PRKN c.758G>A, p.(Cys253Tyr)

Reasoning

Cys253 lies in exon 7 (variant validator: exon 7 of 12) within the RING1 cysteine/histidine-rich zinc-binding domain of parkin. PMID:18519021 (Sironi et al. 2008) states the RING1 domain is 'a cysteine and histidine rich domain... expected to bind two zinc ions in a cross-brace way using a Cys3HisCys4 motif and forming zinc-finger structures' and that 'the three domains together represent the critical zone for topology- and E3 ubiquitin ligase activity of the parkin protein'; it locates p.R275W 'in exon 7 in the RING1 domain' and p.C238W as affecting 'the first cysteine of the RING1 domain' — Cys253 is 15 residues C-terminal of Cys238 and 22 residues N-terminal of Arg275, i.e., within the same RING1 zinc-finger cluster in exon 7. Multiple independent pathogenic missense variants cluster in this domain (p.C238W, p.T240M, p.C253Y, p.R275W; PMID:18519021 lists p.C253Y among 'known pathogenic changes'), consistent with a mutation cluster in a critical, well-established functional domain. No benign variation is documented at residue 253: gnomAD v2.1 AF 1.59e-05 (grpmax FAF 7.02e-06) and gnomAD v4.1 AF 6.81e-06 (FAF 4.29e-06), zero homozygotes, and no benign ClinVar assertion in the bundle. PM1 is therefore met at moderate (native PM) strength. Human review is flagged because the retrieved sources do not explicitly name Cys253 as a zinc-coordinating RING1 residue (membership is inferred from exon 7 position between p.C238, the first RING1 cysteine, and p.R275, in the RING1 domain), and the 'domains must lack benign variants' component was not exhaustively verified with a systematic benign-variant scan.

Rule

ACMG/AMP 2015 (Richards et al., PMID:25741868) PM1: 'Mutational hot spot and/or critical and well-established functional domain... Certain protein domains are known to be critical to protein function and all missense variants identified to date in these domains have been shown to be pathogenic. These domains must also lack benign variants.' Applied at native moderate strength: RING1 zinc-binding domain critical for parkin E3 ligase activity, pathogenic missense clustering (C238W, T240M, C253Y, R275W), no benign variation at Cys253.

Evidence considered

• PMID:18519021 - Discussion (p.330): RING1 domain is a cysteine/histidine-rich domain binding two zinc ions via Cys3HisCys4 motif; TRIAD (RING1-IBR-RING2) is the critical zone for parkin E3 ubiquitin ligase activity; p.R275W located 'in exon 7 in the RING1 domain'; p.C238W affects 'the first cysteine of the RING1 domain'
• PMID:18519021 - Results 3.1 and Table 1/2: pathogenic missense clustering in RING1 region (p.C238W, p.T240M, p.C253Y, p.R275W)
• PMID:16769863 - Figure 1 gene map marks the ubiquitin-like domain and RING-finger motifs; c.758G>A placed on the map (Table 1 Family 4: p.Cys253Tyr)
• gnomad_v2 - variant 6-162206917-C-T: AF 1.59e-05, 4/251238 alleles, 0 homozygotes, grpmax FAF 7.02e-06 — no common/benign allele at this residue
• gnomad_v4 - variant 6-161785885-C-T: AF 6.81e-06, 11/1614112 alleles, 0 homozygotes, grpmax FAF 4.29e-06

Still missing

• Exact zinc-coordinating assignment of Cys253 within the RING1 Cys3HisCys4 motif is not explicitly stated in the retrieved sources; RING1 membership for Cys253 is inferred from exon 7 position between p.C238 (first RING1 cysteine) and p.R275 (RING1 domain)
• 'Without benign variation' component not exhaustively verified — no systematic benign-variant scan of the RING1 domain or residue 253 present in the bundle

