

• gnomAD v4.1: total AF 6.195748477394811e-07 (1/1,614,010 alleles); highest observed population AF 1.6005121638924455e-05 (1/62,480 alleles) in Remaining individuals.
• The variant was absent from gnomAD v2.1 and gnomAD-Canada v1.0.
• The observed frequencies are far below the non-VCEP BS1 operating threshold of >0.3%; ancestry-specific counts do not indicate a higher frequency in any reported major population.

Sources

Reasoning

BS2 is not met. gnomAD v4.1 reports zero homozygotes for the variant, including zero homozygotes in every displayed population, and no unaffected adult homozygote or hemizygote observation is available from the case evidence. The single observed allele is insufficient for BS2.

Rule

Generic ACMG/AMP BS2 rule: requires observation of the variant in healthy adults, typically in homozygous or hemizygous state when relevant; zero homozygotes and no healthy-adult observation do not satisfy BS2.

Evidence considered

• gnomAD v4.1 reports total homozygotes = 0; the highest-frequency Remaining individuals population also has homozygote count 0, and all displayed ancestry groups have homozygote count 0.
• The variant is absent from gnomAD v2.1 and gnomAD-Canada v1.0, so those datasets provide no homozygote observation.
• No unaffected adult cohort, penetrance analysis, or verified healthy homozygote/hemizygote observation is present in the case bundle.

Sources

Reasoning

No functional study of any kind -- benign or pathogenic -- was identified for FLT3 NM_004119.2:c.2515G>A (p.Asp839Asn). BS3 requires a well-established in vitro or in vivo functional study showing no damaging effect on protein function/splicing; no such study, and no PMIDs at all, were returned for this variant (literature_pass.json: total_pmids=0). OncoKB's gene-level page did not provide a variant-specific functional readout in either direction (is_variant_specific=false, not_reviewed=true), so it cannot support a benign functional conclusion. The local FLT3 custom framework provides guidance only for PS3 (pathogenic-direction) in the ITD/activating length-mutation context and contains no BS3 rule or benign functional data source. Absence of functional evidence is not evidence of benign function, so BS3 is not met.

Rule

Generic ACMG/AMP BS3 (well-established functional study showing no damaging effect); no local FLT3 custom framework BS3 rule exists. Not met due to complete absence of functional data.

Evidence considered

• OncoKB variant page for FLT3 D839N returned found=true but is_variant_specific=false and not_reviewed=true, with no biological_effect data returned, meaning no functional readout of any kind (damaging or benign) is available from this source.
• literature_pass.json reports total_pmids=0 and 'No PMIDs found across evidence sources'; the case publications/ directory is empty, so no functional study of any kind (pathogenic or benign) could be reviewed for BS3.

Still missing

• No functional assay data (benign or pathogenic direction) exists for D839N in any bundled source
• No PMIDs returned by the literature search; publications directory is empty
• No CSPEC/VCEP BS3 rule available for FLT3 to consult (local framework only addresses PS3 for the ITD/length-mutation class)

Sources

Reasoning

The same computational evidence used for PP3 also fails to meet BP4. REVEL score 0.481 is above the ~0.290 benign-supporting threshold from the ClinGen SVI-endorsed REVEL calibration (Pejaver et al., PMID 36413997), so it does not support benignity. SpliceAI's max delta of 0.46 (donor-gain) is not near zero and does not indicate a clearly benign splicing prediction. No FLT3-specific VCEP PP3/BP4 missense lookup table exists in the local custom framework used for this case (it only customizes PM1/PM5/PS3/PM4/BP3 for ITD-class events), so generic thresholds were applied. BayesDel is excluded as in PP3 because no verified published calibration threshold is available to this pipeline.

Rule

Generic ACMG/AMP BP4 (calibrated in-silico evidence); no gene-specific VCEP PP3/BP4 rule exists for FLT3 missense variants in this case.

Evidence considered

• REVEL score 0.481 (source_registry key 'revel'), above the ~0.290 BP4-supporting benign threshold from the ClinGen SVI-endorsed REVEL calibration (Pejaver et al., PMID 36413997), so does not support BP4.
• SpliceAI max delta score 0.46 (donor-gain), not consistent with a clearly benign/no-impact splicing prediction, so does not support BP4.
• BayesDel score -0.214309 retrieved but excluded from the decision because no verified published BayesDel threshold/citation is available to this pipeline.

Still missing

• No FLT3-specific VCEP PP3/BP4 missense lookup spreadsheet is available for this case.
• BayesDel score exists but cannot be used because no verified published threshold/citation is available to this pipeline.

Sources

Reasoning

BP6 is restricted here to an exact-variant ClinVar expert-panel classification of Benign or Likely benign. The exact variant NM_004119.2:c.2515G>A is absent from ClinVar, with no variation ID, submissions, review status, or expert-panel assertion. There is therefore no qualifying expert-panel Benign/Likely benign classification, and BP6 does not trigger.

Rule

BP6 only when the exact variant has a ClinVar expert-panel Benign/Likely benign classification; non-expert and aggregate assertions are excluded.

