GBA1 c.1093G>A, p.(Glu365Lys)

Reasoning

Per ACMG/AMP 2015 (PMID 25741868), BS1 is met when allele frequency is greater than expected for the disorder. This variant is far too common to be a disease-causing GBA1 allele: gnomAD v4.1 AF 1.2312% (19,873 alleles), gnomAD v2.1 AF 1.0731%, gnomAD-Canada AF 0.7980%, with highest subpopulation AF 4.4255% (Finnish). These frequencies exceed the pipeline's documented non-VCEP BS1>0.3% convention by 2.5-4x and vastly exceed the expected frequency for a Mendelian disease allele of Gaucher disease (PMID 25741868: BS1 if 'allele frequency is greater than expected for disorder'; the 2015 guidance also applies BS1 when a subpopulation frequency in individuals not known to have the disorder exceeds disease expectation). PMID 12791040 directly demonstrates this for E326K: 'The similar frequencies of this mutation in Gaucher's disease patients (1.3%) and normal controls (0.9%) support this contention' - i.e., the allele is not enriched in affected individuals, supporting benignity in the Gaucher disease (AR) context. Caveat: for the Parkinson's disease phenotype this allele is a risk factor (PMID 26117366), so BS1's benign-supporting weight applies to the Gaucher disease context; the PD EP VCEP rules that would govern this weighting are not available in the bundle, hence human review.

Evidence considered

• gnomAD v4.1 overall AF 1.2312% > BS1>0.3% convention; Finnish AF 4.4255%; grpmax FAF 1.3208%
• gnomAD v2.1 overall AF 1.0731%; gnomAD-Canada AF 0.7980%
• PMID 12791040: E326K frequency 1.3% in Gaucher disease patients vs 0.9% in normal controls - allele frequency not greater in affected individuals
• BS1 definition per ACMG/AMP 2015 (Richards et al., PMID 25741868)

Still missing

• VCEP-specific BS1 threshold/rule for GBA1 in the PD framework not available (rule_payload empty); PD EP may weight population evidence differently for risk alleles such as E365K
• E365K is a documented PD risk allele (PMID 26117366); whether BS1 (supporting benign) should be retained given PD risk status needs human review

Sources

Reasoning

Per ACMG/AMP 2015 (PMID 25741868), BS2 is met when the variant is 'observed in a healthy adult individual for a recessive (homozygous)... disorder, with full penetrance expected at an early age'. This variant is observed homozygous in reference populations at high counts: 196 homozygotes in gnomAD v4.1 (67 Finnish, 120 non-Finnish European), 37 homozygotes in gnomAD v2.1, and 1 in gnomAD-Canada. For autosomal recessive Gaucher disease (a condition under which this variant is curated in ClinVar), biallelic state without disease indicates the allele does not cause Gaucher disease: PMID 12791040 provides in vivo family evidence that 'the recurrent change, E326K, is not a pathogenic mutation' (45-year-old healthy father heteroallelic for G202R/E326K; healthy sister L444P/E326K with ~45% GCase activity); PMID 16293621 notes E326K 'has been found in phenotypically normal individuals as the heteroallele' and that 'normalcy for periods of 1-4 decades can exist in the presence of an E326K allele'; PMID 15146461 found E326K 'was never found alone in a Gaucher disease-causing allele'. Caveats: (1) gnomAD is not a strictly healthy adult cohort; (2) for the late-onset, incompletely penetrant Parkinson's disease phenotype, the BS2 'full penetrance expected at an early age' requirement is not satisfied, so BS2 here reflects the Gaucher disease (AR) context; (3) VCEP rules for GBA1 (PD framework) are unavailable. Human review recommended before finalizing this strong benign call.

Evidence considered

• gnomAD v4.1: 196 homozygotes (67 Finnish, 120 NFE) of 19,873 alleles
• gnomAD v2.1: 37 homozygotes of 3,035 alleles; gnomAD-Canada: 1 homozygote
• PMID 12791040: E326K non-pathogenic in family segregation (healthy G202R/E326K father age 45; L444P/E326K sister with ~45% GCase activity)
• PMID 16293621: E326K in phenotypically normal heteroallele carriers; normalcy for 1-4 decades with an E326K allele
• PMID 15146461: E326K never found alone on a Gaucher disease-causing allele
• BS2 definition per ACMG/AMP 2015 (Richards et al., PMID 25741868)

Still missing

• gnomAD is a reference cohort, not a clinically phenotyped healthy cohort; unaffected-status of homozygotes is inferred
• For late-onset PD, BS2's 'full penetrance expected at an early age' requirement (PMID 25741868) is not met - BS2 met here is specific to the Gaucher disease (AR) context
• PD EP GBA1 VCEP rule payload unavailable - the spec may define different BS2 handling for GBA1 variants

Sources

Reasoning

BS4 (lack of segregation in affected members of a family) is met at strong strength under the generic ACMG/AMP 2015 framework. PMID:12791040 (Zhao et al., Clin Genet 2003; full text available) reports segregation analysis of the E326K allele (= p.Glu365Lys, c.1093G>A, legacy GBA1 numbering) in the family of a Gaucher disease type 3 patient: the affected proband II-4 (genotype G202R/L444P) and affected brother II-3 do not carry E326K, whereas the unaffected father I-1 (G202R/E326K, age 45) and unaffected sister II-5 (L444P/E326K, age 11) do. The paper states the mutations 'segregate independently' and concludes 'This supports E326K as a polymorphic allele that, at least in this family, is non-pathogenic', noting similar allele frequencies in Gaucher patients (1.3%) versus normal controls (0.9%). This is corroborated by the LabCorp ClinVar submission (SCV000697574): 'this variant did not co-segregate with disease in at-least one family with a clinical diagnosis of gaucher's disease'. Human review is requested because (i) the non-segregation evidence is against Gaucher disease while the governing VCEP context is Parkinson's disease risk (the variant is not causal for Gaucher disease, so non-segregation with GD is expected and is the cleanest form of BS4 evidence); (ii) the evidence relies on legacy numbering equivalence E326K = p.Glu365Lys; and (iii) the PD EP GBA1 v1.0 CSPEC rule payload was empty, so the generic BS4 rule was applied.

Rule

ACMG/AMP 2015 BS4: Lack of segregation in affected members of a family (PMID:25741868); strong benign criterion. Caveat: phenocopies for common phenotypes can mimic lack of segregation. No usable VCEP-specific rule text was retrievable; generic rule governs.