Sources

Reasoning

PM2 is met: the variant is present at extremely low frequency in population databases and is absent from a control/genome cohort. gnomAD v2.1 exome total AF 1.592e-05 (0.00159%, 4/251,238 alleles) and gnomAD v4.1 exome total AF 6.815e-06 (0.00068%, 11/1,614,112 alleles) are both well below the project non-VCEP PM2 convention of 0.1% (and far below the commonly used 0.01% recessive threshold). grpmax FAF is 7.02e-06 (v2.1) and 4.29e-06 (v4.1), homozygote count is 0 in both datasets, and the variant is absent from gnomAD-Canada v1.0. One threshold artifact was examined: the gnomAD v2.1 NFE_BGR (Bulgarian) subpopulation shows AF 0.112% (3/2,670 alleles), which exceeds the 0.1% convention if a strict subpopulation-maximum rule were applied. This is a small-sample/founder artifact: it is not replicated in gnomAD v4.1 (NFE AF 8.47e-06 across 1,179,982 alleles), and gnomAD's filtering allele frequency (grpmax FAF 7.02e-06/4.29e-06), which is the metric designed to be robust to such subpopulation noise, remains far below 0.1%. Literature context is consistent: the variant is a genuine disease allele observed homozygous/compound-heterozygous in affected early-onset PD patients (PMID:18519021, PMID:16769863), not a common polymorphism. PM2 applied at supporting strength per generic ACMG/AMP 2015 convention.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID:25741868) PM2: 'Absent from controls (or at extremely low frequency if recessive).' Project non-VCEP convention: PM2 <0.1%. Observed gnomAD total AF 0.00159% (v2.1) / 0.00068% (v4.1) and grpmax FAF 7.02e-06 / 4.29e-06 are well below 0.1%; 0 homozygotes; absent from gnomAD-Canada - met at supporting strength.

Evidence considered

• gnomad_v2 - gnomAD v2.1 exome: total AF 1.592e-05 (0.00159%, 4/251,238), 0 homozygotes, grpmax FAF 7.02e-06
• gnomad_v4 - gnomAD v4.1 exome: total AF 6.815e-06 (0.00068%, 11/1,614,112), 0 homozygotes, joint grpmax FAF 4.29e-06
• gnomad_canada - variant absent from gnomAD-Canada v1.0
• PMID:18519021 - variant observed homozygous in an affected EOPD proband, consistent with a rare recessive disease allele rather than a benign polymorphism
• PMID:16769863 - variant observed compound-heterozygous in affected family members only, consistent with a rare disease allele

Still missing

• The gnomAD v2.1 NFE_BGR (Bulgarian) subpopulation AF of 0.112% (3/2,670 alleles) exceeds the 0.1% convention; treated as a threshold artifact because it is absent from gnomAD v4.1 and inconsistent with grpmax FAF (7.02e-06). If the reviewer applies a strict maximum-subpopulation-AF rule rather than grpmax FAF, this cluster should be reconfirmed against the current gnomAD release

Sources

Reasoning

PRKN (PARK2)-related early-onset parkinsonism (AR-JP2, OMIM 600116) is autosomal recessive, so PM3 applies. The exact variant NM_004562.2:c.758G>A (p.Cys253Tyr / p.C253Y) is documented as HOMOZYGOUS (c.859G>A / c.859G>A in the older transcript numbering used by the authors) in an affected early-onset PD proband (patient K-0010, onset age 20 years) in Sironi et al. 2008 (PMID:18519021, Table 1 and Table 2). Homozygosity places the variant in trans with itself (biallelic) with phase inherently confirmed, in an affected proband with a recessive disorder - the core PM3 observation. Strength is set to moderate: a single affected homozygous proband (equivalent to two in-trans observations under SVI PM3/BS2 counting) supports PM3 at moderate; upgrade to strong would require additional independent affected probands with the variant in trans, which are not present in the available full texts (GenePD PMID:16769863 and Solve-RD PMID:39825153 do not name this variant; full text of PMID:26830385 was not retrieved). The c.859G>A vs c.758G>A numbering difference is a transcript-version artifact (identical protein consequence p.C253Y), anchored by ClinVar's validated PMID trail (SCV000939555, Labcorp) linking PMID:18519021 to this exact genomic record.

Rule

ACMG/AMP 2015 (Richards et al., PMID:25741868) PM3: 'For recessive disorders, detected in trans with a pathogenic variant.' Applied at moderate strength (homozygous in one affected proband = in trans with itself, phase certain); SVI PM3/BS2 counting convention (homozygous observation = two in-trans observations) supports moderate, not strong, without additional probands.