Evidence considered

• ClinVar: 'This variant is absent from ClinVar.'; variation_ids = [], submissions = [], review_status = ''.
• Rule applied: ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions cannot trigger BP6.

Still missing

• No exact-variant ClinVar expert-panel Benign or Likely benign classification exists in the case evidence.

Sources

Reasoning

PS1 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

No proband-level parental testing or validated de novo evidence is documented in the case-evidence bundle. Therefore, PS2 cannot be assessed.

Rule

Generic ACMG/AMP PS2: apply only when the variant is confirmed de novo in the affected proband with confirmed maternity and paternity; strength depends on phenotype specificity and supporting de novo evidence.

Evidence considered

• The consolidated case-evidence bundle contains no proband pedigree, parental genotype, or de novo testing result for NM_004119.2:c.2515G>A (NP_004110.2:p.Asp839Asn).
• The applicable FLT3 local custom framework does not specify PS2; generic ACMG/AMP de novo requirements therefore apply.

Still missing

• Parental genotypes for NM_004119.2:c.2515G>A are not provided.
• Testing methodology, maternity/paternity confirmation, and read-level or laboratory confirmation of a de novo event are not provided.

Reasoning

No applicable FLT3 CSPEC/VCEP specification exists, so generic ACMG/AMP PS4 semantics apply. PS4 requires significant enrichment of the exact variant in affected individuals compared with controls. The evidence bundle contains no germline case-control or cohort enrichment data for NM_004119.2:c.2515G>A (p.Asp839Asn), no ClinVar case submissions, no literature PMIDs, and only four somatic COSMIC observations without a germline disease-control comparison. The available gnomAD data are population-control observations, not an affected-case cohort, so an enrichment calculation cannot be performed. PS4 is therefore not_assessed rather than met or affirmatively refuted.

Rule

Generic ACMG/AMP PS4: significant enrichment of the exact variant in affected individuals compared with controls; case-control or equivalent cohort data are required.

Evidence considered

• ClinVar exact-variant search returned no result and no submissions.
• literature_pass.json reports total_pmids = 0; no case-control or affected-case publication was available.
• gnomAD v2.1 and gnomAD-Canada v1.0 report absence; gnomAD v4.1 reports 1/1,614,010 total alleles (AF 6.19575e-07), but these are population data and do not provide affected-case enrichment.
• COSMIC reports four somatic observations, which are not a germline case-control enrichment dataset.
• Generic ACMG/AMP 2015 is used because no applicable FLT3 CSPEC/VCEP specification exists.

Still missing

• No affected-case case-control or cohort allele counts for NM_004119.2:c.2515G>A.
• No comparable control frequency and statistical enrichment analysis from a disease-specific study.

Sources

Reasoning

PM1 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

PM3 cannot be assessed from the available case evidence. The bundle contains no affected-proband observations, no confirmed diagnosis context, no second pathogenic or likely pathogenic variant, and no segregation or phase information establishing the queried FLT3 variant in trans with a pathogenic variant in a recessive disorder. The variant is a missense substitution, but variant class or general gene-level oncogenic context does not substitute for allelic evidence.

Rule

Generic ACMG/AMP PM3: apply only when the variant is observed in trans with a pathogenic or likely pathogenic variant in an affected individual with a recessive disorder, with strength determined by the quality and number of independent affected observations and phase confirmation.

Evidence considered

• The case bundle reports no ClinVar record or literature PMIDs for NM_004119.2:c.2515G>A and provides no affected-proband, compound-heterozygous, or phase observations.
• The retrieved local FLT3 framework is specific to PM1, PM5, PS3, PM4, and BP3 and does not provide a PM3 specification; therefore generic ACMG/AMP PM3 logic is used for this criterion.
• The available evidence describes the queried variant as NP_004110.2:p.(Asp839Asn) and does not identify a pathogenic allele in trans or cis for allelic adjudication.

Still missing

• Affected-proband observations and confirmed phenotype relevant to the FLT3 disease context
• A second pathogenic or likely pathogenic variant in the same gene, if present
• Experimental or family-based phasing establishing trans configuration
• Inheritance model and segregation information for the relevant disorder

Reasoning

PM5 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

No reliable de novo observation without confirmed maternity and paternity is documented in the case-evidence bundle. Therefore, PM6 cannot be assessed.

Rule

Generic ACMG/AMP PM6: apply for a presumed de novo occurrence when maternity and paternity are not confirmed, with strength determined by the quality and specificity of the de novo observation.

Evidence considered

• The consolidated case-evidence bundle contains no case report, pedigree, parental testing result, or other source documenting an unconfirmed de novo occurrence for this variant.
• The applicable FLT3 local custom framework does not specify PM6; generic ACMG/AMP requirements therefore apply.

Still missing

• A reported de novo occurrence in the proband is not provided.
• Parental testing or other evidence addressing maternity and paternity is not provided.

Reasoning

No affected relatives, informative meioses, or family segregation data are documented. The variant is therefore not assessable for cosegregation evidence.