Evidence considered

• PMID:12791040: 'The E326K mutation was not present in [II-4] or [I-2], but was present in [I-1] and [II-5]. This indicates that the bearing of the mutations encoding E326K, G202R or L444P segregate independently.' (Fig. 1 legend); 'The presence of G202R and the absence of E326K in case 4 (II-4) and II-3, and the L444P/E326K genotype in II-5, confirm the trans orientation of these two alleles in I-1.' and 'This supports E326K as a polymorphic allele that, at least in this family, is non-pathogenic.' (Discussion); abstract: 'isolated E326K is non-pathogenic in this family.'
• ClinVar LabCorp SCV000697574 (Likely benign) assertion signal: 'this variant did not co-segregate with disease in at-least one family with a clinical diagnosis of gaucher's disease'.

Still missing

• No Parkinson's-disease-specific segregation data are available; BS4 rests on non-segregation with Gaucher disease in one family (PMID:12791040) plus a corroborating laboratory assertion (ClinVar SCV000697574).
• Confirm legacy-numbering equivalence E326K = p.Glu365Lys (NM_000157.4:c.1093G>A) for the cited paper, and re-evaluate under the PD EP GBA1 v1.0 BS4 rule once its text is retrievable.

Sources

Reasoning

BP2 is met at supporting strength through the cis clause ('observed in cis with a pathogenic variant in any inheritance pattern'). PMID:15146461 documents that c.1093G>A (p.E326K) occurs in cis on complex Gaucher disease alleles [c.1448T>C; c.1093G>A] ([p.L444P; p.E326K]) and [c.680A>G; c.1093G>A] ([p.N188S; p.E326K]) identified in Gaucher disease patients, and that E326K was never found alone on a Gaucher disease-causing allele — i.e., the variant under evaluation is observed in cis with the established pathogenic L444P (and N188S) change in the same allele, exactly the situation BP2 describes (the co-occurring pathogenic change, not E326K, is the disease-causing component). The first BP2 clause (in trans with a pathogenic variant for a fully penetrant dominant disorder) is not strictly satisfied because Gaucher disease is autosomal recessive; nevertheless PMID:12791040 shows E326K in trans with pathogenic G202R and L444P in phenotypically normal carriers, which is consistent with the benign interpretation and was directly consulted for this criterion.

Rule

Generic ACMG/AMP 2015 (PMID 25741868): BP2 - observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder; or observed in cis with a pathogenic variant in any inheritance pattern.

Evidence considered

• PMID:15146461: E326K observed in cis with pathogenic L444P and N188S on complex alleles [c.1448T>C; c.1093G>A] ([p.L444P; p.E326K]) and [c.680A>G; c.1093G>A] ([p.N188S; p.E326K]) found in Gaucher disease patients; E326K 'never found alone in a Gaucher disease-causing allele'.
• PMID:12791040: E326K in trans with pathogenic G202R and L444P in phenotypically normal carriers of a type 3 Gaucher family (healthy father and sister), directly consulted for BP2 and consistent with a non-pathogenic role.

Still missing

• The GBA1 CSPEC (Parkinson's Disease EP v1.0) supplied no retrievable per-criterion BP2 rule in this case's bundle; generic ACMG/AMP 2015 (PMID 25741868) BP2 was applied.
• Montfort et al. (PMID:15146461) report the [L444P;E326K] and [N188S;E326K] double-mutant alleles using standard complex-allele (in cis) notation; the underlying clinical phase-determination methods for those alleles are not detailed in the abstract.

Sources

Reasoning

BP5 (variant found in a case with an alternate molecular basis for disease) is met from family-level case data in PMID:12791040. In a Gaucher disease family, E326K (c.1093G>A, p.Glu365Lys) was detected in affected individuals whose disease is molecularly explained by other GBA alleles: the E326K change segregated in trans with G202R (case I-1) and L444P (case II-5), with segregation analysis showing G202R and L444P segregate independently of E326K, and the authors conclude isolated E326K is non-pathogenic ('This supports E326K as a polymorphic allele that, at least in this family, is non-pathogenic'). The affected cases therefore carry E326K alongside a documented alternate molecular basis (the trans GBA disease allele), which is the BP5 situation. Applied at supporting strength per generic ACMG. Caveat noted: this is Gaucher-disease-specific family data; no PD case with a documented alternate molecular basis was identified in the bundle.

Rule

Generic ACMG/AMP 2015 BP5 (Richards et al. 2015, PMID:25741868): variant found in a case with an alternate molecular basis for disease.

Evidence considered

• PMID:12791040 (Zhao et al., Hum Mutat 2003): Gaucher disease family - E326K in trans with G202R and L444P; affected members' disease attributable to the alternate GBA allele; isolated E326K non-pathogenic; 'similar frequencies of this mutation in Gaucher's disease patients (1.3%) and normal controls (0.9%)'

Still missing

• No PD-specific case with a documented alternate molecular basis for disease was identified; BP5 met on Gaucher disease family data (PMID:12791040) only
• PD VCEP GBA1 v1.0 rule text for BP5 not retrievable (rule_payload empty); assessed under generic ACMG/AMP 2015

Sources

Reasoning

PS1 requires the same amino acid change as a previously established pathogenic variant, regardless of nucleotide change. Here the amino acid change in question is p.Glu365Lys (legacy p.E326K; same cDNA change c.1093G>A). The literature uniformly shows this change is NOT established as pathogenic: Zhao et al. 2003 (PMID:12791040) states the recurrent E326K change 'is not a pathogenic mutation' and supports it as a polymorphic allele; Montfort et al. 2004 (PMID:15146461) shows c.1093G>A was 'never found alone in a Gaucher disease-causing allele' and calls it a 'modifier variant'; Liou et al. 2006 (PMID:16293621) reports E326K in phenotypically normal individuals. No alternative nucleotide change producing p.Glu365Lys (e.g., at c.1095) was found in any available source, and even if one existed, the prerequisite of an established-pathogenic comparator is not satisfied. PS1 therefore not met.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PS1 - Same amino acid change as a previously established pathogenic variant regardless of nucleotide change.

Evidence considered

• PMID:12791040 - 'In vivo evidence is provided to show that the recurrent change, E326K, is not a pathogenic mutation' (abstract); E326K segregates in trans with G202R/L444P and is non-pathogenic -> the p.Glu365Lys change is not an established pathogenic comparator, so PS1 cannot apply.
• PMID:15146461 - 'the c.1093G>A (p.E326K) change, which was never found alone in a Gaucher disease-causing allele...' - same amino acid change never disease-causing on its own.
• PMID:16293621 - E326K found in phenotypically normal individuals as heteroallele to L444P- or G202R-encoding alleles.
• PMID:25741868 - generic ACMG/AMP PS1 rule definition applied because the CSPEC PD VCEP payload was empty (no PS1 specification retrievable).