Evidence considered

• PMID:18519021 - Table 1: c.G859A p.C253Y / c.G859A p.C253Y, Homozygous, n=1 (affected EOPD proband); Table 2: patient K-0010, proband, 1st parkin mutation p.C253Y / 2nd parkin mutation idem, age at onset 20 years, positive family history for PD
• clinvar - record VCV000645725 submission SCV000939555 (Labcorp) cites PMID:18519021 for this exact variant (exact genomic HGVS match, validated PMID trail), anchoring variant identity

Still missing

• Only one affected proband with the variant biallelic (homozygous) is identified in available full texts; a second affected proband with the variant in trans would support upgrading PM3 to strong
• Full text of PMID:26830385 (juvenile parkinsonism DBS case report) was not retrieved; could not verify whether it adds another biallelic observation of c.758G>A
• No compound-heterozygous (trans with a different pathogenic PRKN variant) observations with explicitly confirmed phase were found in the available sources

Sources

Reasoning

Co-segregation with disease in affected family members is documented in the GenePD study (Sun et al., PMID:16769863). Table 1, Family 3 reports 2 affected members (onset ages 28 and 37) carrying c.758G>A (p.Cys253Tyr) as a compound heterozygous genotype with Ex2-4del, i.e., both affected siblings in the family carry the variant in trans with a large pathogenic deletion. GenePD families were ascertained for >=2 PD-affected members sharing PARK2 alleles IBS, and per-family genotypes in Table 1 apply to the affected members. This meets PP1 at supporting strength (2 affected individuals in 1 family); there is insufficient segregation to upgrade (no additional families or meioses). Corroborating lead: Labcorp/Invitae ClinVar submission SCV000939555 lists a PP1 evidence sentence ('observed to segregate with disease in related individuals') citing this study; the PP1 call here is grounded independently in the GenePD Table 1 data, not in the submitter's classification. Note: Sironi et al. (PMID:18519021) also reports p.C253Y in homozygous and compound heterozygous patients but contains no segregation data for this variant (its co-segregation statement concerns the distinct p.C238W variant, Fam-012).

Rule

PP1: Co-segregation with disease in affected family members - Supporting (Richards et al. 2015, PMID:25741868; generic ACMG fallback - no PRKN VCEP/CSPEC available)

Evidence considered

• GenePD study Table 1, Family 3: 2 affected members, onset ages 28 and 37, mutations Ex2-4del + c.758G>A (p.Cys253Tyr), zygosity Compound Het
• ClinVar SCV000939555 (Labcorp) PP1 evidence signal 'it has also been observed to segregate with disease in related individuals' (used only as a lead; primary support from PMID:16769863 Table 1)

Still missing

• Segregation rests on a single family (GenePD Family 3) with 2 affected siblings; table-level genotype summary, per-individual genotyping of every affected member not explicitly enumerated in text
• No additional segregating families or informative meioses reported for this variant
• Solve-RD submission PMID (39825153) contains no PRKN/c.758G>A content and cannot add segregation support

Sources

Reasoning

PP3 (ACMG/AMP 2015, PMID:25741868) requires multiple lines of computational evidence supporting a deleterious effect on the gene or gene product. NM_004562.2:c.758G>A is a missense substitution (p.Cys253Tyr), so both the missense in-silico path and the splice-impact path were evaluated (this group owns both). Missense path: two independent calibrated meta-predictors support a deleterious effect. (1) REVEL v1.3 = 0.818, which exceeds the ClinGen SVI-calibrated PP3 supporting threshold (>= 0.644) and falls within the PP3 moderate band [0.773, 0.932) (Pejaver et al. 2022 calibration, PMID 36413997 - verified reference for REVEL in this pipeline). (2) BayesDel noAF = 0.395385, which exceeds the SVI-calibrated PP3 supporting threshold (>= 0.13) and falls within the PP3 moderate band [0.27, 0.50). Two independent lines both support a deleterious missense effect, satisfying 'multiple lines'. Splice path: SpliceAI reports max delta score 0.00 (DS_AG 0.0, DS_AL 0.0, DS_DG 0.0, DS_DL 0.0), i.e., no predicted splice impact; the splice path does not add PP3 support, consistent with a protein-level effect of a missense change. Strength is reported as supporting (the generic ACMG/AMP 2015 framework defines PP3 as supporting and no VCEP exists to adopt the SVI moderate-strength mapping; the moderate-band observation is documented for the reviewer). No gene-specific PP3/BP4 lookup table exists in this case to pre-assign a code.

