NF1 c.2544G>A, p.(Gly848=)

Reasoning

NF1 CSPEC (ClinGen Neurofibromatosis and Schwannomatosis Expert Panel v1.0, doc 1553923915) was located but its retrieved ruleset payload is empty (criteria: [], rule_payload: {}, framework_complete: false) and no local gene framework files exist, so generic ACMG/AMP 2015 (PMID:25741868) governs: BA1 is met when allele frequency is >5% in a large general population cohort. The variant is overwhelmingly African/African American (AFR)-specific: gnomAD v4.1 AFR AF = 0.0831 (6225/74954 alleles, 305 homozygotes) with joint grpmax FAF = 0.0813; gnomAD v2.1 AFR AF = 0.0794 (1976/24898, 95 homozygotes) with grpmax FAF = 0.0771; gnomAD-Canada v1.0 AFR AF = 0.10 (102/1020) with grpmax faf95 = 0.0843. All three independent datasets give a highest-ancestry AF (and FAF) above the >5% BA1 threshold, and above this pipeline's more conservative >1% operating threshold. Total AFs (0.45% v4.1, 0.77% v2.1) are lower because the allele is AFR-restricted, so the ancestry-stratified maximum AF / grpmax FAF is the appropriate metric (standard practice for founder/ancestry-specific alleles). BA1 met as stand-alone benign evidence.

Rule

generic ACMG/AMP 2015 BA1 (PMID:25741868): allele frequency >5% in a large general population cohort -> stand-alone benign; NF1 CSPEC doc 1553923915 retrieved with empty rule payload, generic fallback applies

Evidence considered

• ACMG/AMP 2015 (PMID:25741868): 'In general, an allele frequency in a control population that is greater than expected for the disorder (Table 6) is considered strong support for a benign interpretation for a rare Mendelian disorder (BS1) or if over 5% then it is considered as stand-alone support (BA1).'
• gnomAD v4.1: AFR AF = 0.0830509 (6225/74954 alleles, 305 homozygotes), joint grpmax FAF = 0.0813264, total AF = 0.00451209 (7273/1611892) — AFR AF and grpmax FAF both exceed the >5% BA1 threshold
• gnomAD v2.1: AFR AF = 0.0793638 (1976/24898 alleles, 95 homozygotes), grpmax FAF = 0.0771056, total AF = 0.00770786 (2173/281920) — AFR AF and grpmax FAF both exceed the >5% BA1 threshold
• gnomAD-Canada v1.0: grpmax AFR AF = 0.1 (102/1020 alleles, 5 homozygotes), grpmax faf95 = 0.08428927 — exceeds the >5% BA1 threshold
• Highest-ancestry AF / grpmax FAF used instead of total AF because the allele is AFR-restricted (ASJ/EAS/FIN = 0.0); total AF dilutes ancestry-specific frequency and would understate BA1 evidence

Still missing

• Locus overlaps a segmental duplication region (gnomAD-Canada region_flag.segdup=true) and gnomAD v4.1 genome AF (2.37%) exceeds exome AF (0.25%) by ~10x, a pattern that can reflect mild paralog-mapping inflation in whole-genome data; however, AFR AF ~8% is concordant across three datasets with hundreds of homozygotes and remains far above the >5% threshold even under substantial deflation
• NF1 CSPEC v1.0 rule payload was empty at retrieval; if the official panel BA1 threshold becomes available it should be re-checked against this call

Sources

Reasoning

Generic ACMG/AMP 2015 BS1 (PMID:25741868): an allele frequency in a control population greater than expected for the disorder is strong support for a benign interpretation. NF1 (neurofibromatosis type 1) is autosomal dominant with near-complete penetrance by adulthood and a birth incidence/prevalence of roughly 1/2500-1/3000, so the expected pathogenic allele frequency is on the order of 0.02-0.05%. Observed frequencies vastly exceed this: gnomAD v4.1 total AF 0.45% (about 10-20x expected) and AFR AF 8.31% (hundreds-fold); gnomAD v2.1 total AF 0.77% and AFR AF 7.94%; gnomAD-Canada overall AF 0.61% and AFR AF 10%. BS1 is met at strong. Note: BA1 is also met and, per the ACMG/AMP tiering on the frequency axis, BA1 (stand-alone) takes precedence in final aggregation; BS1 should not be counted as an additional strong benign criterion.