Still missing

• No dedicated search for alternative nucleotide changes at codon 365 that also produce p.Glu365Lys was performed; however, this is moot because the amino acid change itself is not established as pathogenic.

Sources

Reasoning

PS2 (de novo with both maternity and paternity confirmed, in a patient with the disease and no family history) requires documented de novo occurrence in the proband. No source in the case-evidence bundle reports de novo occurrence, proband/parental trio testing, or parental genotype data for this variant. On the contrary, the available evidence shows this is a common, transmitted polymorphism: gnomAD v4.1 AF 1.23% (19,873 alleles, 196 homozygotes), gnomAD v2.1 AF 1.07% (3,035 alleles, 37 homozygotes), and the only family-study source with full text (PMID:12791040) documents the allele (as E326K, the legacy designation of p.Glu365Lys) present in the father and a sister of a Gaucher disease family, i.e., inherited rather than de novo. Absence of positive de novo evidence, combined with population and familial transmission data, means PS2 is not met under the generic ACMG/AMP 2015 framework (the PD EP GBA1 v1.0 CSPEC rule payload was empty and could not govern).

Rule

ACMG/AMP 2015 PS2: De novo (both maternity and paternity confirmed) in a patient with the disease and no family history (PMID:25741868). No usable VCEP-specific rule text was retrievable (CSPEC 1833089217 payload empty), so the generic rule governs.

Evidence considered

• PMID:12791040 (Zhao et al., Clin Genet 2003): family segregation study documenting the E326K (= p.Glu365Lys) allele as an inherited allele present in the father (I-1) and sister (II-5) of a Gaucher disease family; no de novo occurrence reported.
• gnomAD v4.1: AF 0.0123122, 19,873/1,614,092 alleles, 196 homozygotes; variant is a common polymorphism incompatible with a de novo event for a proband.
• gnomAD v2.1: AF 0.0107311, 3,035/282,824 alleles, 37 homozygotes.
• ClinVar VCV000199044: 22 submissions contain no de novo or parental-testing evidence for the variant.

Still missing

• No proband or parental (trio) testing data are available in the evidence bundle, so de novo occurrence for the actual proband could not be directly evaluated; PS2 defaults to not_met on absence of positive evidence.

Sources

Reasoning

PS3 (well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product; PMID:25741868) is NOT met for GBA1 c.1093G>A (p.Glu365Lys; E326K in the mature-protein numbering used throughout the Gaucher literature). Two independent, internally controlled recombinant expression studies of this exact allele show substantial residual glucocerebrosidase activity: Montfort et al. 2004 (PMID:15146461, Sf9/baculovirus, CRIM-normalized) measured 42.7% of wild-type activity for E326K alone (54.8% when corrected to CRIM-specific activity), significantly higher than every other tested mutant except N188S, and far above the residual activity of bona fide Gaucher mutations assayed in the same system (N370S 11.6%, D409H 9.3%, L444P not detected); Liou et al. 2006 (PMID:16293621, Sf21/baculovirus, purified enzyme, CRIM specific activity 0.42 +/- 0.09 of wild-type, ~40-50%) describe E326K as a stable enzyme retaining about half of wild-type specific activity. The only in vivo quantitative data (PMID:26117366, dried blood spot glucocerebrosidase activity, PD cohort) show a modest reduction in E326K carriers vs non-carriers (P=0.009), i.e. a partial, not severe, deficit. No study demonstrates a deleterious loss-of-function effect; the authors of PMID:15146461 explicitly classify E326K as a 'modifier variant' that is nonpathogenic on its own (it is never found alone on a Gaucher disease-causing allele), and PMID:12791040 provides in vivo family evidence that isolated E326K is non-pathogenic. Residual activity in the ~43-55% range is incompatible with the near-complete loss of function (typically <15%) seen for disease-causing GBA1 missense alleles, so the functional evidence does not support PS3 at any strength.

Rule

Generic ACMG/AMP 2015 PS3 (PMID:25741868): 'Well-established in vitro or in vivo functional studies supportive of a damaging effect on the gene or gene product' - not met: validated recombinant enzyme assays reproducibly show ~43-55% residual glucocerebrosidase activity for E326K, far above the residual activity range of disease-causing GBA1 alleles measured in the same assays; no functional study demonstrates a deleterious effect.

Evidence considered

• PMID:15146461: E326K (c.1093G>A) expressed in Sf9/baculovirus with CRIM normalization: 42.7% of WT activity uncorrected, 54.8% CRIM-specific; significantly higher than all other mutants except N188S; never found alone in a Gaucher-causing allele; classified as 'modifier variant' - no deleterious LOF effect demonstrated.
• PMID:16293621: E326K purified enzyme: CRIM specific activity 0.42 +/- 0.09 (~40-50% of WT), stable; ~50% of WT phosphatidylserine activation; compound heterozygotes G202R/E326K and L444P/E326K maintain half-normal activity with decades of normalcy - retained substantial function.
• PMID:26117366: In vivo dried blood spot GCase activity modestly reduced in E326K carriers vs non-carriers (P=0.009); partial deficit only, not consistent with a deleterious LOF allele.
• PMID:12791040: In vivo genetic context (healthy 45-year-old compound heterozygote G202R+E326K; E326K in trans with G202R/L444P; isolated E326K non-pathogenic in family) consistent with retained function; not itself a functional assay.

Still missing

• No VCEP-calibrated enzyme-activity threshold for GBA1 PS3/BS3 is available (ClinGen PD EP GBA1 v1.0 rule payload is empty); assignment of an intermediate-zone (~40-55% residual activity) result to PS3 or BS3 would require a published, calibrated threshold.
• No CLIA-grade/clinical-diagnostic-laboratory-validated functional assay for this variant was identified; available assays are research-grade recombinant expression systems and a dried blood spot assay enriched for PD cases.
• If the intended classification context is Parkinson disease risk-factor status rather than Mendelian Gaucher disease, the partial activity reduction in carriers (PMID:26117366) is mechanistically relevant but does not satisfy PS3 as defined for a deleterious effect.

