NF1 c.2190C>G, p.(Asn730Lys)

Reasoning

PM2 is met at supporting strength because the exact variant is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0. These population resources provide no observed allele count or frequency for the variant. No ancestry-specific population observation or frequency artifact was identified that would contradict rarity.

Evidence considered

• The exact variant NM_001042492.2:c.2190C>G / NP_001035957.1:p.(Asn730Lys) is reported as absent from gnomAD v2.1.
• The exact variant is reported as absent from gnomAD v4.1.
• The exact variant is reported as absent from gnomAD-Canada v1.0.
• The applicable NF1 ClinGen Neurofibromatosis and Schwannomatosis Specification, version 1.0, is present in the case bundle, but the extracted framework is incomplete and contains no structured PM2 threshold rule.

Still missing

• The bundle does not provide numeric allele-number denominators, ancestry-stratified counts, or a disease-specific maximum credible allele frequency calculation.
• The NF1 CSPEC v1.0 framework was identified but its extracted ruleset contains no structured population-criterion rules; the PM2 call therefore relies on the documented absence observations and is not a CSPEC-specific threshold adjudication.

Sources

Reasoning

The CSpec NF1 VCEP framework provides no gene-specific PP3/BP4 lookup or rule_payload, so generic ACMG/AMP computational evidence applies. Two independent calibrated computational lines of evidence converge on a benign prediction: (1) SpliceAI predicts no splice impact for this variant (max delta score 0.002, all deltas <0.01), consistent with no aberrant splicing; and (2) REVEL score is 0.06, a low score under the published ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID 36413997), consistent with a benign missense prediction. Both computational predictions point in the same (benign) direction with no conflict, supporting BP4 at supporting strength. BayesDel score (-0.61216) was computed but is not usable because no verified published threshold/table is available to this pipeline for BayesDel; it is not counted toward this determination.

Rule

generic_ACMG_PP3_BP4_computational

Evidence considered

• CSpec NF1 VCEP framework (doc 1553923915) has no gene-specific PP3/BP4 lookup table or rule_payload (framework_complete false), so generic ACMG BP4 logic applies.
• SpliceAI evidence_sentence: 'SpliceAI predicts no significant splice impact for this variant (max delta score = 0.00)'; individual deltas DS_AG=0.0, DS_AL=0.002, DS_DG=0.001, DS_DL=0.001 are all far below any splice-disruption threshold, supporting BP4 via the splice-impact path.
• REVEL score 0.06 (source_registry 'revel') is low and, per the published ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID 36413997), falls in the benign-supporting range for the missense in-silico path.
• BayesDel score -0.61216 was computed but excluded from this determination because no verified published BayesDel calibration threshold/table can be named for this pipeline; not used to escalate strength beyond supporting.

Still missing

• BayesDel score exists but has no verified published calibration threshold available to this pipeline, so it could not be additionally leveraged to strengthen BP4 beyond supporting.

Sources

Reasoning

PS1 requires a different nucleotide change resulting in the same amino acid change (p.Asn730Lys) already established as pathogenic. The variant NM_001042492.2:c.2190C>G / p.(Asn730Lys) is absent from ClinVar entirely, and no ClinVar record for any alternate codon producing p.Asn730Lys was identified in the case bundle. No comparator variant exists to support PS1.

Evidence considered

• case_summary.json/compact_evidence.clinvar reports the variant is absent from ClinVar (evidence_sentence: 'This variant is absent from ClinVar'), so no same-amino-acid-change pathogenic comparator is available for PS1.

Still missing

• No comparator pathogenic ClinVar entry (any nucleotide change) encoding the identical amino acid change p.Asn730Lys was found; PS1 cannot be supported without one.

Sources

Reasoning

PM1 requires the variant to lie in a mutational hotspot or well-established functional domain without benign variation. Residue N730 maps to the N-HEAT domain of neurofibromin (approx. aa 1-1189), a large tandem HEAT/ARM-repeat scaffolding region, rather than the GAP-related domain (GRD, aa ~1197-1546) / RasGAP domain (aa ~1209-1484) that contains the well-characterized catalytic arginine-finger hotspot (R1276) responsible for Ras-GTP hydrolysis. The literature reviewed for this gene explicitly ties recurrent missense pathogenic clustering to the GRD/RasGAP region, not the N-HEAT region. The case's cancer-hotspot and OncoKB lookups for this exact position returned no significant hotspot signal and 'Unknown Oncogenic Effect'. No CSPEC-codified PM1 domain/hotspot rule payload was retrievable to override this generic assessment.

