Five-step scale from benign to pathogenic. This variant is likely pathogenic. Select a step to read what it means.
This variantScenario
BenignPathogenic
Likely pathogenic.A likely pathogenic classification under the recorded framework.
In plain language
Gene-level background from the recorded interpretation.
Words with a dotted underline have a short definition. Hover or tap them.
About FGF3
FGF3 encodes a member of the fibroblast growth factor (FGF) family of signaling proteins, which promote cell growth, survival, and division and participate in embryonic development, morphogenesis, tissue repair, and inner ear formation. It is an oncogene: FGF3 amplification is frequently found in human tumors and may drive neoplastic transformation and tumor progression, and the gene was originally identified through its activation in virus-induced mammary tumors in mice. FGF3 also cooperates with other oncogenic signals, such as WNT1, to promote mammary tumor development, and its signaling is important for normal inner ear development.
LYFE Sciences is an AI system, and it can make mistakes.
Criteria may be applied incorrectly, sources may be misread, and a
confident-looking classification can still be wrong.
Double-check every criterion and its underlying evidence
before relying on any call.
01The reasoning
How it was classified
The recorded framework is Not recorded. The criteria and strengths shown below come from the saved assessment; this page does not recalculate the classification.
Recorded applied criteria
The applied criteria and their strengths are shown as assessed under the stated framework.
Met (moderate): gnomAD v4.1 allele frequency 0.0015% (zero homozygotes) is far below the 0.1% recessive-disorder cutoff.
PM3Moderate
PM3 · moderate
Met (moderate): the variant is in trans with p.Y49C in an affected compound-heterozygous proband and homozygous in additional affected individuals.
PP1Supporting
PP1 · supporting
Met (supporting): cosegregates with LAMM syndrome, homozygous or compound heterozygous in 7 or more affected individuals across 3 families.
PP4Supporting
PP4 · supporting
Met (supporting): the LAMM phenotype (labyrinthine aplasia, microtia, microdontia) is highly specific to biallelic FGF3 mutations and is seen with this exact variant.
Benign criteria applied · 0
No benign criterion marked as applied in the recorded assessment.
Recorded classification: Likely pathogenic
Likely Pathogenic: two moderate (PM2 ultra-rare frequency, PM3 in trans) plus two supporting (PP1 cosegregation, PP4 specific phenotype) criteria map to Likely Pathogenic under ACMG/AMP 2015 with no conflicting benign evidence.
02The sources
Evidence at a glance
Evidence summaries and captured source records, where available. Missing entries should not be interpreted as negative findings.
gnomAD v4.1
gnomAD v2.1
Population frequency
24 / 1,613,542 alleles
gnomAD v4.1: 24 / 1,613,542 alleles. Recorded AF 1.48741e-05. Recorded homozygote count 0.
gnomAD v2.1
This variant is present in gnomAD v2.1 (AF= 2.38588e-05; MAF= 0.00239%, 6/251480 alleles, homozygotes = 0) and has highest observed frequency in the South Asian population (AF= 6.53253e-05; MAF= 0.00653%, 2/30616 alleles, homozygotes = 0); grpmax FAF= 1.121e-05.
gnomAD-Canada
This variant is present in gnomAD-Canada v1.0 (AF= 0.00010866021949364337, 2/18406 alleles, homozygotes = 0).
ACMG/AMP criteria under Not recorded, with the data and sources behind each one.
28 criteria with recorded statuses: 4 met · 8 not met · 4 not assessed · 12 not applicable. Open any row for the full reasoning, the rule and the evidence behind it.
No ClinGen CSPEC/VCEP specification exists for FGF3 (vcep_materials.json: no cspec url, framework_complete=false, precedence_rank 3), so the generic ACMG/AMP 2015 framework (PMID:25741868) governs this group.
FGF3-LAMM (OMIM 610706) is an autosomal recessive disorder requiring biallelic mutations; PS2 and PM6 (de novo criteria) are therefore not applicable: a single de novo allele cannot cause the phenotype, and no de novo occurrence of c.310C>T is reported - all transmissions are from heterozygous carrier parents.
PP1 is met at supporting strength: c.310C>T cosegregates with LAMM in >=7 affected individuals across 3 independent families (3 homozygous in Tekin family B; 4 homozygous in Riazuddin family PKDF295; 1 compound-heterozygous in Sensi Patient B), with carrier parents heterozygous in every family.
BS4 is not met: no non-segregation is documented; unaffected heterozygous carriers are the expected AR carrier configuration, and no affected individual lacks the variant.
Generic ACMG/AMP 2015 fallback applies: no ClinGen CSPEC/VCEP and no gene-specific framework for FGF3 (framework_mode=generic_acmg; vcep_materials.json has no rule payload; final_classification_framework found=false).
PS3 not assessed: every full-text source in the case evidence set was reviewed for functional assay content; none reports a validated functional study of FGF3 c.310C>T (p.Arg104Ter). Available evidence is clinical cosegregation and presumptive/in-silico statements (PMID:17236138 'presumably truncates... nonfunctional product'; PMID:21306635 structural modeling of other mutations; PMID:21480479 lists the allele as truncating). No assay, no controls, no biological-relevance validation - insufficient for PS3.
BS3 not assessed: no functional study shows absence of a damaging effect on FGF3 protein function or splicing. SpliceAI (max delta 0.04) is in-silico and cannot satisfy BS3; the stop-gained null mechanism contradicts the BS3 premise and is owned by PVS1/PM4.
No criterion was inferred from ClinVar classifications or from other sources' final classifications; functional status rests solely on the presence/absence of validated assay evidence.
Strength calibration is moot: neither PS3 nor BS3 is met, so no PS3/BS3 strength assignment (e.g., assay-relevance-based up/downgrade) is made.
Framework: no ClinGen CSPEC/VCEP exists for FGF3 (gene_vcep_dir null, cspec not found); generic ACMG/AMP 2015 (PMID:25741868) per output/generic_acmg_classification_rules.md governs this group.
PS4 (case-control enrichment): not met. Multiple unrelated affected families carry c.310C>T (homozygous: Tekin family B, Riazuddin PKDF295; compound het: Sensi Patient B) and the variant was absent in 162 Pakistani normal-hearing controls, but no formal case-control study with statistical significance exists for this variant; the only quantitative control screen (Sensi 2011, 50 controls) covered novel missense variants, not p.R104X.
PP4 (phenotype specificity): met at supporting. LAMM syndrome (labyrinthine aplasia, microtia, microdontia; OMIM 610706) is a highly specific phenotype caused by biallelic FGF3 mutations - a single genetic etiology - and this exact variant is present in affected individuals with that phenotype.
PP5/BP6: not applicable. ClinVar variation 13838 has zero expert-panel submissions (2 stars, 'criteria provided, multiple submitters, no conflicts'); only ordinary laboratory/literature/research submissions are present, and the aggregate label is Pathogenic/Likely pathogenic. Neither criterion may be triggered by non-expert assertions.
BP5: not assessed. The evidence bundle contains no proband-level data on an alternate molecular basis for disease, so the criterion cannot be evaluated.
Case-observation counts for this variant (approx. 5-6 affected individuals across 3+ unrelated families reported in the literature reviewed here) were considered under PS4/PP4 but were not used to infer any other criterion from external final classifications.
Governing framework: generic ACMG/AMP 2015 (PMID:25741868) — no ClinGen CSPEC/VCEP exists for FGF3 (vcep_materials.json: final_classification_framework found=false, generic_acmg_fallback).
PM3 met (moderate): c.310C>T (p.R104X) detected in trans with p.Y49C in an affected LAMM proband with phase confirmed by parental testing (PMID:21480479), plus homozygous biallelic state in three affected members of family B (PMID:17236138), in the autosomal recessive disorder LAMM (OMIM 610706).
BP2 not met: no cis-configuration with a pathogenic variant is documented anywhere; the in-trans clause of BP2 applies only to fully dominant disorders, and LAMM is autosomal recessive.
PMID:21306635 (Riazuddin 2011) was reviewed for allelic evidence but does not mention c.310C>T/p.R104X in its full text (0 matches for R104/310C/Arg104); excluded from PM3/BP2 evidence per variant-grounding rules.