Rule

ACMG/AMP 2015 PP3 (supporting): multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.) (PMID:25741868). Met via two independent SVI-calibrated missense meta-predictors (REVEL 0.818, BayesDel 0.395385); SpliceAI splice path negative (max delta 0.00). Applied at supporting strength under the generic ACMG/AMP 2015 fallback (no PRKN VCEP/CSPEC retrieved; framework_complete=false).

Evidence considered

• REVEL v1.3 score 0.818 (source_registry 'revel'; prefetch steps.revel, GRCh38 6-161785885-C-T) - exceeds SVI PP3 supporting threshold >= 0.644; within moderate band [0.773, 0.932)
• BayesDel noAF score 0.395385 (source_registry 'bayesdel'; prefetch steps.bayesdel, GRCh37 6-162206917-C-T) - exceeds SVI PP3 supporting threshold >= 0.13; within moderate band [0.27, 0.50)
• SpliceAI max delta 0.00 (DS_AG 0.0, DS_AL 0.0, DS_DG 0.0, DS_DL 0.0) - splice-impact path negative; does not contribute to PP3
• Variant consequence: missense p.Cys253Tyr (prefetch normalization) - makes the missense computational path the primary axis for PP3

Still missing

• aGVGD (owned by this group) was not retrieved in the workspace; not required for the met call because REVEL and BayesDel independently satisfy PP3
• No RNA/splice functional assay for this variant exists in the reviewed literature; PP3 here rests on in-silico evidence only

Sources

Reasoning

No de novo occurrence of c.758G>A (p.Cys253Tyr) is documented in any source in the case bundle. All 6 ClinVar submissions for VCV000645725 record germline origin except SCV005091300 (Solve-RD), which explicitly records origin 'inherited' — affirmative evidence the variant was inherited in that proband, the opposite of de novo. None of the reviewed full texts (GenePD 16769863, Sironi 18519021, Solve-RD 39825153) report parental testing or a de novo event for this variant. Additionally, PRKN early-onset Parkinson disease is autosomal recessive; the ACMG de novo criteria require a family history consistent with de novo inheritance (e.g., unaffected parents for a dominant disorder), which is not the expected mechanism here. Generic ACMG/AMP 2015 (PMID:25741868) fallback applied because no PRKN VCEP/CSPEC exists.

Rule

PS2: De novo (paternity and maternity confirmed) in a patient with the disease and no family history - Strong (Richards et al. 2015, PMID:25741868; generic ACMG fallback - no PRKN VCEP/CSPEC available)

Evidence considered

• ClinVar VCV000645725 submission audit: origins germline (5 submissions) and inherited (1, SCV005091300 Solve-RD); no submission records de novo origin
• Full-text review of PMID:16769863, PMID:18519021, PMID:39825153 found no parental testing or de novo event for c.758G>A

Still missing

• No parental (maternity/paternity) testing data available for any proband carrying this variant
• No de novo occurrence reported in ClinVar submissions or reviewed literature
• PRKN AR inheritance makes de novo occurrence an unexpected mechanism; PS2 not applicable in the classic sense