Rule

generic ACMG/AMP 2015 BS1 (PMID:25741868): allele frequency greater than expected for the disorder -> strong benign

Evidence considered

• ACMG/AMP 2015 (PMID:25741868): 'In general, an allele frequency in a control population that is greater than expected for the disorder (Table 6) is considered strong support for a benign interpretation for a rare Mendelian disorder (BS1) or if over 5% then it is considered as stand-alone support (BA1).'
• NF1 disease context: autosomal dominant, near-complete penetrance, prevalence ~1/3000 -> expected pathogenic allele frequency ~0.02-0.05%; observed gnomAD v4.1 total AF 0.0045 (0.45%) is ~10-20x expected and AFR AF 0.0831 is hundreds-fold above expected
• gnomAD v2.1 total AF 0.0077 (0.77%) and AFR AF 0.0794 likewise far exceed the AF expected for NF1; gnomAD-Canada overall AF 0.0061 (0.61%) and AFR AF 0.10 corroborate

Still missing

• Expected allele frequency for NF1 (~0.02-0.05%) is derived from standard gene-level disease context (AD, near-complete penetrance, prevalence ~1/3000); no variant-specific NF1 prevalence publication was present in the case bundle
• BA1 is also met and subsumes BS1 in the final combination — BS1 should not be double-counted

Sources

Reasoning

Generic ACMG/AMP 2015 BS2 (PMID:25741868): if the disease is fully penetrant at an early age and the variant is observed in a healthy adult (homozygous for a recessive, heterozygous for a dominant condition), this is strong evidence for benignity. NF1 is autosomal dominant and near-fully penetrant by adulthood, so biallelic (homozygous) presence of a pathogenic NF1 allele in a general population cohort at this scale is incompatible with pathogenicity. gnomAD v4.1 reports 310 homozygotes (305 in AFR), gnomAD v2.1 reports 95 homozygotes (all in AFR), and gnomAD-Canada v1.0 reports 5 homozygotes; the gnomAD-Canada age histogram shows homozygotes up to 60-75 years of age. Homozygote counts are consistent with Hardy-Weinberg expectation for the AFR AF (~0.083), supporting genuine polymorphism rather than artifact. BS2 met at strong.

Rule

generic ACMG/AMP 2015 BS2 (PMID:25741868): observed in a healthy adult (or homozygous) for a fully penetrant dominant disorder expected to manifest at an early age -> strong benign

Evidence considered

• ACMG/AMP 2015 (PMID:25741868): 'Furthermore, if the disease under investigation is fully penetrant at an early age and the variant is observed in a well-documented healthy adult individual for a recessive (homozygous), dominant (heterozygous), or X-linked (hemizygous) condition then this is considered strong evidence for a benign interpretation (BS2).'
• gnomAD v4.1: 310 homozygotes total (305 of 74954 AFR alleles), gnomAD v2.1: 95 homozygotes (all in AFR, 1976/24898 alleles), gnomAD-Canada v1.0: 5 homozygotes (112/18418 alleles) — homozygous observation of a fully penetrant dominant-disease allele at this scale strongly supports benignity
• gnomAD-Canada v1.0 age histogram for homozygotes shows carriers across ages 50-75 (age_hist_hom bins), consistent with healthy adult carriers

Still missing

• gnomAD is a general population cohort, not a phenotype-ascertained 'well-documented healthy adult' cohort; BS2 here rests on homozygote observations in a general population database (standard ACMG practice), not on documented healthy-adult status of each carrier
• Homozygosity signal is concentrated in AFR (305/310 homozygotes in gnomAD v4.1); if the segdup flag reflects mapping error the homozygote count could be inflated, though 310 homozygotes far exceeds any plausible NF1-affected count

Sources

Reasoning

The variant is synonymous (p.(Gly848=)) and is therefore evaluated on the SpliceAI path only (generic ACMG fallback, since the NF1 VCEP CSPEC was retrieved without a criterion rule payload and no gene-specific PP3/BP4 lookup spreadsheet exists). SpliceAI max delta is 0.079, below the 0.1 threshold that defines no predicted splice impact (SVI-recommended SpliceAI thresholds; tool per Jaganathan et al. 2019, PMID 30661751), meeting BP4 at supporting strength.

Rule

Generic ACMG fallback (SVI SpliceAI calibration, generic_acmg_combination_rules): SpliceAI max delta < 0.1 -> BP4 (supporting) for non-missense variants; met because max delta = 0.079.

Evidence considered

• SpliceAI lookup (spliceai): DS_AG 0.029, DS_AL 0.079, DS_DG 0.032, DS_DL 0.016; max delta 0.079, with evidence sentence 'SpliceAI predicts no significant splice impact for this variant (max delta score = 0.08)'.
• Generic fallback calibration (generic_acmg_combination_rules): for synonymous (non-missense) variants use SpliceAI only; 'SpliceAI max delta < 0.1 -> BP4 (supporting)'. Max delta 0.079 < 0.1, so BP4 (supporting) is met (SVI-recommended thresholds; SpliceAI tool per Jaganathan et al. 2019, PMID 30661751).
• NF1 VCEP CSPEC (cspec) was consulted and carries no PP3/BP4 rule payload (criteria list empty, framework_complete=false); no gene-specific BP4 assignment overrides the generic calibration.
• The SpliceAI prediction is counted once, for BP4 only; it is not also applied to PP3 or BP7.