PMID:22993869 (GeneReviews) has no extractable full text in this case (PNG only) and was not used for PM3/BP2.
Generic ACMG/AMP 2015 fallback applies: no ClinGen CSPEC/VCEP and no gene-specific framework for FGF3 (framework_mode=generic_acmg; vcep_materials.json has no rule payload).
PP3 not met: SpliceAI predicts no splice impact (max delta 0.04 < 0.20); BayesDel/REVEL are missense predictors and were not applied to this stop-gained variant; the null mechanism is adjudicated under PVS1, not PP3.
BP4 not met: the variant is a stop-gained null change (p.Arg104Ter); computational 'no impact' evidence is inapplicable and the single negative SpliceAI prediction does not satisfy BP4's multiple-lines requirement.
BP7 is not applicable: BP7 is defined for synonymous (silent) variants only and this variant is a nonsense substitution; the deterministic prefilter already marked BP7 not_applicable (requires_synonymous). BP7 is outside this group's output schema and is noted here for completeness.
No double-counting: SpliceAI's negative splice prediction is not credited to PP3, BP4, or BP7 simultaneously; each criterion was adjudicated independently on the same underlying splice data.
No FGF3 VCEP/CSPEC specification exists for this case (vcep_materials.json: 'no cspec url available', gene_vcep_dir=null, precedence_rank=3, framework_mode=generic_acmg); generic ACMG/AMP 2015 (PMID:25741868) with non-VCEP gnomAD cutoffs (PM2 <0.1%, BS1 >0.3%, BA1 >1%) governs the population criteria.
The variant (FGF3 c.310C>T, p.Arg104Ter) is ultra-rare in population databases: gnomAD v2.1 total AF 0.00239% (6/251,480, 0 homozygotes), gnomAD v4.1 total AF 0.00149% (24/1,613,542, 0 homozygotes), gnomAD-Canada 0.0109% (2/18,406, 0 homozygotes); grpmax FAF 1.121e-05 / 5.818e-05.
BA1 is not met: the maximum observed allele frequency (0.033%, MID subpopulation in gnomAD v4.1) is ~30-fold below the 1% BA1 threshold.
BS1 is not met: the highest observed allele frequency (0.033%, MID) remains ~9-fold below the 0.3% BS1 threshold, and is consistent with a rare founder allele for a very rare recessive disorder rather than a benign polymorphism.
BS2 is not met: 0 homozygous carriers in gnomAD v2.1, v4.1 and gnomAD-Canada, and all documented homozygotes of p.Arg104Ter are affected with fully penetrant LAMM syndrome (PMID:17236138 family B; PMID:21306635 family PKDF295); only heterozygous carriers are unaffected, as expected for recessive inheritance.
PM2 is met at moderate strength: all observed allele frequencies (max 0.033%, MID) are below the 0.1% cutoff with 0 homozygotes, and the variant was absent from 162 ethnically matched Pakistani normal-hearing controls (PMID:21306635).
This group contributes one met pathogenic criterion (PM2, moderate) and no benign criteria toward the final classification.
No population criterion was inferred from any other source's final classification; the ClinVar Pathogenic/Likely pathogenic labels were used only to confirm the recessive disease context, and all determinations rest on gnomAD allele counts, homozygote counts, and primary literature control data.
No criteria match.
Met · 4
PM2ModerateMet (moderate): gnomAD v4.1 allele frequency 0.0015% (zero homozygotes) is far below the 0.1% recessive-disorder cutoff.
Details ▾Hide ▴
Reasoning
PM2 is met: the variant is at extremely low frequency in population databases, well below the non-VCEP PM2 cutoff (<0.1%). gnomAD v2.1 total AF = 0.00239% (6/251,480 alleles, grpmax FAF 1.121e-05), gnomAD v4.1 total AF = 0.00149% (24/1,613,542 alleles, grpmax FAF 5.818e-05), gnomAD-Canada AF = 0.0109% (2/18,406 alleles), with 0 homozygotes in all datasets. The highest subpopulation frequency is 0.033% (Middle Eastern, 2/6,062 alleles, gnomAD v4.1), still ~3-fold below the 0.1% cutoff, so no threshold artifact arises. For an autosomal recessive disorder PM2 explicitly permits very low (rather than zero) frequency ('below the expected carrier frequency if recessive'), and the observed frequencies are orders of magnitude below the 0.1% cutoff. The variant was also absent from 162 ethnically matched Pakistani normal-hearing controls (PMID:21306635). Applied at default moderate strength per generic ACMG/AMP 2015 in the absence of an FGF3 VCEP strength calibration.
Rule
Generic ACMG/AMP 2015 PM2 (Richards et al., PMID:25741868; source_registry key generic_acmg_combination_rules): '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).' Applied at the non-VCEP gnomAD cutoff of <0.1%. Met at moderate (default) strength: total AF 0.0015-0.0024%, max subpopulation 0.033% (MID v4.1), grpmax FAF 1.1-5.8e-05, 0 homozygotes; absent from 162 matched Pakistani controls.
Evidence considered
gnomad_v2: total AF 2.38588e-05 (0.00239%, 6/251,480 alleles), 0 homozygotes, grpmax FAF 1.121e-05
gnomad_v4: total AF 1.48741e-05 (0.00149%, 24/1,613,542 alleles), 0 homozygotes, grpmax FAF 5.818e-05; highest population MID 0.000329924 (0.033%, 2/6,062 alleles) — still below 0.1%
gnomad_canada: AF 0.00010866 (0.0109%, 2/18,406 alleles), 0 homozygotes
PMID:21306635: 'These mutations were not found in 162 ethnically matched Pakistani normal-hearing control individuals, indicating that the variants are not common polymorphisms' (p.R104X = c.310C>T among the tested mutations)
generic_acmg_combination_rules + PMID:25741868: PM2 per generic ACMG/AMP 2015 ('absent from (or below the expected carrier frequency if recessive) a large general population or a control cohort (>1000 individuals)'), applied at the non-VCEP gnomAD cutoff of <0.1% at default moderate strength
Still missing
gnomAD v2.1 genome counts are unavailable at this position (genome AN=0), so the v2.1 total is exome-only; gnomAD v4.1 joint (1,613,542 alleles) provides the larger corroborating dataset.
No large race-matched disease-specific control cohort beyond gnomAD and the 162-control Pakistani screen (PMID:21306635) was available; the 50 European controls in PMID:21480479 were screened for the two new missense variants and do not directly cover p.R104X.
PM3ModerateMet (moderate): the variant is in trans with p.Y49C in an affected compound-heterozygous proband and homozygous in additional affected individuals.
Details ▾Hide ▴
Reasoning
FGF3-related LAMM syndrome (OMIM 610706) is an autosomal recessive disorder; no VCEP/CSPEC exists for FGF3, so generic ACMG/AMP 2015 (PMID:25741868) governs. The variant c.310C>T (p.Arg104Ter/p.R104X) is observed biallelic in affected probands in two independent reports. (1) PMID:21480479 (Sensi 2011): affected proband 'Patient B' (Italian, non-consanguineous parents) is compound heterozygous for the paternal missense p.Y49C and the maternal nonsense p.R104X (this variant); phase is established by parental carrier testing (mother heterozygous for p.R104X, father heterozygous for p.Y49C, both unaffected), i.e., the variant is in trans with another FGF3 mutation in an affected individual. Per ACMG 2015, when two heterozygous variants are identified in a recessive disorder gene and one is known pathogenic, determining the other is in trans is moderate evidence for pathogenicity (PM3). (2) PMID:17236138 (Tekin 2007): all three affected members of family B are homozygous for c.310C>T (p.R104X) with parents heterozygous, i.e., the variant occupies both alleles (in trans with itself) in a fully penetrant autosomal recessive phenotype; this adds three biallelic affected-proband observations. p.R104X is a nonsense mutation (exon 2) predicted to truncate >50% of the protein. PM3 is met at the default moderate strength; an upgrade to strong would require the framework to count homozygous observations as multiple 'in trans with other pathogenic variants' observations (ClinGen SVI-style PM3 calibration), which is not available in this case's source set, so strength is conservatively kept at moderate.