Sources

Reasoning

Generic ACMG/AMP 2015 PS4 requires the prevalence of the variant in affected individuals to be significantly increased compared with controls (OR/RR >5, CI excluding 1). No case-control study for c.758G>A (p.Cys253Tyr) exists in this case. The only available observations are case reports/series: PMID:18519021 (Sironi 2008) found 1/146 consecutive early-onset PD probands homozygous for p.C253Y (reported under older transcript numbering as c.G859A), and the variant was a previously reported pathogenic change that was NOT tested against the 50 controls (100 chromosomes) used in that study (controls were reserved for novel variants); PMID:16769863 (GenePD, Sun 2006) found the variant (reported exactly as c.758G>A p.Cys253Tyr) in compound heterozygosity with Ex2-4del in Family 3 of a familial PD cohort, a family-based design without variant-level control comparison. The ClinVar aggregate label (Pathogenic, 5 clinical labs) is not a case-control enrichment study and was not used as PS4 evidence. Population frequency is extremely low (gnomAD v2.1 AF 1.59e-05; v4.1 AF 6.81e-06, 0 homozygotes), which is consistent with rarity but does not by itself demonstrate enrichment. The gene-level statement in PMID:18519021 that parkin mutations account for 10-25% of EOPD is gene-level, not variant-level, and cannot support PS4.

Rule

ACMG/AMP 2015 PS4 (PMID:25741868) applied under generic fallback: prevalence of variant in affected individuals significantly increased vs controls; OR/RR >5 with CI excluding 1

Evidence considered

• PMID:18519021
• PMID:16769863
• clinvar
• gnomad_v2
• gnomad_v4

Still missing

• No case-control study reporting c.758G>A/p.Cys253Tyr allele counts in cases vs controls (no OR/RR or CI available)
• Variant-specific control frequency data absent (Sironi 2008 tested controls only for novel variants; p.C253Y was not among them)
• Only two independent observations (1 homozygous EOPD proband; 1 compound-heterozygous family) - insufficient to establish significant enrichment

Sources

Reasoning

PM5 (ACMG/AMP 2015) applies to a NOVEL missense change at an amino acid residue where a DIFFERENT missense change has been established pathogenic. Two prerequisites fail on the available evidence. First, the variant under evaluation, p.Cys253Tyr, is itself the established pathogenic change (reported as known pathogenic in PMID:18519021; documented in affected patients in PMID:16769863 and PMID:18519021; ClinVar Pathogenic x4 / Likely pathogenic x1 for this exact change), so the 'novel missense' prerequisite is not satisfied. Second, no different amino acid change at residue 253 (e.g., p.Cys253Phe/Trp/Arg) is documented anywhere in the bundle: the pm5_candidates artifact harvested zero same-residue candidates (found=false, total_candidates_seen=0; classic PM5 search skipped with 'Unable to confirm classic same-residue PM5 semantics safely'), and the two available parkin cohort full texts report only Cys253Tyr at codon 253. PM5 is therefore not met on the available evidence.

Rule

ACMG/AMP 2015 (Richards et al., PMID:25741868) PM5: 'Novel missense change at an amino acid residue where a different missense change determined to be pathogenic has been seen before' (moderate) — requires a different pathogenic amino-acid change at residue 253; none documented, and the variant itself is the established change (not novel).

Evidence considered

• pm5_candidates - output/pm5_candidates.json: found=false, total_candidates_seen=0, same_residue_candidates=0; classic same-residue PM5 harvesting skipped
• clinvar - VCV000645725 contains only p.Cys253Tyr at codon 253; no alternate amino acid change at the residue in the retrieved record
• PMID:18519021 - Results 3.1: p.C253Y listed among known pathogenic changes; no other amino acid change at residue 253 reported
• PMID:16769863 - Table 1: only c.758G>A (p.Cys253Tyr) at codon 253

Still missing

• No dedicated same-codon (Cys253) alternate-missense search was performed (pm5_candidates harvest skipped; hermes_instruction: review same-residue comparators manually)
• No ClinVar/PubMed evidence of a different amino acid change at residue 253 (e.g., C253F, C253W, C253R) present in the bundle

Sources

Reasoning

No assumed-de-novo evidence exists for c.758G>A. No source reports the variant as de novo without confirmed parentage, and the only explicit inheritance annotation in ClinVar (SCV005091300, Solve-RD) is 'inherited'. No parental genotype data are present in any ClinVar submission or in the reviewed publications, so the 'assumed de novo' premise cannot even be entertained. As with PS2, PRKN disease is autosomal recessive, so a de novo origin is not the expected disease mechanism. Generic ACMG/AMP 2015 fallback (PMID:25741868) governs since no PRKN VCEP exists.