Evidence considered

• OncoKB lookup for NF1 N730K returned 'Unknown Oncogenic Effect' with no variant-specific reviewed functional evidence, and the case's cancerhotspots.org screen for NF1 N730 found no statistically significant hotspot row.
• pvs1_gene_context.json supporting source PMID:41220108 maps neurofibromin domains (N-HEAT aa1-1189; GRD aa1197-1546; RasGAP aa1209-1484) and identifies the GRD/RasGAP region, centered on the R1276 arginine finger, as the principal recurrent missense hotspot/functional domain for NF1; residue N730 falls within the N-HEAT region, which is not identified as a comparable hotspot or indispensable catalytic domain.
• vcep_materials.json shows cspec found=true but rule_payload is empty (framework_complete=false), so no NF1 VCEP-specific PM1 domain boundary rule was available to apply instead of this generic assessment.

Still missing

• No codified CSPEC PM1 hotspot/domain rule payload was retrievable from vcep_materials.json (rule_payload empty) to confirm/deny NF1 VCEP-specific PM1 domain boundaries.
• No structural/functional study specific to residue N730 was found.

Sources

Reasoning

PP2 requires the gene to have missense variation as a common disease mechanism with a low rate of benign missense variation. NF1-associated tumor-predisposition disease is predominantly driven by loss-of-function via nonsense, frameshift, splice, and large deletion variants; literature reviewed for this gene reports nonsense variants as the most common pathogenic variant class (48% in one cohort) versus missense variants representing a minority (13% in that same cohort, and only 3% of driver mutations in the largest pan-cancer comparator database, versus 78% truncating). This pattern argues against missense change being 'a common mechanism of disease' for NF1 as required by PP2. No NF1 VCEP-specific PP2 recommendation could be retrieved from the case's cspec materials (rule_payload empty), so this assessment relies on the generic ACMG/AMP PP2 criterion and gene-level literature.

Evidence considered

• vcep_materials.json / final_classification_framework shows cspec found=true (NF1 VCEP v1.0) but rule_payload is empty and framework_complete=false, so no codified NF1-specific PP2 statement is available in this bundle.
• pvs1_gene_context.json supporting source PMID:41220108 reports that among 72 NF1 variants in a molecular tumor board cohort, nonsense variants were most common (48%), missense variants a minority (13%), and that the largest comparator pan-cancer database (cBioPortal) shows missense variants at only ~3% versus 78% truncating among driver NF1 mutations, indicating LOF/truncating mechanisms predominate over missense for NF1 pathogenicity.

Still missing

• No codified NF1 VCEP PP2 applicability statement was retrievable from vcep_materials.json (rule_payload empty); an official VCEP-specific PP2 rule (if any) could not be confirmed.

Sources

Reasoning

This is a missense variant (NM_001042492.2:c.2190C>G, p.(Asn730Lys)). The governing ClinGen NF1/Schwannomatosis VCEP specification (CSpec doc 1553923915) was checked and contains no gene-specific PP3/BP4 lookup table or rule payload (rule_payload is empty, no gene VCEP files were retrieved), so generic ACMG/AMP computational-evidence assessment applies. SpliceAI predicts no splice impact (max delta score 0.002, all four delta scores <0.01), well below any threshold used to flag a cryptic/altered splice event, so no PP3 support from the splice-impact path. For the missense path, REVEL score is 0.06, a low score consistent with a benign prediction rather than a deleterious one, so REVEL does not support PP3. BayesDel score (-0.61216) is available but this pipeline has no verified published calibration threshold for BayesDel, so it cannot be used to support PP3. No calibrated in-silico evidence supports pathogenicity for this variant.