Rule
PM3: For recessive disorders, detected in trans with a pathogenic variant (Moderate). ACMG/AMP 2015: when two heterozygous variants are identified in a gene for a recessive disorder and one is known to be pathogenic, determining that the other variant is in trans is moderate evidence for pathogenicity (PM3); may be upgraded to strong with multiple observations of the variant in trans with other pathogenic variants (PMID:25741868).
Evidence considered
Compound heterozygosity p.R104X (maternal) + p.Y49C (paternal) in an affected LAMM proband, phase confirmed by parental carrier genotypes (PMID:21480479)
Homozygous c.310C>T (p.R104X) in all three affected members of family B with heterozygous carrier parents, cosegregating with a fully penetrant autosomal recessive phenotype (PMID:17236138)
ACMG/AMP 2015 PM3 rule text: in-trans detection with a pathogenic variant in a recessive disorder = moderate evidence, upgradable to strong with multiple trans observations (PMID:25741868)
Still missing
The trans partner p.Y49C is a novel variant whose pathogenicity is argued by the authors (residue conservation, receptor-interaction-site location) rather than an independent P/LP classification; confirm p.Y49C carries an established P/LP classification before relying on PM3 at any strength.
Strength upgrade to strong depends on whether homozygous (biallelic) observations are accepted as multiple 'in trans with pathogenic variants' observations per ClinGen SVI PM3 calibration, which is not part of this case's governing source set.
PP1SupportingMet (supporting): cosegregates with LAMM syndrome, homozygous or compound heterozygous in 7 or more affected individuals across 3 families.
Details ▾Hide ▴
Reasoning
PP1 (Richards et al. 2015): cosegregation with disease in multiple affected family members in a gene definitively known to cause the disease. c.310C>T (p.Arg104Ter) cosegregates with LAMM syndrome under a fully penetrant autosomal recessive model: (1) homozygous in all three affected members of Turkish family B with both parents heterozygous carriers and unaffected relatives heterozygous or wild-type (PMID:17236138); (2) homozygous in four affected members (IV:1, IV:2, IV:7, IV:8) of Pakistani family PKDF295, co-segregating with deafness, with heterozygous carrier III-3 clinically and radiologically normal (PMID:21306635); (3) present on the maternal allele of compound-heterozygous Patient B (p.R104X/p.Y49C) whose both parents are unaffected carriers (PMID:21480479). In total >=7 affected individuals carry biallelic p.R104X across 3 independent families with consistent carrier-parent genotypes and no recombination. Under generic ACMG/AMP 2015 PP1 is applied at supporting strength (no FGF3 VCEP exists to justify an upgraded strength).
Rule
generic_acmg_2015 PP1 (cosegregation with disease in multiple affected family members) - supporting; recessive-model cosegregation in 3 families, 7 homozygous + 1 compound-heterozygous affected
Evidence considered
Tekin 2007 family B: 3 affected homozygous c.310C>T, parents heterozygous, unaffected relatives het/wild-type
Riazuddin 2011 family PKDF295: 4 affected homozygous c.310C>T co-segregating with deafness; heterozygous carrier III-3 normal
Sensi 2011 Patient B: compound heterozygote p.R104X/p.Y49C with maternal p.R104X; both parents unaffected carriers
PP4SupportingMet (supporting): the LAMM phenotype (labyrinthine aplasia, microtia, microdontia) is highly specific to biallelic FGF3 mutations and is seen with this exact variant.
Details ▾Hide ▴
Reasoning
LAMM syndrome (OMIM 610706; labyrinthine aplasia, microtia, microdontia) is a highly specific phenotype with a single genetic etiology: biallelic FGF3 mutations. This exact variant (p.R104X) is present in affected individuals with precisely this phenotype: homozygous in family B of Tekin 2007 (type I microtia, microdontia, profound congenital deafness with Michel aplasia - complete absence of inner ear structures), homozygous in family PKDF295 of Riazuddin 2011 (bilateral complete labyrinthine aplasia, microtia, microdontia), and compound heterozygous in Patient B of Sensi 2011 (labyrinthine aplasia, microtia, microdontia). ClinVar lists the condition for this exact variant as 'Deafness with labyrinthine aplasia, microtia, and microdontia'. The phenotype is strongly suggestive of a gene with a single genetic etiology, satisfying PP4 at supporting strength.
Rule
Generic ACMG/AMP 2015 (PMID:25741868) PP4: Variant is present in a case with a highly specific phenotype or family history that is strongly suggestive of a gene with a single genetic etiology (supporting).
Evidence considered
Phenotype of 9 affected individuals (microtia, microdontia, profound deafness, Michel aplasia) with homozygous FGF3 mutations including p.R104X (PMID:17236138)
Bilateral complete labyrinthine aplasia with microtia/microdontia in p.R104X homozygotes, PKDF295 (PMID:21306635)
LAMM phenotype (labyrinthine aplasia, microtia, microdontia) in Patient B with p.R104X compound heterozygosity (PMID:21480479)
ClinVar condition labels for variation 13838: 'Deafness with labyrinthine aplasia, microtia, and microdontia' (clinvar)
PS4Not met: affected families and absence from 162 matched controls are qualitative enrichment, not a case-control study with statistical testing.
Details ▾Hide ▴
Reasoning
No formal case-control study with statistically significant enrichment exists for c.310C>T. The variant is observed in multiple unrelated affected families (homozygous in all three affected members of Turkish family B, PMID:17236138; homozygous in Pakistani family PKDF295, PMID:21306635; compound heterozygous in Patient B, PMID:21480479) and was reported absent in 162 ethnically matched Pakistani normal-hearing control individuals (PMID:21306635), which is qualitative case enrichment but lacks a case-control design with odds-ratio/significance testing. Additionally, the 50-unrelated-control (100 chromosomes) screen in PMID:21480479 covered only the two newly reported missense variants (p.Y106C, p.Y49C), NOT p.R104X. Under generic ACMG/AMP 2015, PS4 (strong) is not met; the observed data could support a lower-strength application only under a VCEP rule, which is absent here.
Rule
Generic ACMG/AMP 2015 (PMID:25741868) PS4: The prevalence of the variant in affected individuals is significantly increased compared with the prevalence in controls (strong).
Evidence considered
Homozygous c.310C>T (p.R104X) in all 3 affected members of Turkish family B (PMID:17236138)
c.310C>T (p.R104X) in family PKDF295; mutations not found in 162 Pakistani normal-hearing controls (PMID:21306635)
p.R104X in compound heterozygosity in Patient B with LAMM (PMID:21480479)
No formal case-control study (affected vs. matched controls) with statistical significance testing for c.310C>T
Control screening specifically for p.R104X is limited to the qualitative absence statement in PMID:21306635 (162 Pakistani controls); no European-background control data for this exact variant
No VCEP rule defining a lower-strength PS4 application for FGF3
PP3Not met: SpliceAI max delta 0.04 is far below the 0.20 splice-altering threshold, and missense predictors do not apply to a stop-gained variant.
Details ▾Hide ▴
Reasoning
No ClinGen CSPEC/VCEP and no gene-specific PP3/BP4 framework exists for FGF3 (framework_mode=generic_acmg; vcep_materials.json empty), so the generic ACMG/AMP 2015 definitions apply (PMID:25741868). Variant NM_005247.2:c.310C>T is a stop-gained (nonsense) change predicting NP_005238.1:p.(Arg104Ter). Splice-impact path (this group owns it for all variant classes): SpliceAI max delta score = 0.04 (DS_AG 0.00, DS_AL 0.02, DS_DG 0.01, DS_DL 0.04), far below the 0.20 threshold conventionally used for splice-altering predictions, so the splice path does not support PP3. Missense in-silico path: a local BayesDel noAF lookup returned score 0.66, and REVEL v1.3 returned no record (score null), but BayesDel/REVEL are missense-oriented predictors; ACMG/AMP 2015 defines the two tool categories as 'those that predict whether a missense change is damaging... and those that predict if there is an effect on splicing', and neither category validates applying missense predictors to a stop-gained variant. The deleterious mechanism of this variant is its protein-truncating/null effect, which is adjudicated under PVS1 (null variant), not under PP3. No computational line of evidence therefore supports PP3; PP3 is not met.