Sources

Reasoning

No statistically significant exact-variant case-control enrichment is available in this case's sources. In the only exact-variant case-control comparison retrieved (PMID:26117366, 517 PD cases vs 252 controls), E326K (c.1093G>A, p.Glu365Lys) was present in 2.5% (13/517) of PD cases vs 1.2% (3/252) of controls, P=0.289 (Table 2) - not significant. The same study's overall GBA mutation/variant carrier enrichment (17.0% vs 6.7%, P=0.001) pools many alleles and is not variant-specific, so it cannot satisfy PS4 for this exact variant. PMID:12791040 reports E326K at similar frequency in Gaucher disease patients (1.3%) and normal controls (0.9%), i.e., no enrichment for the Gaucher phenotype either. The large PD association studies for E326K cited in ClinVar (e.g., PMID:20947659) are not present in this case's source_registry and could not be evaluated. PS4 therefore not met on available exact-variant evidence; flagged for human review because the broader published E326K/PD risk-factor literature lies outside this case bundle.

Rule

Generic ACMG/AMP 2015 PS4 (Richards et al. 2015, PMID:25741868): prevalence of the variant in affected individuals significantly increased versus controls. ClinGen PD VCEP GBA1 v1.0 (cspec doc 1833089217) identified as governing framework but its rule payload was empty in this case; generic rule applied.

Evidence considered

• PMID:26117366 (Alcalay et al., Brain 2015): exact-variant case-control counts for E326K in PD cases vs controls - Table 2, 13/517 (2.5%) vs 3/252 (1.2%), P=0.289
• PMID:12791040 (Zhao et al., Hum Mutat 2003): E326K allele frequency 1.3% in Gaucher disease patients vs 0.9% in normal controls - no enrichment for Gaucher phenotype

Still missing

• No adequately powered exact-variant case-control study for c.1093G>A (E326K) is present in source_registry; the one available cohort comparison (PMID:26117366 Table 2) is non-significant (13/517 vs 3/252, P=0.289)
• Well-known PD association studies for E326K cited on the ClinVar record (e.g., PMID:20947659, PMID:18160183, PMID:29842932) are not in this case's source_registry and were not retrievable
• PD VCEP GBA1 v1.0 rule text for PS4 not retrievable (rule_payload empty), so PS4 was assessed under generic ACMG/AMP 2015

Sources

Reasoning

PM1 requires location in a mutational hotspot and/or critical well-established functional domain WITHOUT benign variation. Neither component is satisfied: (1) No statistical hotspot - cancerhotspots.org returned no result for GBA1 E365. (2) Structural data (Liou et al. 2006, PMID:16293621) show Glu326 (Glu365 in current numbering) is a surface-accessible residue of domain 3, not in the active site; E326K produced the most modest activity reduction (42% of wild-type). (3) The residue carries abundant benign variation: p.Glu365Lys is a common polymorphism (gnomAD v4.1 AF 1.23%, 196 homozygotes, grpmax FAF 1.32%) and is explicitly described as a polymorphic, non-pathogenic allele (PMID:12791040). PM1's 'without benign variation' condition clearly fails. PM1 not met.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PM1 - Located in a mutational hotspot and/or critical and well-established functional domain (e.g., active site of an enzyme) without benign variation.

Evidence considered

• PMID:16293621 - 'side chains of Lys79, Lys198, Tyr304, Glu326, Glu349, and Arg463 are surface-accessible' and 'The most modest reduction in enzymatic activity was associated with the surface mutation E326K (42% activity) in domain 3' - residue is surface-exposed, not active-site/critical domain.
• PMID:12791040 - E326K described as a polymorphic allele (benign variation exists at this residue), violating PM1's 'without benign variation' condition.
• gnomad_v4 - variant present at AF 1.23% with 196 homozygotes, grpmax FAF 1.32% - high population frequency of a missense at this residue, inconsistent with a critical domain devoid of benign variation.
• PMID:25741868 - generic ACMG/AMP PM1 rule definition applied (hotspot check via cancerhotspots.org was negative for GBA1 E365).

Still missing

• CSPEC PD VCEP GBA1 v1.0 may specify a custom PM1 domain list; the retrieved framework payload was empty so no VCEP-specific PM1 definition could be applied.

Sources

Reasoning

PM2 (pathogenic supporting) requires the variant to be absent from, or at extremely low frequency in, reference populations (PMID 25741868: 'Absent from controls (or at extremely low frequency if recessive)'). This variant is common: gnomAD v4.1 AF 1.2312% (19,873/1,614,092 alleles), gnomAD v2.1 AF 1.0731% (3,035/282,824), gnomAD-Canada AF 0.7980% (147/18,422). The frequency is orders of magnitude above the pipeline's documented PM2<0.1% non-VCEP convention and above any recessive-disease allele expectation, so PM2 is definitively not met regardless of framework. No population-frequency support for pathogenicity exists for this variant.

Evidence considered

• gnomAD v4.1 AF 1.2312% (19,873 alleles) - far above PM2<0.1% convention and any 'extremely low frequency' standard
• gnomAD v2.1 AF 1.0731%; gnomAD-Canada AF 0.7980%
• PM2 definition per ACMG/AMP 2015 (Richards et al., PMID 25741868)

Sources

Reasoning

PM3 (recessive disorder: variant detected in trans with a pathogenic variant) is not met. The governing ClinGen Parkinson's Disease EP GBA1 CSPEC v1.0 was identified (cspec doc 1833089217) but its rule payload was empty (framework_complete=false), so the generic ACMG/AMP 2015 definition (PMID 25741868) was applied. Direct phase evidence from PMID:12791040 (single-allele sequencing of a type 3 Gaucher family) shows c.1093G>A (p.Glu365Lys, legacy E326K) in trans with the pathogenic G202R allele in the phenotypically normal father (healthy at 45 y) and in trans with the pathogenic L444P allele in the proband's healthy 11-year-old sister (L444P/E326K genotype, leukocyte GCase activity ~45% of normal); the affected proband (case 4) and affected brother carry G202R/L444P without E326K. The in-trans compound-heterozygous carriers are unaffected, so the observation actively refutes a recessive pathogenic role rather than supporting PM3, and no affected proband with E326K in trans with a pathogenic variant was identified in any reviewed source. Consistently, gnomAD v2.1 reports 37 homozygotes and gnomAD v4.1 reports 196 homozygotes (homozygous E326K individuals occur in the general population without Gaucher disease); PMID:15146461 reports E326K was never found alone on a Gaucher disease-causing allele; PMID:26117366 categorizes E326K among variants associated with Parkinson's risk but not with Gaucher disease. All available allelic evidence points against PM3.

Rule

Generic ACMG/AMP 2015 (PMID 25741868): PM3 - for recessive disorders, detected in trans with a pathogenic variant. Applied because the governing ClinGen Parkinson's Disease EP GBA1 CSPEC v1.0 had an empty rule payload (framework_complete=false).