Rule

generic_ACMG_PP3_BP4_computational

Evidence considered

• CSpec NF1 VCEP framework (doc 1553923915) contains no gene-specific PP3/BP4 lookup table or rule_payload; framework_complete is false and no gene VCEP user files were found, so generic ACMG PP3/BP4 logic with published calibrated thresholds applies.
• SpliceAI evidence_sentence: 'SpliceAI predicts no significant splice impact for this variant (max delta score = 0.00)'; scores DS_AG=0.0, DS_AL=0.002, DS_DG=0.001, DS_DL=0.001, well below the 0.2 threshold commonly used to flag splice-altering variants, so the splice-impact path does not support PP3.
• REVEL score 0.06 (source_registry 'revel'), a low score under the ClinGen SVI REVEL calibration (Pejaver et al. 2022, PMID 36413997), which does not meet any pathogenic-direction (PP3) threshold, so the missense in-silico path does not support PP3.
• BayesDel score -0.61216 was computed but is treated as not_available/insufficiently calibrated for PP3/BP4 per pipeline policy, since no verified published BayesDel threshold publication/table can be named.

Still missing

• No published, pipeline-verified BayesDel threshold is available, so the BayesDel score could not be used to assess PP3 despite being computed.

Sources

Reasoning

PP5 is not met. The exact variant NM_001042492.2:c.2190C>G is reported as absent from ClinVar, with no expert-panel classification. Ordinary laboratory submissions, aggregate labels, and non-expert assertions would not qualify under the governing rule, and none are present here.

Rule

PP5 supports Pathogenic/Likely pathogenic only when the exact variant has a ClinVar expert-panel Pathogenic or Likely pathogenic classification; all other ClinVar evidence is insufficient.

Evidence considered

• ClinVar exact-variant retrieval found no record for NM_001042492.2:c.2190C>G; variation IDs, classifications, review status, submissions, and PMIDs are all absent.
• Under the instructed PP5 rule, only an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification can support PP5.

Sources

Reasoning

BA1 is not met. The variant is absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0; therefore no population frequency exceeds a stand-alone benign threshold.

Evidence considered

• The consolidated case evidence reports the exact variant NM_001042492.2:c.2190C>G / NC_000017.11:g.31226623C>G as absent from gnomAD v2.1.
• The consolidated case evidence reports the exact variant as absent from gnomAD v4.1.
• The consolidated case evidence reports the exact variant as absent from gnomAD-Canada v1.0.

Still missing

• No ancestry-specific allele counts or population frequency above a BA1 threshold were identified.

Sources

Reasoning

BS1 is not met. The variant is absent from the available gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0 datasets, so there is no observed allele frequency above a disease-specific benign threshold.

Evidence considered

• The exact variant is reported as absent from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0.
• Because no allele frequency was observed, the available population data do not support BS1.

Still missing

• No disease-specific maximum credible allele frequency calculation or population frequency above a BS1 threshold was identified.

Sources

Reasoning

BP6 is not met. The exact variant NM_001042492.2:c.2190C>G is reported as absent from ClinVar, with no expert-panel classification. Ordinary laboratory submissions, aggregate labels, and non-expert assertions would not qualify under the governing rule, and none are present here.

Rule

BP6 supports Benign/Likely benign only when the exact variant has a ClinVar expert-panel Benign or Likely benign classification; all other ClinVar evidence is insufficient.

Evidence considered

• ClinVar exact-variant retrieval found no record for NM_001042492.2:c.2190C>G; variation IDs, classifications, review status, submissions, and PMIDs are all absent.
• Under the instructed BP6 rule, only an exact-variant ClinVar expert-panel Benign or Likely benign classification can support BP6.

Sources

Reasoning

The case-evidence bundle provides no proband-level de novo observation, parental genotypes, maternity/paternity confirmation, or phenotype documentation sufficient to establish a confirmed de novo variant. PS2 therefore cannot be assessed.

Evidence considered

• The retrieved NF1 ClinGen Neurofibromatosis and Schwannomatosis Expert Panel specification is present, but its extracted rule payload is empty and the framework is marked incomplete.
• The consolidated case-evidence bundle contains no proband, parental-testing, maternity/paternity, or confirmed de novo evidence for NM_001042492.2:c.2190C>G.