Sources

Reasoning

Generic ACMG/AMP 2015 PM2 (PMID:25741868): a variant absent from (or below the expected carrier frequency if recessive) a large, race-matched general population or control cohort is moderate (supporting) evidence of pathogenicity. This variant is common: gnomAD v4.1 AF = 0.45% (7273/1611892 alleles), gnomAD v2.1 AF = 0.77% (2173/281920 alleles), and up to 8.3-10% in the AFR population — it is neither absent nor at extremely low frequency, so PM2 is directly refuted. Additionally, BA1/BS1 are met, and PM2 is mutually exclusive with BA1/BS1 for the same variant on the frequency axis. PM2 not met.

Rule

generic ACMG/AMP 2015 PM2 (PMID:25741868): absent from (or below expected carrier frequency if recessive) a large race-matched general population or control cohort -> moderate (supporting) pathogenic evidence; not met when the variant is present at high frequency

Evidence considered

• ACMG/AMP 2015 (PMID:25741868): 'If a variant is absent from (or below the expected carrier frequency if recessive) a large general population or a control cohort (>1000 individuals) and the population is race-matched to the patient harboring the identified variant, then this observation can be considered a moderate piece of evidence for pathogenicity (PM2).'
• gnomAD v4.1: variant present at 7273/1611892 alleles (AF 0.0045, 0.45%) with 310 homozygotes; gnomAD v2.1: 2173/281920 alleles (AF 0.0077, 0.77%) with 95 homozygotes — the variant is common, directly contradicting the 'absent or extremely low frequency' prerequisite of PM2
• BA1 and BS1 are met for the same variant; PM2 cannot co-apply with BA1/BS1 on the population-frequency axis

Sources

Reasoning

The variant NM_001042492.2:c.2544G>A is a synonymous (silent) change (NP_001035957.1:p.(Gly848=)). The NF1 VCEP CSPEC was retrieved but carries no criterion-level rule payload (criteria.json empty, framework_complete=false) and no gene-specific PP3/BP4 lookup spreadsheet exists in this case, so the generic ACMG fallback calibration governs: non-missense variants are evaluated on the SpliceAI path only (single-mechanism rule). SpliceAI max delta is 0.079 (DS_AG 0.029, DS_AL 0.079, DS_DG 0.032, DS_DL 0.016), far below the PP3 threshold of max delta > 0.2 (SVI-recommended SpliceAI thresholds encoded in the generic rules; tool per Jaganathan et al. 2019, PMID 30661751). The missense in-silico path (REVEL/BayesDel/aGVGD) is not applicable to a synonymous variant, and no such scores are available (REVEL lookup returned found:false; BayesDel has no verified published threshold). No computational evidence in the case bundle supports a deleterious effect, so PP3 is not met.

Rule

Generic ACMG fallback (SVI SpliceAI calibration, generic_acmg_combination_rules): SpliceAI max delta > 0.2 -> PP3 (supporting) for non-missense variants; not met because max delta = 0.079.

Evidence considered

• Variant is a synonymous (silent) change, NP_001035957.1:p.(Gly848=); per the governing generic ACMG fallback calibration (generic_acmg_combination_rules), non-missense variants are evaluated with the SpliceAI path only — exactly one mechanism per variant, never combining missense-path and splice-path scores.
• SpliceAI lookup (spliceai): DS_AG 0.029, DS_AL 0.079, DS_DG 0.032, DS_DL 0.016; max delta 0.079 (evidence sentence: 'SpliceAI predicts no significant splice impact for this variant (max delta score = 0.08)'). This is well below the PP3 threshold of SpliceAI max delta > 0.2 (SVI-recommended SpliceAI thresholds as encoded in generic_acmg_combination_rules; SpliceAI tool per Jaganathan et al. 2019, PMID 30661751), so the splice path gives no support for PP3.
• Missense in-silico path (REVEL/BayesDel/aGVGD) is not applicable to this synonymous variant under the single-mechanism rule; regardless, no scores are available in this case: REVEL v1.3 lookup returned found:false (queried at GRCh38 17-31229159-G-A) and BayesDel has no verified published threshold for PP3/BP4 in this pipeline.
• No other computational evidence in the case bundle (conservation, evolutionary, splicing) supports a deleterious effect; PP3 not met.