Rule
ACMG/AMP 2015 (PMID:25741868) PP3: multiple lines of computational evidence support a deleterious effect on the gene or gene product (missense in-silico and/or splice predictors); not met for this stop-gained variant
BA1Not met: the highest population frequency is 0.033% (gnomAD v4.1 Middle Eastern), roughly 30-fold below the >1% BA1 threshold.
Details ▾Hide ▴
Reasoning
BA1 requires an allele frequency in population databases incompatible with pathogenicity. No FGF3 VCEP/CSPEC exists for this case (vcep_materials.json: 'no cspec url available', gene_vcep_dir=null, framework_mode=generic_acmg), so the generic non-VCEP gnomAD cutoff applies: BA1 >1%. The variant (FGF3 c.310C>T, p.Arg104Ter, autosomal recessive LAMM syndrome) is far below this threshold in every dataset: gnomAD v2.1 total AF = 0.00239% (6/251,480 alleles, 0 homozygotes), gnomAD v4.1 total AF = 0.00149% (24/1,613,542 alleles, 0 homozygotes), and gnomAD-Canada v1.0 AF = 0.0109% (2/18,406 alleles, 0 homozygotes). The highest continental/subpopulation frequency anywhere is 0.033% (Middle Eastern, 2/6,062 alleles, gnomAD v4.1), still ~30-fold below the 1% BA1 threshold. BA1 is therefore not met.
Rule
Generic ACMG/AMP 2015 BA1 (Richards et al., PMID:25741868; source_registry key generic_acmg_combination_rules): allele frequency >1% in population databases is stand-alone benign evidence (non-VCEP gnomAD cutoff). Not met: maximum observed allele frequency is 0.033% (MID, gnomAD v4.1), far below 1%.
Evidence considered
gnomad_v2: total AF 2.38588e-05 (0.00239%, 6/251,480 alleles), 0 homozygotes, grpmax FAF 1.121e-05; highest continental population SAS 6.53e-05 (0.0065%)
gnomad_v4: total AF 1.48741e-05 (0.00149%, 24/1,613,542 alleles), 0 homozygotes, grpmax FAF 5.818e-05; highest population MID 0.000329924 (0.033%, 2/6,062 alleles)
gnomad_canada: AF 0.00010866 (0.0109%, 2/18,406 alleles), 0 homozygotes
generic_acmg_combination_rules + PMID:25741868: BA1 per generic ACMG/AMP 2015, applied at the non-VCEP gnomAD cutoff of >1%
BS1Not met: the highest observed allele frequency (0.033%) is about 9-fold below the >0.3% BS1 cutoff.
Details ▾Hide ▴
Reasoning
BS1 requires an allele frequency greater than expected for the disorder (non-VCEP gnomAD cutoff: >0.3%). The variant is a founder/recurrent disease allele for LAMM syndrome (OMIM 610706), an extremely rare fully penetrant autosomal recessive disorder, so the expected allele frequency is far below 0.3%. Observed frequencies: gnomAD v2.1 total AF 0.00239% (max continental SAS 0.0065%), gnomAD v4.1 total AF 0.00149% (max population MID 0.033%, 2/6,062 alleles), gnomAD-Canada 0.0109%. The highest frequency in any dataset (0.033%) is ~9-fold below the 0.3% BS1 cutoff, so BS1 is not met.
Rule
Generic ACMG/AMP 2015 BS1 (Richards et al., PMID:25741868; source_registry key generic_acmg_combination_rules): '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)'. Applied at the non-VCEP gnomAD cutoff of >0.3%. Not met: maximum observed AF is 0.033% (MID, gnomAD v4.1), below the cutoff and consistent with a rare recessive disease allele.
Evidence considered
gnomad_v2: total AF 0.00239%; continental popmax 0.0065% (SAS); country-level max NFE_NWE 0.0071% (3/42,220) — all below 0.3%
gnomad_v4: total AF 0.00149%; population max 0.033% (MID, 2/6,062 alleles) — below 0.3%
gnomad_canada: AF 0.0109% (2/18,406) — below 0.3%
generic_acmg_combination_rules + PMID:25741868: BS1 per generic ACMG/AMP 2015, applied at the non-VCEP gnomAD cutoff of >0.3%
BS2Not met: zero homozygotes in gnomAD v2.1, v4.1, and gnomAD-Canada, and documented homozygotes are all affected with LAMM syndrome.
Details ▾Hide ▴
Reasoning
BS2 requires observation of the variant in a well-documented healthy adult in the genotype consistent with the inheritance pattern — for this autosomal recessive disorder, in the homozygous state. gnomAD v2.1, v4.1 and gnomAD-Canada all report 0 homozygotes (0/251,480; 0/1,613,542; 0/18,406 alleles). Moreover, the literature shows the opposite pattern: homozygotes for p.Arg104Ter are affected with fully penetrant LAMM syndrome (PMID:17236138: c.310C>T homozygous in all three affected members of Turkish family B; PMID:21306635: bilateral complete labyrinthine aplasia in all four deaf members of Pakistani family PKDF295), and the only unaffected carriers documented are heterozygotes (PMID:21306635 III-3; PMID:21480479 heterozygous parent of Patient B), which is the expected carrier state for a recessive disorder and does not satisfy BS2. BS2 is therefore not met.
Rule
Generic ACMG/AMP 2015 BS2 (Richards et al., PMID:25741868; source_registry key generic_acmg_combination_rules): '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)'. Not met: 0 homozygotes in gnomAD v2.1/v4.1/Canada and no healthy homozygous adults reported; all documented homozygotes are affected with LAMM syndrome.
Evidence considered
gnomad_v2: 0 homozygotes (6/251,480 alleles)
gnomad_v4: 0 homozygotes (24/1,613,542 alleles)
gnomad_canada: 0 homozygotes (2/18,406 alleles)
PMID:17236138: p.R104X (c.310C>T) homozygous in all three affected members of family B; parents heterozygous; unaffected relatives heterozygous or homozygous wild-type — homozygotes are affected, not healthy
PMID:21306635: homozygous p.R104X in all affected members of PKDF295 (bilateral CLA); heterozygous carrier III-3 unaffected — no healthy homozygotes
PMID:21480479: p.R104X in compound heterozygosity with p.Y49C in affected Patient B; heterozygous mother unaffected
generic_acmg_combination_rules + PMID:25741868: BS2 per generic ACMG/AMP 2015 requires a healthy adult in the homozygous state for recessive disorders; none observed
Still missing
No dedicated healthy-control cohort was explicitly screened for homozygous carriers of this specific variant beyond gnomAD (1.6M alleles, 0 homozygotes) and the small control screens in PMID:21306635 (162 Pakistani normal-hearing controls) and PMID:21480479 (50 European controls, covering the two new missense variants); gnomAD remains the primary population source for BS2.
BS4Not met: no affected individual lacks the variant; unaffected relatives are heterozygous carriers or wild-type, as expected for recessive inheritance.
Details ▾Hide ▴
Reasoning
BS4 (Richards et al. 2015) requires non-segregation of the variant with disease in affected family members. No non-segregation is documented for c.310C>T: no affected individual lacks the variant (all affected are homozygous or compound-heterozygous carriers), and unaffected relatives are either heterozygous carriers or homozygous wild-type, which is exactly the pattern expected for a fully penetrant autosomal recessive disorder (PMID:17236138, PMID:21306635, PMID:21480479). Heterozygous carriers being unaffected (e.g., PKDF295 III-3; Sensi mother) is consistent with recessive carrier status, not evidence against pathogenicity. The heterozygous mother of Patient B in PMID:21480479 had surgically corrected unilateral microtia, which the authors discuss as possibly related to heterozygosity but she lacks hearing loss and dental findings; this does not constitute non-segregation. Population data (gnomAD v2.1 AF=2.39e-05, 0 homozygotes) likewise show no healthy homozygotes contradicting the recessive model, though BS1/BA1 are handled by the population-frequency group.
Rule
generic_acmg_2015 BS4 (non-segregation with disease) - not met: no evidence of non-segregation in any reported family
Evidence considered
Segregation data fully consistent with AR inheritance in all 3 families; unaffected heterozygous carriers present but expected for AR carrier state; no affected individual without the variant
BP2Not met: the documented in-trans configurations (compound heterozygosity, homozygosity) are evidence of pathogenicity, the opposite of this criterion.