Still missing

• Proband phenotype and diagnostic consistency are not provided.
• Maternal and paternal testing results are not provided.
• Confirmation of maternity and paternity is not provided.
• Read-depth, allele-balance, and assay-quality information for parental testing is not provided.

Sources

Reasoning

No functional/biochemical assay data (e.g., Ras-GAP activity assays, in vitro splicing assays, minigene assays) for NM_001042492.2:c.2190C>G (p.Asn730Lys) were identified in the case-evidence bundle. The literature_pass.json search found zero PMIDs overall ('No PMIDs found across evidence sources'), and no functional-study PMIDs were surfaced elsewhere in the bundle. OncoKB explicitly reports no variant-specific reviewed functional evidence. The NF1 VCEP (ClinGen Neurofibromatosis and Schwannomatosis Expert Panel Specifications v1.0) cspec framework material retrieved for this case contains no structured PS3 rule payload (rule_payload is empty) to apply even if assay data existed.

Still missing

• No published functional assay (Ras-GAP activity, neurofibromin stability/localization, splicing minigene, or other validated functional study) identified for this specific variant.
• NF1 VCEP PS3/BS3 assay-validation table (with specificity/sensitivity-calibrated strength levels) was not retrievable in structured form from this case's vcep_materials/cspec bundle; only the framework header/metadata was captured, not the rule content.
• A targeted literature search specifically for NF1 functional/splicing assay studies of c.2190C>G or p.Asn730Lys was not performed in this evidence-gathering pass (literature_pass returned 0 PMIDs across all sources).

Reasoning

No affected-case series, case-control comparison, statistically significant enrichment analysis, or variant-specific affected-case count was available in the case-evidence bundle. PS4 cannot be applied from the absence of the variant in population databases or from the absence of a ClinVar record.

Rule

Apply PS4 only when statistically significant enrichment of the variant is demonstrated in affected individuals or an appropriate affected-case series/case-control dataset supports the disease association; no such data are available here.

Evidence considered

• The applicable NF1 ClinGen Neurofibromatosis and Schwannomatosis Expert Panel specification was reviewed first; its retrieved criteria payload is empty/incomplete and provides no usable PS4-specific rule for this case.
• The consolidated bundle reports no literature PMIDs, no variant-specific clinical cases, and no case-control data for NM_001042492.2:c.2190C>G.

Still missing

• Variant-specific affected-case observations and an appropriate case-control enrichment analysis are unavailable.

Sources

Reasoning

The NF1-specific CSPEC framework was retrieved, but its PM3 rule payload is empty and the case-evidence bundle contains no affected-proband observations, second pathogenic/likely pathogenic allele, phase information, or segregation/inheritance data. PM3 therefore cannot be assessed.

Rule

No explicit NF1-specific PM3 rule was retrieved; generic ACMG/AMP PM3 semantics require observation of the variant in trans with a pathogenic variant in an affected individual, with strength determined by the applicable disease/inheritance framework.

Evidence considered

• The retrieved NF1 CSPEC is Version 1.0 for autosomal dominant neurofibromatosis type 1, but the framework is marked incomplete and has an empty rule_payload.
• The consolidated case-evidence bundle reports no literature PMIDs and contains no documented affected-proband, compound-heterozygous, phase, or inheritance observation for NM_001042492.2:c.2190C>G (p.Asn730Lys).

Still missing

• No affected proband or affected-family-member genotype/phenotype observations are documented.
• No pathogenic or likely pathogenic variant in the same gene is documented as being in trans or cis.
• No parental testing, segregation, or phase data are available.
• The retrieved NF1 CSPEC rule payload does not specify a PM3 rule.

Sources

Reasoning

PM5 requires a different missense change at the same residue (Asn730) previously established as pathogenic, distinct from the observed p.Asn730Lys. The case's dedicated PM5 candidate-harvesting step explicitly flagged that it could not confirm classic same-residue PM5 semantics safely and skipped automatic candidate collection (pm5_mode='unknown', eligible_for_classic_pm5_search=false, 0 candidates seen), instructing reviewers not to apply PM5 from that artifact alone. No other same-residue comparator variant data was found elsewhere in the bundle (ClinVar is empty for this locus).