Sources

Reasoning

PP5 (reputable source recently reports variant as pathogenic without published evidence) applies in this pipeline only for an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic. The ClinVar submission audit for VCV000183949 (NM_001042492.3(NF1):c.2544G>A (p.Gly848=)) shows submissions_total 21, expert_panel_submissions 0, validated_pmids 0, criterion_lead_submissions 0; every submission is an ordinary clinical-testing laboratory with single-submitter review status. The aggregate ClinVar label is 'Benign' (2 review stars) from 20 clinical laboratories — an aggregate, non-expert assertion. Per the adjudication rule, ordinary laboratory submissions and aggregate ClinVar labels must never trigger PP5, and the label direction (Benign) would not support PP5 in any case. PP5 is therefore not met.

Evidence considered

• ClinVar submission audit (VCV000183949): submissions_total 21, usable_for_classification 18, expert_panel_submissions 0, validated_pmids 0, criterion_lead_submissions 0; all submitters are clinical laboratories (e.g., Ambry Genetics, GeneDx, Labcorp, Illumina, Mayo Clinic, ARUP), none is an expert panel
• ClinVar aggregate classification 'Benign', review_status 'criteria provided, multiple submitters, no conflicts', review_stars 2; per group rule aggregate ClinVar labels never trigger PP5
• cspec: NF1 Expert Panel v1.0 framework identified but no rule payload available; no VCEP-specific PP5 modification could be applied, so the generic PP5 rule (expert-panel classification required) governs

Still missing

• No ClinGen expert-panel (3-star) classification exists for this exact variant; none of the 21 ClinVar submissions is from an expert panel

Sources

Reasoning

BP6 (reputable source recently reports variant as benign without published evidence) applies in this pipeline only for an exact-variant ClinVar expert-panel classification of Benign/Likely benign. Although the ClinVar aggregate label for VCV000183949 is 'Benign', the submission audit shows expert_panel_submissions 0: the label is an aggregate of 20 ordinary clinical-laboratory submissions (review stars 2, 'criteria provided, multiple submitters, no conflicts'), none from an expert panel, and 0 validated PMIDs. Per the adjudication rule, ordinary laboratory submissions and aggregate ClinVar labels must never trigger BP6, regardless of how concordant the Benign label is. BP6 is therefore not met. (The benign assessment is instead supported by population-frequency evidence under other criteria groups, e.g., gnomAD frequencies and the synonymous, non-splice-impacting nature of the variant.)

Evidence considered

• ClinVar submission audit (VCV000183949): expert_panel_submissions 0; 18 usable submissions all from clinical laboratories with single-submitter review status; aggregate 'Benign' with review_stars 2
• Per group rule, aggregate ClinVar labels and ordinary laboratory submissions never trigger BP6; only an exact-variant ClinVar expert-panel Benign/LB classification would
• cspec: NF1 Expert Panel v1.0 framework identified but rule payload unavailable; no VCEP-specific BP6 modification could be applied

Still missing

• No ClinGen expert-panel (3-star) Benign/Likely benign classification exists for this exact variant; the concordant Benign label is from clinical laboratories only and is excluded from BP6 by rule

Sources

Reasoning

No proband or parental genotype data is present anywhere in the case bundle. PS2 requires a de novo occurrence of the variant in an affected proband with confirmed maternity and paternity (ACMG/AMP 2015, PMID 25741868). The ClinVar record for VCV000183949 (20 germline submissions, all Benign) contains no de novo annotation, no parental-testing results, and no proband phenotype information, so a de novo event cannot be confirmed or excluded. No VCEP-specific PS2 rule was retrievable (NF1 cspec v1.0 record found but rule payload empty), so the generic ACMG/AMP 2015 definition governs. Evidence is insufficient; criterion not assessed.

Rule

ACMG/AMP 2015 (PMID 25741868): PS2 = de novo variant in the proband with maternity and paternity confirmed. No applicable NF1 VCEP-specific rule text was retrievable from the cspec record in this case.

Evidence considered

• ClinVar VCV000183949 submission records were reviewed in full for de novo or parental-testing annotations: all 20 submissions list origin 'germline' and none report de novo status, parental testing, or parentage confirmation; the retrieved record contains no such fields.