Evidence considered

• PMID:12791040: family segregation with single-allele/allele-specific sequencing shows E326K in trans with pathogenic G202R (healthy father, 45 y) and in trans with pathogenic L444P (healthy sister, 11 y, GCase activity ~45% of normal); affected sibs carry G202R/L444P without E326K; authors conclude isolated E326K is non-pathogenic in this family.
• PMID:15146461: c.1093G>A (p.E326K) was never found alone on a Gaucher disease-causing allele; it occurs only in cis within complex alleles, so it is not itself a recessive disease allele.
• PMID:26117366: E326K explicitly grouped with GBA variants 'associated with Parkinson's risk but not with Gaucher disease'.
• gnomad_v2: AF=1.073% (3035/282824 alleles), 37 homozygotes; gnomad_v4: AF=1.231% (19873/1614092 alleles), 196 homozygotes — homozygous E326K carriers present in the general population.

Still missing

• No affected proband with Gaucher disease carrying E326K in trans with a pathogenic GBA1 variant was identified in the available sources; every in-trans observation (with G202R and L444P) is in a phenotypically normal carrier (PMID:12791040).
• The GBA1 CSPEC (Parkinson's Disease EP v1.0) supplied no retrievable per-criterion PM3 rule in this case's bundle; generic ACMG/AMP 2015 (PMID 25741868) was used.

Sources

Reasoning

PM5 requires a novel missense change at an amino acid residue where a DIFFERENT missense change has been determined to be pathogenic. No different missense change at residue Glu365 was identified in any available source: the PM5 comparator candidate collection (pm5_candidates) contains 0 same-residue candidates, and none of the consulted full texts report any alternate substitution at residue 365. The only characterized change at this residue, p.Glu365Lys, is non-pathogenic (PMID:12791040; PMID:15146461), and the residue is a common polymorphic site (gnomAD v4.1 AF 1.23%), making a pathogenic same-residue comparator unsupported by the available evidence. PM5 not met on available evidence; a dedicated same-residue variant sweep remains an open gap.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PM5 - Novel missense change at an amino acid residue where a different missense change determined to be pathogenic has been seen before.

Evidence considered

• pm5_candidates - candidate collection contains 0 total/same-residue candidates for residue Glu365; no pathogenic comparator identified.
• PMID:12791040 + PMID:15146461 - only documented change at this residue is p.Glu365Lys (legacy E326K), which is non-pathogenic/polymorphic; no different pathogenic missense at the same residue reported.
• PMID:25741868 - generic ACMG/AMP PM5 rule definition applied.

Still missing

• No systematic ClinVar/LOVD/literature sweep for other missense changes at residue 365 (e.g., p.Glu365Asp, p.Glu365Ala) was completed - the automated PM5 candidate harvest was skipped (pm5_candidates: mode 'unknown', 0 candidates, queries not executed). If a pathogenic same-residue comparator exists outside the reviewed sources, PM5 would need re-assessment.

Sources

Reasoning

PM6 (assumed de novo, without confirmation of paternity/maternity) requires evidence that the variant arose de novo in the proband. No such evidence exists in the bundle: no proband or parental samples are described, and no ClinVar submission or publication reports an assumed de novo occurrence. Moreover, the variant is a common polymorphism (gnomAD v4.1 AF 1.23%, 196 homozygotes; gnomAD v2.1 AF 1.07%, 37 homozygotes) and is documented as inherited from the father in the PMID:12791040 family (E326K legacy = p.Glu365Lys). A variant at ~1.2% population frequency with numerous homozygotes is implausible as an assumed de novo event, so PM6 is not met under the generic ACMG/AMP 2015 fallback.

Rule

ACMG/AMP 2015 PM6: Assumed de novo, but without confirmation of paternity and maternity (PMID:25741868). No usable VCEP-specific rule text was retrievable; generic rule governs.

Evidence considered

• PMID:12791040: E326K allele transmitted in a family (father I-1, sister II-5), i.e., inherited, not de novo.
• gnomAD v4.1: AF 0.0123122 with 196 homozygotes and gnomAD v2.1: AF 0.0107311 with 37 homozygotes; allele is a common polymorphism, making assumed de novo origin implausible.
• ClinVar VCV000199044: no submission reports de novo or parental-testing data.

Still missing

• No proband/parental trio data available to directly test assumed de novo status; criterion defaults to not_met on absence of any de novo evidence and is affirmatively contradicted by population frequency and documented familial transmission.

Sources

Reasoning

PP1 (cosegregation with disease in multiple affected family members in a gene definitively known to cause the disease) is not met. No source provides co-segregation of this variant with disease in affected family members; the only segregation data available show the opposite. PMID:12791040 (full text) demonstrates that in a Gaucher disease type 3 family, the affected proband (II-4, G202R/L444P) does not carry E326K (= p.Glu365Lys) while unaffected relatives (father I-1, G202R/E326K; sister II-5, L444P/E326K, age 11) do, with the authors concluding the alleles 'segregate independently' and that isolated E326K is non-pathogenic in this family. This is corroborated by the LabCorp ClinVar submission (SCV000697574) whose assertion text states the variant 'did not co-segregate with disease in at-least one family with a clinical diagnosis of gaucher's disease'. Co-segregation is affirmatively contradicted rather than merely absent, so PP1 is not met under the generic ACMG/AMP 2015 fallback.

Rule

ACMG/AMP 2015 PP1: Cosegregation with disease in multiple affected family members in a gene definitively known to cause the disease (PMID:25741868). No usable VCEP-specific rule text was retrievable; generic rule governs.

Evidence considered

• PMID:12791040 (Fig. 1 legend and Results/Discussion): affected proband II-4 lacks E326K while unaffected I-1 and II-5 carry it; 'the bearing of the mutations encoding E326K, G202R or L444P segregate independently'; 'This supports E326K as a polymorphic allele that, at least in this family, is non-pathogenic.'
• ClinVar LabCorp SCV000697574 assertion signal: 'this variant did not co-segregate with disease in at-least one family with a clinical diagnosis of gaucher's disease'.

Still missing

• No segregation data specific to Parkinson's disease (the phenotype targeted by the governing PD EP VCEP) are available; only Gaucher-disease-family non-segregation exists, which contradicts rather than supports PP1.

Sources

Reasoning

PP2 is a two-pronged criterion requiring (a) a gene with a LOW rate of benign missense variation AND (b) missense as a common mechanism of disease. Prong (b) is satisfied for GBA1 (Gaucher disease and GBA1-related parkinsonism are predominantly driven by missense alleles, e.g., N409S/N370S, L483P/L444P). Prong (a) fails: GBA1 has a well-documented high rate of benign/polymorphic missense variation. p.Glu365Lys itself is a common polymorphism - gnomAD v4.1 AF 1.23% with 196 homozygotes (grpmax FAF 1.32%) - described as a 'polymorphic allele' that is 'non-pathogenic' (PMID:12791040, with similar frequencies in Gaucher patients 1.3% vs controls 0.9%) and as a 'modifier variant rather than a neutral polymorphism' that was 'never found alone in a Gaucher disease-causing allele' (PMID:15146461); it is found in phenotypically normal individuals (PMID:16293621). Because the gene does not have a low rate of benign missense variation, PP2 is not met.