Evidence considered

• pm5_candidates.json reports pm5_mode='unknown', eligible_for_classic_pm5_search=false, reason='Unable to confirm classic same-residue PM5 semantics safely; skipping automatic candidate harvesting', and recommendation.level='not_applicable' with hermes_instruction to not apply PM5 from this artifact alone.
• The case's ClinVar evidence_sentence confirms the queried variant itself is absent from ClinVar, and no other codon-730 missense entries were surfaced anywhere in the bundle.

Still missing

• No classic same-residue PM5 candidate search was completed; pm5_candidates.json explicitly withheld automated harvesting.
• Requires manual/targeted ClinVar search for other pathogenic missense variants at NF1 codon 730 before PM5 can be adjudicated.

Sources

Reasoning

No case report or clinical record documents this variant as presumed de novo, and no parental testing or inheritance information is available. PM6 therefore cannot be assessed.

Evidence considered

• The retrieved NF1 ClinGen Neurofibromatosis and Schwannomatosis Expert Panel specification is present, but its extracted rule payload is empty and the framework is marked incomplete.
• The consolidated case-evidence bundle reports no literature PMIDs and contains no documentation of presumed de novo status for this variant.

Still missing

• A clinical report documenting presumed de novo occurrence is not provided.
• Parental genotypes and the extent of parental testing are not provided.
• Proband phenotype consistency with NF1 is not provided.

Sources

Reasoning

No affected relatives, informative meioses, cosegregation observations, or family phenotypes are documented in the case-evidence bundle. PP1 cannot be assessed.

Evidence considered

• The retrieved NF1 ClinGen Neurofibromatosis and Schwannomatosis Expert Panel specification is present, but its extracted rule payload is empty and the framework is marked incomplete.
• The consolidated case-evidence bundle contains no pedigree, affected-relative genotypes, family phenotype data, or meiosis count relevant to cosegregation.

Still missing

• Family pedigree and number of informative meioses are not provided.
• Genotypes of affected relatives are not provided.
• Phenotype status and clinical consistency of relatives are not provided.
• Any segregation analysis or phase information is not provided.

Sources

Reasoning

The bundle contains no patient phenotype, clinical diagnostic features, or phenotype-specific case report for this exact variant. A gene-level description of NF1 is not sufficient to establish a highly specific phenotype for PP4.

Rule

Apply PP4 only when the patient's phenotype or family history is highly specific for NF1 and the applicable disease-specific framework permits its use; no patient phenotype is available here.

Evidence considered

• The applicable NF1 ClinGen Neurofibromatosis and Schwannomatosis Expert Panel specification was reviewed first; its retrieved criteria payload is empty/incomplete and provides no usable PP4-specific rule for this case.
• The case-evidence bundle does not provide a clinical phenotype or exact-variant phenotype correlation.

Still missing

• Patient-level clinical phenotype and diagnostic specificity are unavailable.

Sources

Reasoning

BS2 was not assessed because the case bundle contains no reliable information about unaffected adults homozygous for this variant, no relevant disease-cohort homozygote observations, and no penetrance or phenotype ascertainment data that would establish the BS2 condition.

Evidence considered

• The consolidated case-evidence bundle contains no BS2-relevant homozygote or unaffected-adult observation.

Still missing

• No unaffected adult homozygotes were documented.
• No disease-specific cohort or family data addressing homozygous occurrence were available.
• No penetrance, ascertainment, or phenotype information was available to determine whether any homozygotes would be expected to be affected.

Reasoning

No functional/biochemical assay data showing normal (wild-type-like) neurofibromin function for NM_001042492.2:c.2190C>G (p.Asn730Lys) were identified in the case-evidence bundle. As with PS3, the literature_pass.json search found zero PMIDs overall, no functional-study PMIDs were surfaced elsewhere in the bundle, and OncoKB reports no variant-specific reviewed functional evidence. No structured NF1 VCEP BS3 rule content was retrievable from the cspec/vcep_materials bundle for this case.

Still missing

• No published functional assay demonstrating normal/benign functional readout for this specific variant identified.
• NF1 VCEP PS3/BS3 assay-validation table was not retrievable in structured form from this case's vcep_materials/cspec bundle.
• A targeted literature search specifically for NF1 functional assay studies of c.2190C>G or p.Asn730Lys was not performed in this evidence-gathering pass.