Details ▾Hide ▴
Reasoning
BP2 has two clauses: (1) observed in trans with a pathogenic variant for a fully dominant gene/disorder, and (2) observed in cis with a pathogenic variant in any inheritance pattern. Neither applies. FGF3-related LAMM syndrome is autosomal recessive (not fully dominant), so the in-trans clause cannot be invoked; in fact, the variant's documented in-trans configuration (compound heterozygosity with p.Y49C, PMID:21480479; homozygous biallelic state, PMID:17236138) is evidence in the pathogenic direction (PM3), the opposite of BP2. No report places c.310C>T in cis with a pathogenic variant: all parental-genotype data show the two mutations segregating on opposite chromosomes (maternal p.R104X / paternal p.Y49C in PMID:21480479) or homozygosity (PMID:17236138). BP2 is therefore not met.
Rule
BP2: Observed in trans with a pathogenic variant for a fully dominant gene/disorder; or observed in cis with a pathogenic variant in any inheritance pattern (Supporting, benign). ACMG/AMP 2015: finding the second variant in cis is supporting, though not definitive, evidence for a benign role (BP2); in dominant disorders, detection of a variant in trans with a pathogenic variant can be supporting evidence for a benign impact (BP2) (PMID:25741868).
Evidence considered
ACMG/AMP 2015 BP2 rule text: in-trans observation is benign-supporting only in fully dominant disorders; in-cis observation is benign-supporting in any inheritance pattern (PMID:25741868)
No cis-configured observations: p.R104X is in trans with p.Y49C (parental phase confirmed) in PMID:21480479 and homozygous (in trans with itself) in PMID:17236138
BP4Not met: a single negative SpliceAI prediction (max delta 0.04) cannot offset the unambiguous protein-truncating consequence of a stop-gained variant.
Details ▾Hide ▴
Reasoning
BP4 (ACMG/AMP 2015, PMID:25741868) requires multiple lines of computational evidence suggesting no impact on the gene or gene product. This variant is a stop-gained null change (NP_005238.1:p.(Arg104Ter)) whose protein-truncating consequence is unambiguous at the sequence level; computational evidence of 'no impact' cannot be reconciled with the established loss-of-function mechanism, and ACMG/AMP 2015 cautions that in-silico predictions should not be used as the sole basis for an assertion. The splice path was evaluated: SpliceAI max delta 0.04 is a single negative splice prediction and does not constitute 'multiple lines of computational evidence suggesting no impact'; moreover, absence of a predicted splice effect does not make a truncating variant benign. Missense benign predictors are not applicable to a nonsense variant. Applying BP4 here would conflict with the null mechanism and would inappropriately offset pathogenic criteria; BP4 is not met.
Rule
ACMG/AMP 2015 (PMID:25741868) BP4: multiple lines of computational evidence suggest no impact on the gene or gene product; not met/inapplicable for a stop-gained null variant
Evidence considered
SpliceAI max delta score 0.04 - single negative splice prediction; insufficient for BP4 and inapplicable given the truncating mechanism
Variant consequence class: stop_gained (p.Arg104Ter) - null variant; BP4 'no impact' premise contradicted by the variant class itself
ACMG/AMP 2015: in-silico predictions are a single piece of evidence and not recommended as the sole basis for a clinical assertion
PS3Not assessed: no well-established functional study of this variant was found in the reviewed literature.
Details ▾Hide ▴
Reasoning
No ClinGen CSPEC/VCEP and no gene-specific framework exists for FGF3 (framework_mode=generic_acmg; vcep_materials.json final_classification_framework found=false, generic_acmg_fallback), so the generic ACMG/AMP 2015 definition applies (PMID:25741868): PS3 = well-established functional studies show a deleterious effect. All full-text publications in the case evidence set (Tekin 2007 PMID:17236138; Riazuddin 2011 PMID:21306635; Sensi 2011 PMID:21480479) were read for functional assay content. None contains a validated functional assay for FGF3 c.310C>T (p.Arg104Ter): no transfection/expression, secretion, receptor-binding, signaling, proliferation, or mRNA/minigene experiments, and no assay controls, were reported. Tekin 2007 documents homozygous cosegregation of p.R104X in three affected members of family B and states only that the mutation 'presumably truncates more than one-half of the protein, apparently leaving behind a nonfunctional product' - a presumptive in-silico statement, not an experimental result. Riazuddin 2011 contains structural modeling of other FGF3 mutations (p.R95W, p.G66C, p.S156P) and a predicted-truncation statement that refers to p.R132GfsX26, not p.R104X; no wet-lab assays. Sensi 2011 is a clinical report listing p.R104X among known truncating mutations. SpliceAI (max delta 0.04) is an in-silico splice predictor, not a functional study. Because no well-established functional study demonstrating a deleterious effect is available, the evidence is insufficient to apply PS3: PS3 = not_assessed.
Rule
ACMG/AMP 2015 (PMID:25741868) PS3: well-established functional studies show a deleterious effect. Applied via generic fallback - no ClinGen CSPEC/VCEP or gene-specific framework for FGF3 (vcep_materials.json found=false, generic_acmg_fallback)
Evidence considered
Tekin 2007 (PMID:17236138): homozygous c.310C>T (p.R104X) in all three affected members of Turkish family B with complete cosegregation; 'This mutation presumably truncates more than one-half of the protein, apparently leaving behind a nonfunctional product' is presumptive wording, not assay data; no functional experiments in the paper
Riazuddin 2011 (PMID:21306635): family PKDF295 segregates p.R104X (c.310C>T) in four deaf members with bilateral complete labyrinthine aplasia; no functional assays; structural modeling addresses p.R95W/p.G66C/p.S156P, and the 'truncation of one third of the FGF3 protein' statement refers to p.R132GfsX26, not p.R104X
Sensi 2011 (PMID:21480479): p.R104X listed among three known truncating FGF3 mutations and found as the recurrent allele in European compound heterozygotes; clinical report, no functional studies
SpliceAI max delta 0.04 (in-silico splice prediction) - not a functional study and does not speak to a deleterious protein effect; not used to meet PS3
No validated assay with controls demonstrating a deleterious effect of p.Arg104Ter on FGF3 function exists anywhere in the case evidence set; clinical observation alone cannot satisfy PS3
Still missing
No well-established functional study (in vitro or in vivo, with assay controls) of FGF3 p.Arg104Ter exists in the case evidence set; only clinical cosegregation (PMID:17236138, PMID:21306635, PMID:21480479) and presumptive/in-silico statements are available
A validated FGF3 functional assay (e.g., secretion/processing or FGF-receptor signaling assay of the truncated protein, or mRNA NMD/minigene assessment) would be required before PS3 could be met or refuted
BS3Not assessed: no functional study demonstrating retained or normal activity of the truncated protein was available.
Details ▾Hide ▴
Reasoning
BS3 (generic ACMG/AMP 2015, PMID:25741868) requires well-established functional studies showing no damaging effect on protein function or splicing. No functional study of any kind was identified for FGF3 c.310C>T (p.Arg104Ter) in the case evidence set: none of the retrieved publications (PMID:17236138, PMID:21306635, PMID:21480479) performed any assay of FGF3 function, and nothing demonstrates retained or normal activity of the truncated protein. The only splice-related datum is SpliceAI (max delta score 0.04, no predicted splice impact), which is an in-silico prediction and explicitly not a well-established functional study; moreover, absence of a predicted splice effect says nothing about the protein-truncating consequence of this stop-gained allele and cannot establish benignity. The variant is a nonsense (null) change (NP_005238.1:p.(Arg104Ter)) whose expected consequence is loss of function - the opposite of the BS3 premise - and adjudication of that null mechanism belongs to PVS1/PM4 (other groups). Evidence is therefore insufficient to assess BS3: BS3 = not_assessed.