Still missing

• No proband genotype/phenotype data in the case bundle
• No parental testing results (required to confirm de novo status and maternity/paternity)
• NF1 cspec v1.0 ruleset payload was empty in this case; a VCEP-specific PS2 rule, if any, could not be applied

Sources

Reasoning

PS3 (ACMG/AMP 2015, PMID:25741868) requires well-established functional studies demonstrating a deleterious effect on gene/protein function. The governing NF1 framework is the ClinGen Neurofibromatosis and Schwannomatosis Expert Panel Specification v1.0 (cspec), but its rule payload was not retrievable in this case (vcep_materials.json: rule_payload empty, framework_complete=false), so the generic ACMG/AMP 2015 PS3 definition applies. No functional assay evidence for NM_001042492.2:c.2544G>A exists anywhere in the case bundle. The only functional-flagged item is a ClinVar single-submitter note (SCV000696390, LabCorp) stating that 3/5 splice prediction tools predict gain of a splice acceptor site 'however these predictions have not been confirmed by functional studies' — i.e., an unconfirmed in silico splice prediction, which explicitly does not constitute a functional study and cannot support PS3 (and its direction conflicts with SpliceAI, which predicts no significant splice impact, max delta 0.079). OncoKB lists NF1 G848= as 'Unknown Oncogenic Effect' with non-variant-specific context; COSMIC somatic count (n=4) is an observation, not an assay; the ClinVar Benign aggregate (20 submissions) is a final classification and is never used to infer a criterion. All five full-text papers in the case publication set were searched for the variant ('2544', 'G848', 'Gly848': 0 hits); the two functional-theme papers (PMID:25741868, PMID:24033266) are methodology papers with no variant-specific assay data. Because no well-established functional study of this variant is available, PS3 is not assessed and the evidence gap is recorded.

Rule

generic_acmg_2015_ps3: well-established functional studies show a deleterious effect (PMID:25741868); NF1 VCEP rule payload unavailable in this case

Evidence considered

• No functional assay evidence for NM_001042492.2:c.2544G>A (p.(Gly848=)) is present anywhere in the case evidence bundle.
• ClinVar submission SCV000696390 (LabCorp) states '3/5 splice prediction tools predict the gain of a splice acceptor site, however these predictions have not been confirmed by functional studies' — an in silico splice prediction explicitly lacking functional confirmation; does not satisfy PS3.
• SpliceAI lookup predicts no significant splice impact: max delta 0.079 (DS_AL 0.079, DS_AG 0.029, DS_DG 0.032, DS_DL 0.016) — in silico prediction only, not a functional study.
• OncoKB lists NF1 G848= as 'Unknown Oncogenic Effect' with non-variant-specific curated literature; no variant-specific functional assay data.
• All five full-text publications in the case set were searched for '2544', 'G848', 'Gly848' (0 hits); none report functional studies of this variant. Functional-theme papers PMID:25741868 and PMID:24033266 are methodology papers and contain no variant-specific assay data.
• ClinVar Benign classification (20 submissions, 15 submitters) is a final classification and was not used to infer PS3/BS3, per policy that a criterion is never inferred from another source's final classification.

Still missing

• No validated functional assay data for c.2544G>A (no minigene/RT-PCR splicing assay, no neurofibromin GAP-activity, stability, or expression assay) — PS3 cannot be applied until such evidence is available

Sources

Reasoning

PS4 requires significantly increased prevalence of the variant in affected individuals versus controls (case-control enrichment). The case bundle contains no case-control study, no affected-cohort allele counts, and no proband cohort data for NF1 c.2544G>A. The only prevalence data available are control-population frequencies: gnomAD v4.1 total AF 0.451% (7273/1611892 alleles, 310 homozygotes, AFR AF 8.31%) and gnomAD v2.1 total AF 0.771% (2173/281920 alleles, 95 homozygotes, AFR AF 7.94%). These high control frequencies, including multiple homozygotes, are incompatible with a disease-causing enrichment signal, but because no affected-individual prevalence data exist the criterion cannot be formally evaluated. Literature_pass returned only NF1/cancer genetic-counseling practice guidelines and ACMG standards papers (PMID 17636453, 25394175, 32602153, 25741868, 24033266); none provide case-prevalence data and none mention this variant.

Evidence considered

• gnomAD v4.1 (17-31229159-G-A): total AF 0.00451209 (7273/1611892 alleles), 310 homozygotes, grpmax FAF 0.0813264, highest AF in African/African American (0.0830509); reviewed as control-population context, inconsistent with PS4 enrichment
• gnomAD v2.1 (17-29556177-G-A): total AF 0.00770786 (2173/281920 alleles), 95 homozygotes, grpmax FAF 0.0771056, highest AF in African/African American (0.0793638); reviewed as control-population context
• No case-control study or affected-individual prevalence data for this variant were found in review_manifest.json, case_summary.json, prefetch.json, evidence.json, or literature_pass.json
• Retrieved literature (PMID 17636453, 25394175, 32602153) consists of genetic-counseling practice guidelines with no cohort or variant-level prevalence data for c.2544G>A