Rule

Generic ACMG/AMP 2015 (PMID:25741868): PP2 - Missense variant in a gene that has a low rate of benign missense variation and in which missense variants are a common mechanism of disease (both prongs required).

Evidence considered

• PMID:12791040 - 'The similar frequencies of this mutation in Gaucher's disease patients (1.3%) and normal controls (0.9%) support this contention... This supports E326K as a polymorphic allele that, at least in this family, is non-pathogenic' - direct evidence of benign missense variation in GBA1.
• PMID:15146461 - c.1093G>A (p.E326K) 'was never found alone in a Gaucher disease-causing allele' and is a 'modifier variant' - non-disease-causing missense in GBA1.
• PMID:16293621 - E326K present in phenotypically normal individuals; missense is the common mechanism of Gaucher disease (52 missense alleles analyzed).
• gnomad_v4 - AF 1.23% (19873/1614092 alleles), 196 homozygotes - common benign missense allele, incompatible with a low rate of benign missense variation.
• PMID:25741868 - generic ACMG/AMP PP2 rule definition applied.

Sources

Reasoning

No calibrated computational evidence supports a deleterious effect on the gene or gene product. Splice path: SpliceAI max delta score = 0.00 (pangolin_SG 0.01, pangolin_SL -0.01) for NM_000157.4:c.1093G>A, far below the delta >= 0.2 threshold for potential splice alteration (SpliceAI, Jaganathan et al. 2019, PMID 30661751); no splice impact predicted. Missense path: REVEL score 0.595 falls in the uninformative middle range of the ClinGen SVI calibration (PP3 supporting >= 0.773, PP3 strong >= 0.932; Pejaver et al. 2022, PMID 36413997); BayesDel score -0.0147355 has no verified published threshold available to this pipeline and is treated as insufficiently calibrated (unusable); aGVGD was not fetched. ClinVar submission audit corroborates: no submitter reports computational support for pathogenicity - LabCorp (SCV000697574) states 3/5 in silico tools predict a benign effect, Broad (SCV001422690) states tools/conservation provide no strong support for or against impact, VCGS (SCV005086522) states the missense variant is predicted tolerated. PP3 therefore not met.

Rule

Generic ACMG/AMP 2015 PP3: 'Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.)' (PMID 25741868); applied with published calibrated predictor thresholds because the governing VCEP (ClinGen PD EP GBA1 v1.0) rule payload was not retrievable

Evidence considered

• SpliceAI max delta score 0.00 (pangolin_SG 0.01, pangolin_SL -0.01) for NM_000157.4:c.1093G>A; no splice impact predicted; threshold delta >= 0.2 for potential splice alteration per SpliceAI publication (Jaganathan et al., Nat Genet 2019, PMID 30661751); 0.00 is far below 0.2, so this does not support PP3
• REVEL score 0.595; ClinGen SVI calibrated thresholds (Pejaver et al., Am J Hum Genet 2022, PMID 36413997): PP3 supporting >= 0.773, PP3 strong >= 0.932, BP4 supporting <= 0.073, BP4 strong <= 0.016; 0.595 lies in the uninformative range, so it neither supports nor opposes PP3
• BayesDel score -0.0147355; no verified published threshold available to this pipeline (no named publication/table), so treated as insufficiently calibrated and not usable for PP3
• ClinVar submission audit (exact-variant submissions): no submitter cites computational evidence supporting a deleterious effect; SCV000697574 (LabCorp) 'three of five in-silico tools predict a benign effect of the variant on protein function'; SCV001422690 (Broad) 'computational prediction tools and conservation analyses do not provide strong support for or against an impact to the protein'; SCV005086522 (VCGS) 'missense variant consistently predicted to be tolerated by multiple in silico tools'; corroborating context only (unnamed tools, no scores/thresholds)
• ACMG/AMP 2015 standards (Richards et al., Genet Med 2015, PMID 25741868) define PP3 as 'Multiple lines of computational evidence support a deleterious effect on the gene or gene product' and caution that in silico algorithms sharing inputs are not independent and PP3 may be used only once; applied as the generic framework because the governing VCEP rule payload was not retrievable

Still missing

• aGVGD missense predictor not fetched for this case; would have provided an additional missense-path line
• Governing VCEP (ClinGen Parkinson's Disease EP GBA1 v1.0) rule payload not retrievable in this case (framework_complete=false, rule_payload empty); VCEP-specific PP3 treatment unknown, so generic ACMG/AMP 2015 PP3 was applied

Sources

Reasoning

PP4 requires the patient's phenotype or family history to be highly specific for a disease with a single genetic etiology. No such phenotype specificity exists for this variant. Family segregation data show that isolated E326K (c.1093G>A, p.Glu365Lys) is non-pathogenic for Gaucher disease: PMID:12791040 found E326K in trans with the disease-causing alleles G202R and L444P and concluded it is 'a polymorphic allele that, at least in this family, is non-pathogenic'. PMID:15146461 (expression studies) states the c.1093G>A (p.E326K) change 'was never found alone in a Gaucher disease-causing allele'. For Parkinson disease the variant behaves as a risk factor (PMID:26117366), and PD is a genetically heterogeneous disorder, not a single-genetic-etiology phenotype. Consistent with this, the variant is common in the general population (gnomAD v2.1 AF 1.07%, 37 homozygotes; Finnish AF 4.32%, 25 homozygotes), incompatible with a highly penetrant single-etiology disease phenotype. PP4 not met.

Rule

Generic ACMG/AMP 2015 PP4 (Richards et al. 2015, PMID:25741868): patient's phenotype or family history highly specific for a disease with a single genetic etiology. ClinGen PD VCEP GBA1 v1.0 (cspec doc 1833089217) identified as governing framework but its rule payload was empty in this case; generic rule applied.