Rule
ACMG/AMP 2015 (PMID:25741868) BS3: well-established functional studies show no damaging effect on protein function or splicing. Applied via generic fallback - no ClinGen CSPEC/VCEP or gene-specific framework for FGF3 (vcep_materials.json found=false, generic_acmg_fallback)
Evidence considered
No functional study of any kind identified for FGF3 c.310C>T (p.Arg104Ter) across all retrieved sources; nothing shows retained/normal FGF3 function
SpliceAI max delta 0.04 (DS_AG 0.00, DS_AL 0.02, DS_DG 0.01, DS_DL 0.04) - single in-silico negative splice prediction; insufficient for BS3, which requires well-established functional studies (PMID:25741868)
Variant class: stop_gained (p.Arg104Ter) null allele; expected loss-of-function mechanism contradicts the BS3 'no damaging effect' premise and is adjudicated under PVS1/PM4, not BS3
Still missing
No functional study demonstrating absence of a damaging effect on FGF3 protein function or splicing exists for this variant; there is no benign assay evidence at all
In-silico SpliceAI (max delta 0.04) was consulted for the splicing component of BS3 but cannot substitute for a well-established functional study (e.g., mRNA/cDNA analysis, minigene assay, or protein function assay with controls)
BP5Not assessed: insufficient data were available to determine whether an alternate molecular basis of disease existed in affected cases.
Details ▾Hide ▴
Reasoning
BP5 requires evidence that the variant was found in a case with an alternate molecular basis for disease. The case bundle contains no proband-level phenotype or molecular-workup data (no phenopacket, no alternate-diagnosis investigation), so the presence or absence of an alternate molecular basis cannot be evaluated. In the reported literature, all affected carriers of p.R104X have biallelic FGF3 mutations consistent with LAMM and no alternate molecular cause was identified, but this does not constitute evidence for BP5 and is not a formal assessment.
Rule
Generic ACMG/AMP 2015 (PMID:25741868) BP5: Variant found in a case with an alternate molecular basis for disease (supporting, benign).
Still missing
No proband-level data on whether an alternate molecular basis for the phenotype exists (e.g., alternate gene/variant identified, non-genetic etiology)
No case-level molecular workup records in the evidence bundle
Not applicable · 12· recorded as not applicableShow reasons ▾Hide reasons ▴PS1N/ANot applicable: the nonsense change p.(Arg104Ter) produces no altered amino acid to compare against previously established pathogenic missense variants.
Details ▾Hide ▴
Reasoning
PS1 (Richards et al. 2015, PMID:25741868) requires the same amino acid change as a previously established pathogenic variant, evaluated regardless of the underlying nucleotide change. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, no altered amino acid exists at this position to compare against any previously established pathogenic missense variant, so PS1's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: nonsense (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
PS2N/ANot applicable: LAMM syndrome is autosomal recessive, so a single de novo heterozygous allele cannot fulfill the disease model, and none is documented.
Details ▾Hide ▴
Reasoning
PS2 (Richards et al. 2015) requires a confirmed de novo occurrence (paternity and maternity confirmed) in a patient with the disease and no family history. FGF3-related LAMM syndrome (OMIM 610706) is an autosomal recessive disorder requiring biallelic loss-of-function alleles, so a single de novo heterozygous allele cannot fulfill the disease model and PS2 does not apply to this variant class/disease context. Moreover, no de novo occurrence of NM_005247.2:c.310C>T (p.Arg104Ter) is documented in any reviewed source; every reported transmission is inherited from heterozygous carrier parents: both parents of family B are heterozygous carriers (PMID:17236138), the maternal allele in compound-heterozygous Patient B (PMID:21480479), and a heterozygous carrier parent III-3 tested in family PKDF295 (PMID:21306635). No VCEP/CSPEC specification exists for FGF3 (vcep_materials.json: no cspec url, framework_mode=generic_acmg, precedence_rank 3), so generic ACMG/AMP 2015 rules apply.
Rule
generic_acmg_2015 PS2 (de novo, paternity/maternity confirmed) - not applicable: autosomal recessive disorder; biallelic requirement means a single de novo allele cannot cause LAMM
Evidence considered
All reported p.R104X alleles are inherited from heterozygous carrier parents; no de novo occurrence in any family (Tekin 2007 family B parents heterozygous; Sensi 2011 Patient B maternal p.R104X; Riazuddin 2011 PKDF295 carrier III-3 heterozygous and clinically normal)
PM1N/ANot applicable: as a nonsense variant, no altered residue exists to evaluate for mutational hotspot or critical-domain membership.
Details ▾Hide ▴
Reasoning
PM1 (Richards et al. 2015, PMID:25741868) applies to a missense variant located in a mutational hot spot and/or a critical, well-established functional domain without benign variation. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, no altered residue exists to evaluate for mutational hot-spot or critical-domain membership, so PM1's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: nonsense (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
PM4N/ANot applicable: p.Arg104Ter truncates the protein rather than altering its length, so the protein-length-change criterion does not apply.
Details ▾Hide ▴
Reasoning
PM4 (Richards et al. 2015, PMID:25741868) applies to protein length changes: in-frame insertions/deletions in a non-repetitive region, or stop-loss variants that extend the protein past the reference stop codon. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, no change in protein length occurs, so there is nothing for this criterion to evaluate, so PM4's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
PM5N/ANot applicable: no missense change exists at Arg104 to compare against a different pathogenic missense change at the same residue.
Details ▾Hide ▴
Reasoning
PM5 (Richards et al. 2015, PMID:25741868) requires a novel missense change at an amino acid residue where a different missense change has previously been determined to be pathogenic. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, no missense change exists at this residue to compare against a different pathogenic missense change at the same position, so PM5's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: nonsense (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
PM6N/ANot applicable: LAMM syndrome is autosomal recessive and no de novo occurrence of this variant is documented.
Details ▾Hide ▴
Reasoning
PM6 (Richards et al. 2015) covers assumed de novo occurrence without confirmation of paternity/maternity. For the same reason as PS2, PM6 does not apply: FGF3-LAMM is autosomal recessive and disease requires biallelic FGF3 mutations, so an unconfirmed de novo heterozygous allele is not a disease-causing configuration in this disorder. No assumed-de-novo case of c.310C>T is reported in any source; all documented transmissions are from heterozygous carrier parents (PMID:17236138 family B; PMID:21480479 maternal allele in Patient B; PMID:21306635 PKDF295). Adjudicated under generic ACMG/AMP 2015 (no FGF3 VCEP/CSPEC available).
Rule
generic_acmg_2015 PM6 (assumed de novo, unconfirmed) - not applicable: autosomal recessive disorder; inherited carrier transmission documented
Evidence considered
No assumed de novo occurrence reported; variant consistently inherited from heterozygous carrier parents across all families with parental testing
PP2N/ANot applicable: the gene's missense-constraint properties are irrelevant because no missense change is present to evaluate.
Details ▾Hide ▴
Reasoning
PP2 (Richards et al. 2015, PMID:25741868) applies to a missense variant in a gene with a low rate of benign missense variation, where missense variants are a common mechanism of disease. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, the gene's missense-constraint properties are irrelevant because no missense change is present to evaluate, so PP2's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: nonsense (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
PP5N/ANot applicable: ClinVar has no expert-panel submissions for this variant, and laboratory classifications alone cannot trigger PP5.
Details ▾Hide ▴
Reasoning
PP5 is applied only for an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic. ClinVar variation 13838 has zero expert-panel submissions (expert_panel_submissions: 0; review status 'criteria provided, multiple submitters, no conflicts', 2 stars). All seven submissions are ordinary clinical laboratories (PreventionGenetics, GeneDx, 3billion, Victorian Clinical Genetics Services), OMIM literature, GeneReviews, and a research submission (Baylor-Hopkins). Aggregate laboratory classifications (2 Pathogenic + 2 Likely pathogenic) must never trigger PP5. Therefore PP5 is not applicable.
Rule
Generic ACMG/AMP 2015 (PMID:25741868) PP5: Reputable source recently reports variant as pathogenic, but the evidence is not available to the laboratory to perform an independent evaluation (supporting). Per governing instructions, applied only for an exact-variant ClinVar expert-panel Pathogenic/Likely pathogenic classification.
Evidence considered
ClinVar submission audit: expert_panel_submissions = 0; all submitters are clinical labs/literature/research (clinvar)
Still missing
No ClinGen/other expert-panel classification exists for this exact variant (would be required to apply PP5)
BP1N/ANot applicable: the criterion concerns missense variants, and this is a nonsense change (p.Arg104Ter).