Still missing

• No case-control or affected-cohort allele-count data for NF1 c.2544G>A (e.g., NF1 patient cohort sequencing study) are available in this case bundle
• No proband phenotype or indication data that would define an affected case group

Sources

Reasoning

PM6 requires an assumed de novo occurrence without confirmation of parental identity (ACMG/AMP 2015, PMID 25741868). No proband genotype data and no parental testing results are present in the case bundle; the ClinVar record contains no de novo or parentage annotations. There is therefore no basis to assume a de novo event. Evidence is insufficient; criterion not assessed.

Rule

ACMG/AMP 2015 (PMID 25741868): PM6 = assumed de novo variant without confirmation of maternity and paternity. No applicable NF1 VCEP-specific rule text was retrievable from the cspec record in this case.

Evidence considered

• ClinVar VCV000183949 submission records were reviewed in full for de novo or parental-testing annotations: none of the 20 submissions (all origin 'germline') report de novo status or parental testing.

Still missing

• No proband genotype data in the case bundle
• No parental testing results to support an assumed de novo occurrence
• NF1 cspec v1.0 ruleset payload was empty in this case; a VCEP-specific PM6 rule, if any, could not be applied

Sources

Reasoning

PP1 requires co-segregation of the variant with disease in multiple affected family members (ACMG/AMP 2015, PMID 25741868). No family, pedigree, or segregation data of any kind exists in the case bundle: no affected relatives genotyped, no meioses counted, and no segregation observations in the ClinVar record or any retrieved publication. Evidence is insufficient; criterion not assessed.

Rule

ACMG/AMP 2015 (PMID 25741868): PP1 = co-segregation with disease in multiple affected family members in a gene definitively known to cause the disease. No applicable NF1 VCEP-specific rule text was retrievable from the cspec record in this case.

Evidence considered

• ClinVar VCV000183949 was reviewed for segregation annotations: none of the 20 submissions report segregation in affected family members; no family studies are referenced in the retrieved record.
• None of the five full-text publications available in the case (24033266, 25741868, 17636453, 32602153, 25394175) report segregation of c.2544G>A with NF1; they are general methodology or practice-guideline documents without variant-specific family data.

Still missing

• No segregation data in affected family members (meioses count unavailable)
• No pedigree or family testing records in the case bundle
• NF1 cspec v1.0 ruleset payload was empty in this case; a VCEP-specific PP1 rule, if any, could not be applied

Sources

Reasoning

PP4 requires that the variant's phenotype or family history is highly specific for the gene's disease (here, neurofibromatosis type 1). The case bundle contains no proband phenotype data whatsoever: review_manifest.json and case_summary.json carry only the variant string and pipeline metadata; there is no clinical features list, no family history, and no phenotype file. The NF1 VCEP specification (cspec, v1.0) was identified but its rule payload was not retrievable (framework_complete: false, criteria: [], rule_payload: {}), so no gene-specific phenotype-specificity rule could be applied. With no phenotype information, PP4 cannot be assessed.

Evidence considered

• No phenotype source exists in the bundle; review_manifest.json and case_summary.json contain only variant and pipeline metadata
• vcep_materials.json confirms the NF1 Expert Panel v1.0 framework was found but rule_payload is empty and framework_complete is false, so generic ACMG/AMP 2015 (PMID 25741868) PP4 definition was used as the evaluation frame

Still missing

• No proband clinical phenotype, diagnostic criteria status (e.g., NIH/NF1 diagnostic criteria), or family history present in the case bundle
• NF1 VCEP (cspec doc 1553923915) rule payload unavailable (framework_complete: false), so no VCEP-specific PP4 specification could be consulted

Reasoning

BS3 (ACMG/AMP 2015, PMID:25741868) requires well-established functional studies showing no damaging effect on protein function or splicing. The governing NF1 CSPEC v1.0 rule payload was not retrievable in this case (framework_complete=false), so the generic definition applies. No well-established functional study exists in the bundle for c.2544G>A. SpliceAI predicts no significant splice impact (max delta 0.079, below the commonly used 0.10 threshold), but this is an in silico prediction, not a functional assay, and in silico splicing prediction alone cannot satisfy the BS3 evidence bar; the ACMG/AMP framework (PMID:25741868) restricts BS3 to assay-based evidence. The only functional-flagged ClinVar statement (SCV000696390) explicitly says splice predictions were not confirmed by functional studies, so it provides no BS3 support either. Population frequency (gnomAD v2.1 total AF 0.77%, AFR AF 7.9%; gnomAD v4.1 AFR AF 8.3%, 310 homozygotes) and the ClinVar Benign aggregate are evaluated by other criteria groups (BA1/BS1; PP5/BP6) and are not functional evidence. Because the required functional study is absent, BS3 is not assessed and the evidence gap is recorded.