Rule

PM6: Assumed de novo, but without confirmation of paternity and maternity - Moderate (Richards et al. 2015, PMID:25741868; generic ACMG fallback - no PRKN VCEP/CSPEC available)

Evidence considered

• ClinVar VCV000645725: no submission or evidence signal indicates de novo or assumed-de-novo origin; one submission (SCV005091300) records origin 'inherited'
• Full-text review of PMID:16769863, PMID:18519021, PMID:39825153: no parental testing described for this variant

Still missing

• No parental genotype data documented for any reported carrier of c.758G>A
• No reports of an assumed-de-novo occurrence in ClinVar or literature
• AR inheritance model limits applicability of PM6

Sources

Reasoning

PP5 is applied ONLY for an exact-variant ClinVar expert-panel classification (Pathogenic/Likely pathogenic) per the governing framework rule; ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions must never trigger PP5. ClinVar VCV000645725 (and VCV000903817) for NM_004562.3(PRKN):c.758G>A (p.Cys253Tyr) has 6 submissions from 5 clinical laboratories plus the Solve-RD consortium (4 Pathogenic, 1 Likely pathogenic usable, 1 non-usable) with review status 'criteria provided, single submitter' / 2 stars on the API, and the submission audit records expert_panel_submissions: 0 and expert_panels: {} (evidence.json source_data.expert_panels is empty). No ClinGen VCEP/expert-panel classification of this variant exists, so the PP5 trigger condition is absent. The aggregate Pathogenic label from clinical laboratories is explicitly excluded.

Rule

Framework-restricted PP5: exact-variant ClinVar expert-panel (3-star) Pathogenic/Likely pathogenic classification required; lab-only and aggregate assertions excluded (ACMG/AMP 2015 PP5, PMID:25741868)

Evidence considered

• clinvar

Sources

Reasoning

BA1 (standalone benign) requires an allele frequency above the population threshold for the disorder. Under the project non-VCEP convention BA1 applies at AF >1% (the original ACMG/AMP 2015 text uses >5% in ESP/1000G/ExAC; not met under either). Observed frequencies are orders of magnitude lower: gnomAD v2.1 exome total AF 1.592e-05 (0.00159%, 4/251,238 alleles), gnomAD v4.1 exome total AF 6.815e-06 (0.00068%, 11/1,614,112 alleles), 0 homozygotes in both. The highest observed subpopulation frequency is the gnomAD v2.1 NFE_BGR (Bulgarian) cluster at 0.112% (3/2,670 alleles), which is a small-sample threshold artifact not replicated in v4.1 (NFE AF 8.47e-06 across 1,179,982 alleles) and is still ~10-fold below the 1% convention. grpmax FAF (the gnomAD-provided metric designed for BA1/BS1/PM2 decisions) is 7.02e-06 (v2.1) and 4.29e-06 (v4.1). The variant is also absent from gnomAD-Canada v1.0. BA1 is not met.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID:25741868) BA1: 'Allele frequency is >5% in Exome Sequencing Project, 1000 Genomes Project, or Exome Aggregation Consortium.' Project non-VCEP convention: BA1 threshold 1%. Observed gnomAD total AF 0.0016% (v2.1) / 0.0007% (v4.1) and grpmax FAF <=7.02e-06 are far below both thresholds - not met.

Evidence considered

• gnomad_v2 - gnomAD v2.1 exome: total AF 1.592e-05 (4/251,238), 0 homozygotes; highest subpopulation AMR AF 2.893e-05 (1/34,566); NFE_BGR AF 0.112% (3/2,670); exome grpmax FAF 7.02e-06
• gnomad_v4 - gnomAD v4.1 exome: total AF 6.815e-06 (11/1,614,112), 0 homozygotes; highest subpopulation AMR AF 1.666e-05 (1/60,026); NFE AF 8.47e-06 (10/1,179,982); joint grpmax FAF 4.29e-06
• gnomad_canada - variant absent from gnomAD-Canada v1.0 (HostSeq genomes)