Details ▾Hide ▴
Reasoning
BP1 (Richards et al. 2015, PMID:25741868) applies to a missense variant in a gene for which primarily truncating variants are known to cause disease. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, the gene's truncating-variant mechanism is irrelevant because no missense change is present to evaluate, so BP1's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: nonsense (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
BP3N/ANot applicable: the criterion concerns in-frame indels in repetitive regions; this is a nonsense substitution.
Details ▾Hide ▴
Reasoning
BP3 (Richards et al. 2015, PMID:25741868) applies to in-frame insertions/deletions located in a repetitive region without a known function. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, the variant does not alter protein length within a repeat region, so there is nothing for this criterion to evaluate, so BP3's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
BP6N/ANot applicable: no expert-panel benign classification exists for this variant; the ClinVar record is Pathogenic or Likely Pathogenic.
Details ▾Hide ▴
Reasoning
BP6 is applied only for an exact-variant ClinVar expert-panel classification of Benign/Likely benign. No expert-panel submissions exist for ClinVar variation 13838 (expert_panel_submissions: 0), and the aggregate record is Pathogenic/Likely pathogenic (4 clinical laboratory submissions), not benign. Ordinary laboratory submissions and aggregate ClinVar labels must never trigger BP6. Therefore BP6 is not applicable.
Rule
Generic ACMG/AMP 2015 (PMID:25741868) BP6: Reputable source recently reports variant as benign, but the evidence is not available to the laboratory to perform an independent evaluation (supporting, benign). Per governing instructions, applied only for an exact-variant ClinVar expert-panel Benign/Likely benign classification.
BP7N/ANot applicable: the criterion applies only to synonymous variants; this nonsense change alters the protein sequence.
Details ▾Hide ▴
Reasoning
BP7 (Richards et al. 2015, PMID:25741868) is defined for a synonymous (silent) variant for which splicing prediction algorithms predict no impact to the splice consensus sequence and the nucleotide is not highly conserved. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, the encoded protein sequence is altered, so the criterion's silent-variant premise does not hold regardless of any splicing prediction, so BP7's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_synonymous
Evidence considered
Deterministic consequence classification: nonsense (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
Identifiers
Gene
FGF3
Transcript
NM_005247.2
Coding (HGVS c.)
NM_005247.2:c.310C>T
Protein (HGVS p.)
NP_005238.1:p.(Arg104Ter)
GRCh38
chr11:69816334 G>A
GRCh37
chr11:69631102 G>A
ClinVar
13838
COSMIC
COSV100490202
Gene diagram
Every interpreted variant in FGF3, mapped onto the exon structure of NM_005247.2. Select a row to find it on the plot.
FGF3NM_005247.2
Fetching transcript structure from UCSC…
Exons
—
Transcript span
—
Strand
—
Variants mapped
—
Source
UCSC ncbiRefSeqCurated
All variants in FGF3—Select a row to find it on the plot; use the link to open its page.
AF 1.5e-05 · 24 / 1,613,542 alleles · 0 hom · FAF 5.8e-05
Middle Eastern2 / 6,062
0.00033
African/African American5 / 75,014
6.7e-05
South Asian3 / 91,062
3.3e-05
European (non-Finnish)14 / 1,179,496
1.2e-05
0 alternate alleles recorded in 6 other groups: Remaining individuals, Admixed American, European (Finnish), Amish, East Asian, Ashkenazi Jewish.
gnomAD v2.1
AF 2.4e-05 · 6 / 251,480 alleles · 0 hom · FAF 1.1e-05
South Asian2 / 30,616
6.5e-05
European (non-Finnish)4 / 113,766
3.5e-05
0 alternate alleles recorded in 6 other groups: African/African American, Admixed American, Ashkenazi Jewish, East Asian, European (Finnish), Remaining individuals.
gnomAD-Canada
AF not recorded · ? / ? alleles
Statements
gnomAD v4.1
This variant is present in gnomAD v4.1 (AF= 1.48741e-05; MAF= 0.00149%, 24/1613542 alleles, homozygotes = 0) and has highest observed frequency in the Middle Eastern population (AF= 0.000329924; MAF= 0.03299%, 2/6062 alleles, homozygotes = 0); grpmax FAF= 5.818e-05.
gnomAD v2.1
This variant is present in gnomAD v2.1 (AF= 2.38588e-05; MAF= 0.00239%, 6/251480 alleles, homozygotes = 0) and has highest observed frequency in the South Asian population (AF= 6.53253e-05; MAF= 0.00653%, 2/30616 alleles, homozygotes = 0); grpmax FAF= 1.121e-05.
gnomAD-Canada
This variant is present in gnomAD-Canada v1.0 (AF= 0.00010866021949364337, 2/18406 alleles, homozygotes = 0).
PP3Not met: SpliceAI max delta 0.04 is far below the 0.20 splice-altering threshold, and missense predictors do not apply to a stop-gained variant.
Details ▾Hide ▴
Reasoning
No ClinGen CSPEC/VCEP and no gene-specific PP3/BP4 framework exists for FGF3 (framework_mode=generic_acmg; vcep_materials.json empty), so the generic ACMG/AMP 2015 definitions apply (PMID:25741868). Variant NM_005247.2:c.310C>T is a stop-gained (nonsense) change predicting NP_005238.1:p.(Arg104Ter). Splice-impact path (this group owns it for all variant classes): SpliceAI max delta score = 0.04 (DS_AG 0.00, DS_AL 0.02, DS_DG 0.01, DS_DL 0.04), far below the 0.20 threshold conventionally used for splice-altering predictions, so the splice path does not support PP3. Missense in-silico path: a local BayesDel noAF lookup returned score 0.66, and REVEL v1.3 returned no record (score null), but BayesDel/REVEL are missense-oriented predictors; ACMG/AMP 2015 defines the two tool categories as 'those that predict whether a missense change is damaging... and those that predict if there is an effect on splicing', and neither category validates applying missense predictors to a stop-gained variant. The deleterious mechanism of this variant is its protein-truncating/null effect, which is adjudicated under PVS1 (null variant), not under PP3. No computational line of evidence therefore supports PP3; PP3 is not met.
Rule
ACMG/AMP 2015 (PMID:25741868) PP3: multiple lines of computational evidence support a deleterious effect on the gene or gene product (missense in-silico and/or splice predictors); not met for this stop-gained variant
BP4Not met: a single negative SpliceAI prediction (max delta 0.04) cannot offset the unambiguous protein-truncating consequence of a stop-gained variant.
Details ▾Hide ▴
Reasoning
BP4 (ACMG/AMP 2015, PMID:25741868) requires multiple lines of computational evidence suggesting no impact on the gene or gene product. This variant is a stop-gained null change (NP_005238.1:p.(Arg104Ter)) whose protein-truncating consequence is unambiguous at the sequence level; computational evidence of 'no impact' cannot be reconciled with the established loss-of-function mechanism, and ACMG/AMP 2015 cautions that in-silico predictions should not be used as the sole basis for an assertion. The splice path was evaluated: SpliceAI max delta 0.04 is a single negative splice prediction and does not constitute 'multiple lines of computational evidence suggesting no impact'; moreover, absence of a predicted splice effect does not make a truncating variant benign. Missense benign predictors are not applicable to a nonsense variant. Applying BP4 here would conflict with the null mechanism and would inappropriately offset pathogenic criteria; BP4 is not met.
Rule
ACMG/AMP 2015 (PMID:25741868) BP4: multiple lines of computational evidence suggest no impact on the gene or gene product; not met/inapplicable for a stop-gained null variant
Evidence considered
SpliceAI max delta score 0.04 - single negative splice prediction; insufficient for BP4 and inapplicable given the truncating mechanism
Variant consequence class: stop_gained (p.Arg104Ter) - null variant; BP4 'no impact' premise contradicted by the variant class itself
ACMG/AMP 2015: in-silico predictions are a single piece of evidence and not recommended as the sole basis for a clinical assertion
BP7N/ANot applicable: the criterion applies only to synonymous variants; this nonsense change alters the protein sequence.