Rule

generic_acmg_2015_bs3: well-established functional studies show no damaging effect on protein function or splicing (PMID:25741868); NF1 VCEP rule payload unavailable in this case

Evidence considered

• BS3 requires well-established functional studies showing no damaging effect on protein function or splicing (ACMG/AMP 2015, PMID:25741868); no such study exists in the case evidence for c.2544G>A.
• SpliceAI max delta 0.079 (<0.10) is a negative in silico splicing prediction; it cannot substitute for a functional assay and does not meet the BS3 evidence bar.
• The only functional-flagged statement in ClinVar (SCV000696390, LabCorp) explicitly says the splice predictions were not confirmed by functional studies, so it provides neither BS3 nor PS3 support.
• gnomAD v2.1 (total AF 0.00771, AFR AF 0.0794, 95 homozygotes) and gnomAD v4.1 (total AF 0.00451, AFR AF 0.0831, 310 homozygotes) population frequencies and the ClinVar Benign aggregate are handled by other criteria groups and were not used as functional evidence.

Still missing

• No well-established functional study demonstrating absence of a damaging effect on protein function or splicing for c.2544G>A (e.g., minigene/RT-PCR splicing assay, neurofibromin protein function assay); the negative SpliceAI in silico result alone is insufficient for BS3

Sources

Reasoning

BS4 requires lack of segregation with disease in affected members of a family (i.e., affected individuals who do not carry the variant) (ACMG/AMP 2015, PMID 25741868). No affected family members were tested in any source in the case bundle, so no non-segregation observations exist to evaluate. Note: the high population frequency of this variant (gnomAD v2.1 AF 0.77%, grpmax FAF 0.077; gnomAD v4.1 AF 0.45%, AFR AF 8.3%, 310 homozygotes) is population-frequency evidence relevant to BA1/BS1, not to BS4, and was not used to support this criterion. Evidence is insufficient; criterion not assessed.

Rule

ACMG/AMP 2015 (PMID 25741868): BS4 = lack of segregation in affected members of a family. No applicable NF1 VCEP-specific rule text was retrievable from the cspec record in this case.

Evidence considered

• ClinVar VCV000183949 was reviewed for non-segregation annotations: none of the 20 submissions report affected family members lacking the variant; no family studies are referenced in the retrieved record.
• gnomAD v2.1 and v4.1 frequencies were considered but are population-frequency evidence (BA1/BS1 domain), not BS4 non-segregation evidence, and therefore do not contribute to this criterion.

Still missing

• No affected family members tested without the variant (non-segregation observations unavailable)
• No pedigree or family testing records in the case bundle
• NF1 cspec v1.0 ruleset payload was empty in this case; a VCEP-specific BS4 rule, if any, could not be applied

Sources

Reasoning

BP2 is met when a variant is observed in trans with a pathogenic variant in the context of a dominant disorder (supporting evidence for a benign role), or observed in cis with a pathogenic variant. No cis/trans phase evidence for NM_001042492.2:c.2544G>A exists in any bundle source: the ClinVar record (VCV000183949, 20 germline submissions from 15 submitters, all Benign, review status 'criteria provided, multiple submitters, no conflicts') provides no phase observations and yields 0 usable criterion-lead submissions; no publication in the case bundle mentions this variant; and no proband/family segregation or phase data was retrieved. gnomAD v4.1 homozygosity (310 homozygotes; AFR AF 8.3%) reflects population frequency and is not a documented in-trans observation with a pathogenic variant, so it cannot satisfy BP2. Evidence is insufficient to assess BP2; the required phase data is absent.

Rule

BP2: Observed in trans with a pathogenic variant for a fully dominant disease/mechanism; or observed in cis with a pathogenic variant for any mechanism (ACMG/AMP 2015, PMID 25741868).