Rule

Generic ACMG/AMP PP1: apply when a variant cosegregates with disease in affected family members; strength may be increased according to the number and informativeness of meioses and the specificity of the phenotype.

Evidence considered

• The consolidated case-evidence bundle contains no family segregation observations for NM_004119.2:c.2515G>A.
• The applicable FLT3 local custom framework does not specify PP1; generic ACMG/AMP segregation requirements therefore apply.

Still missing

• No pedigree or family structure is provided.
• No genotypes and phenotypes for affected or unaffected relatives are provided.
• No count of informative meioses or segregation likelihood assessment is provided.

Reasoning

PP2 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

No applicable FLT3 CSPEC/VCEP specification exists, so generic ACMG/AMP PP4 semantics apply. PP4 requires a highly specific phenotype or family history in a variant carrier that points to a single-gene etiology. The case bundle contains no proband phenotype, family history, tumor phenotype, segregation information, or clinical case report for NM_004119.2:c.2515G>A (p.Asp839Asn). FLT3 being associated with myeloid malignancies in general is gene-level context and is not sufficient to establish PP4 for this carrier. The criterion cannot be evaluated from the available evidence.

Rule

Generic ACMG/AMP PP4: phenotype or family history highly specific for a disease with a single genetic etiology; variant-carrier clinical data are required.

Evidence considered

• case_summary.json provides no proband phenotype, family history, segregation, or tumor findings.
• ClinVar exact-variant search returned no result and therefore provides no carrier phenotype.
• literature_pass.json reports total_pmids = 0; no variant-specific phenotype publication was available.
• Gene-level OncoKB context for FLT3 malignancies is not variant-carrier phenotype evidence and was not used to trigger PP4.

Still missing

• No proband phenotype or family-history information is present for the variant carrier.
• No variant-specific clinical case report with a phenotype highly specific for FLT3-related disease.

Sources

Reasoning

No unaffected relatives who were adequately tested for the variant, and no informative non-segregation observations, are documented. BS4 cannot be assessed.

Rule

Generic ACMG/AMP BS4: apply when a variant does not segregate with disease in affected family members or is observed in unaffected members in a manner inconsistent with disease, taking penetrance, age, phenotype specificity, and phenocopies into account.

Evidence considered

• The consolidated case-evidence bundle contains no non-segregation data from unaffected relatives for NM_004119.2:c.2515G>A.
• The applicable FLT3 local custom framework does not specify BS4; generic ACMG/AMP non-segregation requirements therefore apply.

Still missing

• No family pedigree or phenotype data are provided.
• No tested unaffected relatives carrying or lacking the variant are identified.
• No assessment of age-related penetrance, phenocopies, or other reasons for possible non-segregation is provided.

Reasoning

BP1 could not be assessed because its agent did not produce valid output.

Still missing

• attempt 1: evidence_refs must contain only source_registry keys
• attempt 2: evidence_refs must contain only source_registry keys
• attempt 3: evidence_refs must contain only source_registry keys

Reasoning

BP2 cannot be assessed from the available case evidence. No individual-level genotype data, affected or unaffected status, second variant, or experimentally established cis/trans phase is provided. The population observations and somatic database entry do not establish the allelic configuration required for BP2.

Rule

Generic ACMG/AMP BP2: consider only when the variant is observed in trans with a pathogenic variant in a fully penetrant dominant disorder or in cis with a pathogenic variant, with phase and the relevant clinical/inheritance context established.

Evidence considered

• The case bundle contains no family or proband-level genotype records showing the queried variant with another variant in cis or trans.
• The bundle reports only population-level gnomAD observations, absence from ClinVar, and a somatic COSMIC occurrence count; these do not provide the phase and clinical context required for BP2.
• The retrieved local FLT3 framework does not specify BP2, so generic ACMG/AMP BP2 logic is used for this criterion.

Still missing

• A documented second variant in the same individual
• Reliable cis/trans phasing for the queried variant and the second variant
• Clinical phenotype and inheritance context of the individual
• Pathogenicity classification of the second allele, if BP2 is being considered in trans

Reasoning

Generic ACMG/AMP BP5 requires evidence that an alternate molecular basis explains the phenotype in a reported case carrying the variant. No carrier case, proband phenotype, family history, or other-gene testing result is present for NM_004119.2:c.2515G>A. The four COSMIC somatic observations do not establish an alternate germline cause in a clinical proband. Consequently, BP5 cannot be evaluated.

Rule

Generic ACMG/AMP BP5: an alternate molecular basis of disease explains the phenotype in a variant carrier; carrier-level clinical and molecular data are required.

Evidence considered

• ClinVar exact-variant search returned no result and no carrier case.
• literature_pass.json reports total_pmids = 0; no case report with alternate molecular diagnosis was available.
• COSMIC somatic observations are not proband-level evidence of an alternate germline molecular basis.
• No phenotype or other-gene testing information is present in case_summary.json.

Still missing

• No reported carrier case with proband-level phenotype and testing for an alternate molecular basis.
• No confirmed pathogenic variant in another gene or other established alternate cause documented for a carrier.

Sources