Evidence considered

• PMID:12791040 (Zhao et al., Hum Mutat 2003): family data - isolated E326K non-pathogenic, segregates in trans with G202R/L444P; supports polymorphic, non-disease-causing status
• PMID:15146461 (Montfort et al., Hum Mutat 2004): expression studies - c.1093G>A (p.E326K) never found alone in a Gaucher disease-causing allele
• PMID:26117366 (Alcalay et al., Brain 2015): E326K described as a GBA variant associated with PD risk but not with Gaucher disease
• gnomad_v2 (gnomAD v2.1, variant 1-155206167-C-T): AF 0.0107311 (3035/282824 alleles, 37 homozygotes); highest in Finnish population AF 0.0432497

Still missing

• PD VCEP GBA1 v1.0 rule text for PP4 not retrievable (rule_payload empty); assessed under generic ACMG/AMP 2015

Sources

Reasoning

PP5 applies only when an exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic exists. The ClinVar record for this variant (VCV000199044, variation 199044) has zero expert-panel submissions (expert_panels: {}; submission audit expert_panel_submissions: 0; review_stars: 1). All 22 submissions are from clinical laboratories/centers (8 Benign, 5 Likely benign, 4 risk factor, 2 Uncertain significance, 1 Likely pathogenic) with 'criteria provided, single submitter' or 'no assertion criteria provided' status. Per the restricted PP5 policy, ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions must never trigger PP5. No qualifying expert-panel classification exists, so PP5 is not met.

Rule

PP5 restricted per pipeline policy: only an exact-variant ClinVar expert-panel classification (Pathogenic/Likely pathogenic) qualifies (Richards et al. 2015 PP5, PMID:25741868).

Evidence considered

• clinvar (VCV000199044): expert_panels empty, expert_panel_submissions 0, review_stars 1; submissions summary 8 Benign / 5 Likely benign / 4 risk factor / 2 VUS / 1 Likely pathogenic from single-submitter laboratories

Still missing

• No ClinVar expert-panel (3-star) Pathogenic/Likely pathogenic classification exists for c.1093G>A to enable PP5

Sources

Reasoning

Governing framework: ClinGen Parkinson's Disease Expert Panel GBA1 v1.0 specification (cspec doc 1833089217) is identified but its rule_payload is empty (framework_complete=false), so generic ACMG/AMP 2015 (Richards et al., PMID 25741868) population criteria apply. BA1 requires allele frequency >5% in a reference population (PMID 25741868, Table 4). This variant is below that threshold in every dataset: gnomAD v4.1 overall AF 1.2312% (grpmax FAF 1.3208%), highest subpopulation (Finnish) AF 4.4255%; gnomAD v2.1 overall AF 1.0731% (grpmax FAF 1.2600%), Finnish 4.3250%; gnomAD-Canada 0.7980%. BA1 is additionally inappropriate here because c.1093G>A (p.Glu365Lys, legacy E326K) is a documented low-penetrance Parkinson's disease risk allele rather than a benign variant: PMID 26117366 states E326K 'predisposes to Parkinson's disease, but does not cause Gaucher's disease' and shows reduced glucocerebrosidase activity in E326K carriers (P=0.009), and PMID 15146461 concludes E326K should be considered a 'modifier variant' rather than a neutral polymorphism. The high population frequency therefore reflects common risk-allele status, not stand-alone benignity. Note: the variant's AF (1.23%) exceeds this pipeline's documented non-VCEP BA1>1% convention, but applying stand-alone benignity at that frequency would conflict with the risk-allele evidence; the threshold conflict and the missing VCEP rule payload require human adjudication.

Evidence considered

• gnomAD v4.1 overall AF 1.2312% (19,873/1,614,092 alleles), 196 homozygotes, grpmax FAF 1.3208%, Finnish AF 4.4255% (2,833/64,016)
• gnomAD v2.1 overall AF 1.0731% (3,035/282,824 alleles), 37 homozygotes, grpmax FAF 1.2600%, Finnish AF 4.3250% (1,086/25,110)
• gnomAD-Canada AF 0.7980% (147/18,422 alleles), 1 homozygote
• BA1 >5% allele-frequency stand-alone threshold per ACMG/AMP 2015 (Richards et al., PMID 25741868)
• E365K documented as PD risk allele, not benign: PMID 26117366 ('predisposes to Parkinson's disease, but does not cause Gaucher's disease'; reduced GCase activity P=0.009); PMID 15146461 ('modifier variant')

Still missing

• PD EP GBA1 VCEP rule payload not available in case bundle (cspec doc 1833089217, rule_payload empty) - VCEP-specific BA1 treatment for GBA1 risk alleles (e.g., E365K) is unknown
• Variant AF 1.23% exceeds the pipeline's non-VCEP BA1>1% convention but is below the published ACMG/AMP >5% stand-alone threshold (PMID 25741868); which threshold governs for this GBA1 PD-framework case is unresolved without VCEP rules
• gnomAD flags segdup=true at this locus (GBA1/GBAP1 pseudogene paralogy); the frequency is corroborated by literature-based Sanger genotyping (PMID 12791040, PMID 26117366), but site-quality is a residual consideration

Sources

Reasoning

BS3 (well-established in vitro or in vivo functional studies show no damaging effect on protein function or splicing; PMID:25741868) is NOT met. The same well-established functional studies that argue against PS3 demonstrate a reproducible, measurable partial functional deficit for E326K: ~2-fold reduction in enzyme activity (42.7% uncorrected / 54.8% CRIM-specific of WT in PMID:15146461; ~40-50% CRIM specific activity, 0.42 +/- 0.09, in PMID:16293621), significantly diminished phosphatidylserine activation (~50% of WT; PMID:16293621), significantly lower in vivo dried blood spot glucocerebrosidase activity in carriers vs non-carriers (P=0.009; PMID:26117366), and an additive activity decrease when E326K is present in cis with N188S or L444P double-mutant alleles (PMID:15146461). E326K therefore is not functionally neutral - the residual activity of ~40-55% of WT falls in an intermediate zone between the near-complete loss seen in pathogenic GBA1 alleles (<~15%) and the normal range, so the criterion 'no damaging effect on protein function' is not established by the available data. Flagged for human review because the interpretation is context-dependent: for Mendelian Gaucher disease pathogenicity the retained ~50% activity argues against a pathogenic allele (consistent with ClinVar's predominantly benign/likely benign submissions), whereas for Parkinson disease risk-factor biology the partial loss of activity is mechanistically relevant, and this variant's ClinVar submissions are conflicting (8 Benign, 5 Likely benign, 4 risk factor, 2 VUS, 1 Likely pathogenic).

Rule

Generic ACMG/AMP 2015 BS3 (PMID:25741868): 'Well-established in vitro or in vivo functional studies show no damaging effect on protein function or splicing' - not met: the available well-established studies reproducibly show a partial but real functional deficit (approximately 2-fold reduced enzyme activity, reduced phosphatidylserine activation, reduced in vivo activity in carriers), so 'no damaging effect' is not established; the effect is intermediate between benign and pathogenic ranges.