Reasoning

No unaffected relatives tested for the variant, non-segregation observation, or reliable phenotype/genotype comparison is documented. BS4 cannot be assessed.

Evidence considered

• The retrieved NF1 ClinGen Neurofibromatosis and Schwannomatosis Expert Panel specification is present, but its extracted rule payload is empty and the framework is marked incomplete.
• The consolidated case-evidence bundle contains no unaffected-relative testing or documented non-segregation evidence for this variant.

Still missing

• Genotypes of unaffected relatives are not provided.
• Phenotype assessments and ages of relatives are not provided.
• The number of informative unaffected relatives and meioses is not provided.
• Potential age-dependent or variably expressive NF1 features are not addressed.

Sources

Reasoning

BP1 requires a missense variant in a gene where primarily truncating variants are known to cause disease, generally supported by a validated low pathogenic-missense rate specific to that gene. NF1's disease mechanism is predominantly truncating/LOF, which partially supports BP1's premise, but the gene also harbors well-documented recurrent pathogenic missense variants clustering in the GAP-related domain / RasGAP region (e.g., R1276Q, R1276P, Q1447H), meaning missense pathogenicity is not uniformly negligible across the gene. No NF1 VCEP-calibrated BP1 rule or gene-specific benign-missense-rate threshold was retrievable from the case's cspec materials (rule_payload empty), so a reliable determination cannot be made from the evidence in this bundle alone.

Evidence considered

• vcep_materials.json / final_classification_framework shows cspec found=true (NF1 VCEP v1.0) but rule_payload is empty and framework_complete=false, so no codified NF1-specific BP1 statement or missense benign-rate calibration is available in this bundle.
• pvs1_gene_context.json supporting source PMID:41220108 documents recurrent, well-characterized pathogenic missense variants in NF1 (R1276Q, R1276P, Q1447H) clustering in the GAP-related/RasGAP domain, showing that missense pathogenicity is not negligible gene-wide despite truncating variants predominating overall.

Still missing

• No NF1 VCEP-calibrated BP1 threshold or statistical validation study (e.g., missense z-score / benign-missense-rate calculation) was retrievable from vcep_materials.json (rule_payload empty).
• Residue-730/N-HEAT-domain-specific benign missense background rate was not assessed in this bundle.

Sources

Reasoning

The NF1-specific CSPEC framework was retrieved, but its BP2 rule payload is empty and the case-evidence bundle contains no observation of the variant in cis with a pathogenic variant, no affected/unaffected carrier phase data, and no family testing. BP2 therefore cannot be assessed.

Rule

No explicit NF1-specific BP2 rule was retrieved; generic ACMG/AMP BP2 semantics require observation of the variant in cis with a pathogenic variant, with interpretation dependent on the disease mechanism and phase evidence.

Evidence considered

• The retrieved NF1 CSPEC is Version 1.0 for autosomal dominant neurofibromatosis type 1, but the framework is marked incomplete and has an empty rule_payload.
• The consolidated case-evidence bundle contains no documented cis-phase observation, pathogenic/likely pathogenic second allele, or family genotype data for NM_001042492.2:c.2190C>G (p.Asn730Lys).

Still missing

• No cis-phase observation with a pathogenic or likely pathogenic variant is documented.
• No family or proband genotype data establish the relevant allelic configuration.
• No parental testing, segregation, or inheritance data are available.
• The retrieved NF1 CSPEC rule payload does not specify a BP2 rule.

Sources

Reasoning

No affected individual, phenotype, or molecular finding is available to establish an alternative molecular explanation for the disease. BP5 therefore cannot be assessed from this bundle.

Rule

Apply BP5 only when an alternative molecular explanation accounts for the affected individual's phenotype; no such individual-level evidence is available here.

Evidence considered

• The applicable NF1 ClinGen Neurofibromatosis and Schwannomatosis Expert Panel specification was reviewed first; its retrieved criteria payload is empty/incomplete and provides no usable BP5-specific rule for this case.
• The case-evidence bundle contains no patient-level diagnostic data or alternative pathogenic molecular finding.

Still missing

• An affected individual's phenotype and an established alternative molecular explanation are unavailable.

Sources