Still missing

• None material: combined gnomAD v2.1 + v4.1 sample sizes (251K and 1.6M alleles) provide robust AF estimates; AF is 3-4 orders of magnitude below any BA1 threshold

Sources

Reasoning

BS1 requires an allele frequency greater than expected for the disorder. PRKN-related early-onset Parkinson disease (AR-JP2, OMIM 600116) is autosomal recessive and rare (estimated prevalence ~1-2/100,000), so the maximum credible carrier allele frequency for a fully penetrant recessive disorder is on the order of 0.5-1% (sqrt of prevalence) - the project non-VCEP convention sets BS1 at AF >0.3%. Observed total AF is 0.00159% (gnomAD v2.1, 4/251,238) and 0.00068% (gnomAD v4.1, 11/1,614,112), ~200-400x below the 0.3% convention. Even the maximum observed subpopulation frequency (gnomAD v2.1 NFE_BGR 0.112%, 3/2,670 alleles) is below 0.3%, and that enrichment is a small-sample artifact that does not replicate in gnomAD v4.1 (NFE AF 8.47e-06). grpmax FAF 7.02e-06 (v2.1) / 4.29e-06 (v4.1) and absence from gnomAD-Canada further confirm the variant is far too rare for BS1. Disease prevalence and penetrance considerations (recessive, rare, with heterozygous carriers at increased but incomplete risk per GenePD literature) do not lower the BS1 threshold below the observed AF. BS1 is not met.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID:25741868) BS1: 'Allele frequency is greater than expected for disorder.' Project non-VCEP convention: BS1 threshold 0.3%. Observed total AF 0.0016%/0.0007% and max credible subpopulation AF 0.112% (artifact) are all below 0.3%; recessive-disease carrier-frequency expectation (~0.5-1%) is also above the observed AF - not met.

Evidence considered

• gnomad_v2 - gnomAD v2.1 exome: total AF 1.592e-05 (0.00159%, 4/251,238), 0 homozygotes; max subpopulation AMR 2.893e-05; NFE_BGR 0.112% (3/2,670) below 0.3% BS1 convention; grpmax FAF 7.02e-06
• gnomad_v4 - gnomAD v4.1 exome: total AF 6.815e-06 (0.00068%, 11/1,614,112), 0 homozygotes; NFE AF 8.47e-06 (10/1,179,982); joint grpmax FAF 4.29e-06
• gnomad_canada - variant absent from gnomAD-Canada v1.0

Still missing

• No population-specific disease prevalence estimate for AR-JP2 was available in the case bundle; using a conservative ~1/100,000 recessive prevalence, the expected carrier AF threshold (~0.5-1%) remains far above the observed AF, so the not_met call is robust to this gap

Sources

Reasoning

BS2 requires the variant to be observed in a healthy adult individual in a state consistent with the disorder's inheritance - for this autosomal recessive disorder, a homozygous (biallelic) observation in an unaffected individual. gnomAD v2.1 and v4.1 both report 0 homozygotes for this variant (4/251,238 and 11/1,614,112 alleles respectively, all heterozygous), and it is absent from gnomAD-Canada. The literature observations of this variant are exclusively in AFFECTED individuals: a homozygous affected early-onset PD proband (K-0010) in Sironi et al. 2008 (PMID:18519021, Table 1) and an affected compound-heterozygous family (c.758G>A with Ex3del, 2 affected members, onset ages 34 and 32) in the GenePD study (PMID:16769863, Table 1, Family 4). No healthy adult homozygous or biallelic carrier of c.758G>A has been reported in any population or control cohort in the available sources. BS2 is not met.

Rule

Generic ACMG/AMP 2015 (Richards et al., PMID:25741868) BS2: 'Observed in a healthy adult individual for a recessive disorder in homozygous state, or in a dominant disorder in heterozygous state.' No homozygous (or biallelic) observation in a healthy individual exists: gnomAD homozygote count is 0 in both v2.1 and v4.1, and all literature observations are in affected patients - not met.