Evidence considered

• PMID:15146461: E326K alone retains 42.7% (54.8% CRIM-specific) of WT activity - a ~2-fold reduction, i.e., a demonstrable effect on protein function; double mutants [N188S;E326K] 23.2% and [L444P;E326K] 6-14% show additive loss, arguing against functional neutrality.
• PMID:16293621: E326K CRIM specific activity ~40-50% of WT with significantly diminished phosphatidylserine activation - partial functional impairment, not 'no damaging effect'; stability is normal.
• PMID:26117366: E326K carriers have significantly lower in vivo GCase activity than non-carriers (P=0.009) - direct evidence of a functional effect in vivo.
• PMID:12791040: Healthy 45-year-old with G202R+E326K and segregation data indicate isolated E326K is non-pathogenic for Gaucher disease in this family; consistent with retained function but does not establish 'no damaging effect' at the molecular level.

Still missing

• A calibrated benign-range enzyme-activity threshold for GBA1 (e.g., activity overlapping wild-type control distribution, per a published functional-assay calibration) would be required to upgrade this intermediate-effect finding to BS3; no such VCEP/SVI-calibrated threshold is available in this case's sources.
• The dried blood spot assay (PMID:26117366) is research-grade, not CLIA-validated, and the E326K carrier group is enriched for PD cases (82.7% of heterozygotes), limiting its use as a quantitative benign-function assay.
• No assay directly interrogates splicing; SpliceAI max delta score 0.00 (evidence.json) is computational only and was not used as functional evidence for BS3.

Sources

Reasoning

BP4 requires multiple lines of computational evidence suggesting no impact on the gene or gene product, and per ACMG/AMP 2015 the predictions should not be used when in silico tools disagree. Only one calibrated line is available here: SpliceAI max delta 0.00 (no splice impact, far below the delta >= 0.2 actionable threshold, PMID 30661751). The missense path does not add a line: REVEL 0.595 is uninformative under the ClinGen SVI calibration (PMID 36413997); BayesDel -0.0147355 is unusable because no verified published threshold is available to this pipeline; aGVGD was not fetched. ClinVar submission audit shows the in silico evidence as neutral-to-benign but non-consensual - LabCorp reports only 3/5 tools predict benign (i.e., predictions disagree, which per ACMG 2015 means the evidence should not be used), Broad reports no strong support either way, VCGS reports tolerated. The 'multiple lines' requirement of BP4 is therefore not satisfied with the currently available calibrated evidence, so BP4 is not met. Human review is flagged because the governing VCEP (ClinGen PD EP GBA1 v1.0) PP3/BP4 rules could not be retrieved and may pre-assign treatment for this common variant (E365K/E326K).

Rule

Generic ACMG/AMP 2015 BP4: 'Multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.)' (PMID 25741868); applied with published calibrated predictor thresholds because the governing VCEP (ClinGen PD EP GBA1 v1.0) rule payload was not retrievable

Evidence considered

• SpliceAI max delta score 0.00 (pangolin_SG 0.01, pangolin_SL -0.01) for NM_000157.4:c.1093G>A; no splice impact predicted; threshold delta >= 0.2 for potential splice alteration per SpliceAI publication (Jaganathan et al., Nat Genet 2019, PMID 30661751); 0.00 << 0.2 provides one line of support for BP4 (splice path)
• REVEL score 0.595; ClinGen SVI calibrated thresholds (Pejaver et al., Am J Hum Genet 2022, PMID 36413997): BP4 supporting <= 0.073, BP4 strong <= 0.016; 0.595 is far above these, so REVEL provides no support for BP4
• BayesDel score -0.0147355 (benign direction); no verified published threshold available to this pipeline (no named publication/table), so treated as insufficiently calibrated and not usable for BP4
• ClinVar submission audit (exact-variant submissions): SCV000697574 (LabCorp) 'three of five in-silico tools predict a benign effect of the variant on protein function' (only 3/5 = disagreement among tools, which per ACMG 2015 means in silico evidence should not be used); SCV001422690 (Broad) 'computational prediction tools and conservation analyses do not provide strong support for or against an impact to the protein'; SCV005086522 (VCGS) 'missense variant consistently predicted to be tolerated by multiple in silico tools'; corroborating context only, not counted as calibrated lines
• ACMG/AMP 2015 standards (Richards et al., Genet Med 2015, PMID 25741868) define BP4 as 'Multiple lines of computational evidence suggest no impact on gene or gene product' and state: 'If all of the in silico programs tested agree on the prediction, then this evidence can be counted as supporting. If in silico predictions disagree, however, then this evidence should not be used in classifying a variant.' Applied as the generic framework because the governing VCEP rule payload was not retrievable

Still missing

• Only one calibrated line of computational evidence (SpliceAI no-impact) is available; REVEL is uninformative, BayesDel is uncalibrated for this pipeline, and aGVGD was not fetched, so the BP4 'multiple lines' requirement cannot be satisfied with current data
• Governing VCEP (ClinGen Parkinson's Disease EP GBA1 v1.0) rule payload not retrievable in this case (framework_complete=false, rule_payload empty); a gene-specific PP3/BP4 lookup or pre-assignment could not be checked - flagged for human review
• ClinVar submitter in silico statements do not name tools, scores, or thresholds, so they cannot be counted as calibrated lines of evidence

Sources

Reasoning

BP6 applies only when an exact-variant ClinVar expert-panel classification of Benign/Likely benign exists. The ClinVar record (VCV000199044) has zero expert-panel submissions (expert_panels: {}; expert_panel_submissions: 0; review_stars: 1). The aggregate ClinVar label is 'Benign'/'Conflicting classifications of pathogenicity; risk factor' derived from single-submitter laboratory submissions (8 Benign, 5 Likely benign, 4 risk factor, 2 VUS, 1 Likely pathogenic) - none of which qualifies as an expert-panel assertion. Per the restricted BP6 policy, ordinary laboratory submissions and aggregate ClinVar labels must never trigger BP6. BP6 not met.

Rule

BP6 restricted per pipeline policy: only an exact-variant ClinVar expert-panel classification (Benign/Likely benign) qualifies (Richards et al. 2015 BP6, PMID:25741868).

Evidence considered

• clinvar (VCV000199044): expert_panels empty, expert_panel_submissions 0, review_stars 1; aggregate label 'Conflicting classifications of pathogenicity; risk factor' from single-submitter laboratory submissions

Still missing

• No ClinVar expert-panel (3-star) Benign/Likely benign classification exists for c.1093G>A to enable BP6

Sources