Evidence considered

• BP2 (ACMG/AMP 2015, PMID 25741868): in a dominant disorder, 'the detection of a variant in trans with a pathogenic variant can be considered supporting evidence for a benign impact (BP2)'; additionally 'finding the second variant in cis would be supporting evidence, though not definitive, for a benign role (BP2)'.
• ClinVar VCV000183949 was reviewed for phase data: 18 usable submissions from 20 clinical laboratories, all classified Benign (review status 'criteria provided, multiple submitters, no conflicts'); the submission audit found 0 expert-panel submissions, 0 criterion-lead submissions, and 0 validated PMIDs, and no submission reports cis/trans phase with a pathogenic variant.
• NF1 CSPEC (doc 1553923915) confirms the dominant-disease context (autosomal dominant inheritance) in which BP2's in-trans-with-pathogenic clause would apply.
• gnomAD v4.1 homozygote counts (310 total, 305 in the AFR population) are population-frequency evidence, not a documented in-trans observation with a pathogenic NF1 variant, and therefore cannot be used to meet BP2.
• No literature in the case bundle (literature_pass.json, fulltext_map.json) mentions c.2544G>A/p.Gly848=, so no publication-based phase or co-occurrence observation exists.

Still missing

• No proband/family phase data (variant in cis or trans with a pathogenic NF1 variant) is available for this variant in any bundle source; targeted family testing or a clinical report documenting phase with a pathogenic NF1 variant would be required to evaluate BP2.

Sources

Reasoning

BP5 requires that an alternate molecular basis for disease (a different pathogenic variant or non-genetic cause) has been established in the proband. The case bundle contains no proband-level data of any kind — no other variant findings, no alternate diagnosis, no segregation or diagnostic workup results. Because there is no information about whether an alternate molecular basis exists, BP5 cannot be assessed. The high control frequency of this variant (gnomAD v4.1 total AF 0.451%, AFR AF 8.31%) is suggestive that it is a benign polymorphism, but that is population-frequency evidence (BS1/BA1 territory) and is not itself a documented alternate molecular basis for a specific proband.

Evidence considered

• review_manifest.json, case_summary.json, prefetch.json, and evidence.json contain no proband genetic findings beyond the index variant and no alternate-diagnosis information
• No source in the bundle documents an alternate molecular basis for any affected individual carrying this variant

Still missing

• No proband-level evidence of an alternate molecular basis or alternate diagnosis (e.g., a second pathogenic variant or confirmed non-genetic etiology) is present in the case bundle

Reasoning

The variant is synonymous (p.(Gly848=)) and SpliceAI predicts no splice impact (max delta 0.079 < 0.1), satisfying the first component of BP7 (ACMG 2015, PMID 25741868: a synonymous variant for which splicing prediction algorithms do not predict impact on the splice consensus sequence nor creation of a new splice site). However, the second component of BP7 — 'AND the nucleotide is not highly conserved' — cannot be verified: no conservation score (e.g., PhyloP/GERP) is present anywhere in the case bundle, and the NF1 VCEP CSPEC was retrieved without a rule payload, so no gene-specific BP7 modification is available. Population frequency (gnomAD AF up to 8.3% in AFR) is population-based evidence assessed under BA1/BS1 by another group and is not a conservation score; it is not used here to satisfy BP7's conservation component. BP7 is therefore not assessed pending conservation data.

Rule

ACMG 2015 BP7 (PMID 25741868): synonymous (silent) variant with no predicted splice impact AND nucleotide not highly conserved; splice component met (SpliceAI max delta 0.079 < 0.1), conservation component unverifiable -> not assessed.

Evidence considered

• Variant is synonymous: NP_001035957.1:p.(Gly848=) (mutalyzer normalization p.(=)); gnomAD-Canada VEP most_severe_consequence = 'synonymous_variant'.
• SpliceAI lookup (spliceai): max delta 0.079 (DS_AG 0.029, DS_AL 0.079, DS_DG 0.032, DS_DL 0.016) — no predicted impact on the splice consensus sequence and no new splice site predicted; satisfies the splicing component of BP7.
• BP7 definition (ACMG 2015, PMID 25741868): 'A synonymous (silent) variant for which splicing prediction algorithms do not predict impact on the splice consensus sequence nor the creation of a new splice site AND the nucleotide is not highly conserved.' The 'nucleotide is not highly conserved' component is unverifiable here: no PhyloP/GERP/conservation data in the case bundle.
• Population frequency (gnomAD v4.1 total AF 0.451%, AFR AF 8.31%, 310 homozygotes) demonstrates polymorphism but is population-frequency evidence that belongs to the BA1/BS1 assessment path; it is not a conservation score and is not used to satisfy BP7's conservation component.
• NF1 VCEP CSPEC (cspec) consulted: no BP7 rule or modification retrieved (criteria payload empty); generic BP7 definition applies.

Still missing

• No conservation score (e.g., PhyloP100way/GERP++) for the variant position is present in the case bundle — required to verify the 'nucleotide is not highly conserved' component of BP7.
• NF1 VCEP CSPEC (doc 1553923915) was retrieved with an empty criteria payload; no gene-specific BP7 modification or lookup entry is available to consult.

Sources