Details ▾Hide ▴
Reasoning
BP7 (Richards et al. 2015, PMID:25741868) is defined for a synonymous (silent) variant for which splicing prediction algorithms predict no impact to the splice consensus sequence and the nucleotide is not highly conserved. NM_005247.2:c.310C>T in FGF3 is a nonsense substitution introducing a premature stop codon predicted to produce NP_005238.1:p.(Arg104Ter). As a result, the encoded protein sequence is altered, so the criterion's silent-variant premise does not hold regardless of any splicing prediction, so BP7's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_synonymous
Evidence considered
Deterministic consequence classification: nonsense (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
Statement
In silico
SpliceAI predicts no significant splice impact for this variant (max delta score = 0.04). BayesDel score = 0.66.
5 paper records listed. 4 cited in the classification.
Cited in the classification
Homozygous mutations in fibroblast growth factor 3 are associated with a new form of syndromic deafness characterized by inner ear agenesis, microtia, and microdontia.
Tekin M et al. · 2007 Feb
Original report of FGF3 mutations in LAMM syndrome. In Turkish family B, the nonsense mutation c.310C>T (p.R104X) is homozygous in all three affected members; all available parents are heterozygous carriers and unaffected relatives are heterozygous or homozygous wild-type (pedigree fig. 1). Segregation is fully consistent with completely penetrant autosomal recessive inheritance, confirming the variant is inherited (not de novo). Family A (p.S156P) and family C (p.V206SfsX117) carry different mutations and are not directly relevant to this variant. Note: the extracted OCR text renders the transition as 'c.310CrT' where 'r' is the OCR artifact for '>'.
We found a nonsense homozygous mutation, c.310CrT (p.R104X), in the second exon of FGF3 in all three of the affected members of family B (fig. 7). All available parents were heterozygous, and unaffected relatives were heterozygous or homozygous for the wild-type allele in this family (fig. 1).
Location Results section, Family B paragraph; Figure 1 (pedigrees of three Turkish families); Figure 7 (mutation electropherograms)
Context Three unrelated Turkish families with autosomal recessive syndromic deafness (Michel aplasia, microtia, microdontia); homozygosity mapping with 10K SNP arrays followed by Sanger sequencing of FGF3 exons in all available family members; mutations screened in 300 healthy Turkish controls
Variable expressivity of FGF3 mutations associated with deafness and LAMM syndrome.
Riazuddin S et al. · 2011 Feb 9
Prospective molecular genetic study of Pakistani families segregating deafness linked to the DFNB63 locus on 11q13. Family PKDF295 segregates homozygous c.310C>T (p.R104X) in four affected members (IV:1, IV:2, IV:7, IV:8) with complete labyrinthine aplasia; the mutation was previously reported and was not found in 162 ethnically matched normal-hearing controls. A heterozygous carrier (III-3, PKDF295) showed no abnormal temporal bone findings on imaging, consistent with recessive carrier status. This adds a second family (4 affected homozygotes) with cosegregation of the variant with disease.
One mutation (p.R132GfsX26 (c.394delC) from family PKDF887) was novel and the other two mutations (p.R95W (c.283 C > T) from PKDF817 and p.R104X (c.310C > T) from PKDF295) were previously reported (Table 1 ; Figure 1 )
Location Results ('FGF3 mutations in DFNB63-linked families'); Table 1 (family PKDF295 rows IV:1, IV:2, IV:7, IV:8); temporal bone imaging paragraph on heterozygous carriers
Context Five Pakistani families linked to DFNB63 locus; homozygosity mapping and FGF3 sequencing; 162 ethnically matched Pakistani normal-hearing controls; temporal bone CT/3D-FIESTA MRI in 10 affected and 3 heterozygous individuals
LAMM syndrome with middle ear dysplasia associated with compound heterozygosity for FGF3 mutations.
Sensi A et al. · 2011 May
First report of FGF3 compound heterozygotes in two European non-consanguineous families. Patient B (Italian) carries the paternal missense p.Y49C in compound heterozygosity with the maternal nonsense p.R104X (c.310C>T), confirming maternal transmission of the variant in a third independent family; both parents are normal-hearing, unaffected carriers. The heterozygous mother had surgically corrected unilateral microtia without hearing loss or dental findings, discussed by the authors as possibly related to heterozygosity but not constituting non-segregation. Family A carries different mutations (p.Y106C/p.W153VfsX51) and is not relevant to this variant.
In Patient B the paternal missense mutation p.Y49C was found in compound heterozygosity with the maternal nonsense mutation (p.R104X).
Location Results section, Patient B (Italian family); Discussion paragraph on the heterozygous mother's unilateral microtia
Context Two European families from non-consanguineous marriages with LAMM syndrome; FGF3 Sanger sequencing of propositi and parents (NM_005247.2); 50 unrelated European control subjects (100 chromosomes) tested for the novel missense variants
Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.
Richards S et al. · 2015 May
Richards et al. 2015 (ACMG/AMP joint consensus standards) is the governing framework for this case because no ClinGen CSPEC/VCEP or gene-specific specification exists for FGF3. It defines PS3 ('well-established functional studies show a deleterious effect') and BS3 ('well-established functional studies show no damaging effect on protein function or splicing'), and its 'Functional studies' paragraph sets the assay-quality standard for applying them: assays must be validated, reproducible, and should reflect the full biological function of the protein in an environment close to the biological one. This is the basis for requiring actual assay evidence (not in-silico prediction or clinical observation) before either criterion can be met. The paper does not discuss this variant (framework/guideline document only).
→ PM2 · moderate→ PM3 · moderate→ PP4 · supportingDoes not name this variant
Functional studies can be a powerful tool in support of pathogenicity; however, not all functional studies are effective in predicting an impact on a gene or protein function. For example, certain enzymatic assays offer well-established approaches to assess the impact of a missense variant on enzymatic function in a metabolic pathway... To assess the validity of a functional assay, one must consider how closely the functional assay reflects the biological environment.
Location Full text (publications/25741868.txt), 'Functional studies' paragraph under the PS3/BS3 criterion heading (criterion definitions in Table 5: PS3 and BS3)
Context ACMG/AMP 2015 joint consensus recommendation for sequence variant interpretation (guideline document; PS3/BS3 criteria and functional-assay validity discussion)
Other recorded publications
Congenital Deafness with Labyrinthine Aplasia, Microtia, and Microdontia.
The evidence statements recorded by the interpretation pipeline. They have not been reinterpreted by the page.
gnomAD v4.1
This variant is present in gnomAD v4.1 (AF= 1.48741e-05; MAF= 0.00149%, 24/1613542 alleles, homozygotes = 0) and has highest observed frequency in the Middle Eastern population (AF= 0.000329924; MAF= 0.03299%, 2/6062 alleles, homozygotes = 0); grpmax FAF= 5.818e-05.
gnomAD v2.1
This variant is present in gnomAD v2.1 (AF= 2.38588e-05; MAF= 0.00239%, 6/251480 alleles, homozygotes = 0) and has highest observed frequency in the South Asian population (AF= 6.53253e-05; MAF= 0.00653%, 2/30616 alleles, homozygotes = 0); grpmax FAF= 1.121e-05.
gnomAD-Canada
This variant is present in gnomAD-Canada v1.0 (AF= 0.00010866021949364337, 2/18406 alleles, homozygotes = 0).
ClinVar
This variant has been reported in ClinVar as Pathogenic (2 clinical laboratories) and as Likely pathogenic (2 clinical laboratories). (ClinVarID = 13838)
In silico
SpliceAI predicts no significant splice impact for this variant (max delta score = 0.04). BayesDel score = 0.66.
Functional
OncoKB identified curated literature and non-variant-specific oncogenicity context for review; listed oncogenicity label: Unknown Oncogenic Effect.
Somatic
This variant does not lie in a statistically significant hotspot. This variant has previously been reported in somatic cancers (COSMIC; COSV100490202, n = 3 times).
Cancer Hotspots
This variant does not lie in a statistically significant hotspot.
This variant was interpreted by pipeline 7.1.0.
The current version is 11.0.07.
Re-running queues a fresh interpretation; the result replaces this page when complete.
Re-runs are not free. Blank draws on today's public interpretation allowance; with a code, it uses one of that code's uses.