Evidence considered

• gnomad_v2 - 0 homozygotes (4/251,238 alleles)
• gnomad_v4 - 0 homozygotes (11/1,614,112 alleles)
• PMID:18519021 - p.C253Y homozygous in an affected EOPD proband (K-0010); all observations are in affected individuals, not healthy controls
• PMID:16769863 - Table 1, Family 4: c.758G>A (p.Cys253Tyr) compound heterozygous with Ex3del in affected family members (onset ages 34 and 32); no healthy homozygous carriers

Still missing

• No dedicated healthy-control cohort genotyping for this specific variant was found in the reviewed sources (in PMID:18519021 only the two novel variants and p.A230T were screened against 100 control chromosomes; p.C253Y was previously reported and not control-screened there); gnomAD homozygote count of 0 across 251K+1.6M alleles is the primary evidence and is sufficient for not_met

Sources

Reasoning

BP2 (ACMG/AMP 2015) requires either (a) observation in trans with a pathogenic variant for a fully dominant disease/mechanism, or (b) observation in cis with a pathogenic variant in any inheritance pattern. PRKN-associated AR-JP2 is autosomal recessive, not a fully dominant mechanism, so clause (a) is inapplicable. No cis-phase evidence exists for c.758G>A: the only phase-defining observation (PMID:18519021, homozygous affected proband K-0010) demonstrates the variant is in trans (biallelic), which is the opposite of the cis configuration BP2 addresses. The heterozygous-carrier earlier-onset association in the GenePD study (PMID:16769863) is a gene-level/heterozygosity-risk observation that does not name this variant and provides no cis/trans data; it cannot support BP2. BP2 is therefore not met.

Rule

ACMG/AMP 2015 (Richards et al., PMID:25741868) BP2: 'Observed in trans with a pathogenic variant for a fully dominant disease/mechanism; or observed in cis with a pathogenic variant in any inheritance pattern.' Neither clause is satisfied: disease is autosomal recessive and the observed phase is trans, not cis.

Evidence considered

• PMID:18519021 - homozygous observation (in trans) of the variant in an affected proband; no cis observations reported
• PMID:16769863 - GenePD study: parkin heterozygosity influences onset age; exact variant not named; no cis/trans data

Still missing

• No familial segregation data with phased parental alleles for c.758G>A were available to formally exclude cis configuration in additional carriers
• Parental genotyping/phase data for the homozygous proband (K-0010) were not reported in PMID:18519021

Sources

Reasoning

BP4 (ACMG/AMP 2015, PMID:25741868) requires multiple lines of computational evidence suggesting no impact on the gene or gene product. No benign-direction computational line exists for NM_004562.2:c.758G>A. The two calibrated missense meta-predictors both predict a deleterious effect (REVEL 0.818, BayesDel 0.395385) - the opposite direction from BP4. The only no-impact signal is SpliceAI (max delta 0.00), which addresses solely the splicing axis; its four delta sub-scores are outputs of one prediction and cannot be counted as separate lines (the ACMG 2015 text itself cautions that the combination of predictions from different in-silico tools counts as a single piece of evidence). For a missense variant the protein-level consequence is the relevant axis, and both missense predictors contradict 'no impact on gene or gene product'. No conservation score (phyloP/GERP/phastCons) was retrieved in the workspace to provide an additional benign-direction line. BP4 is therefore not met.

Rule

ACMG/AMP 2015 BP4 (supporting): multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.) (PMID:25741868). Not met: no benign-direction line; both missense predictors support a deleterious effect and the single SpliceAI no-impact prediction is insufficient and is not double-counted.

Evidence considered

• REVEL 0.818 and BayesDel 0.395385 both predict a deleterious missense effect - contradict BP4's no-impact requirement
• SpliceAI max delta 0.00 - single predictor addressing only the splicing axis; four sub-scores are outputs of one prediction, not independent lines
• No conservation score (phyloP/GERP/phastCons) retrieved in workspace artifacts - no additional benign-direction line available

Still missing

• A benign-direction computational line (e.g., low REVEL/BayesDel scores or a conservation score) would be required to reconsider BP4; none is present in the workspace

Sources

Reasoning

BP6 is applied ONLY for an exact-variant ClinVar expert-panel classification (Benign/Likely benign) per the governing framework rule; ordinary laboratory submissions, aggregate ClinVar labels, and non-expert assertions must never trigger BP6. The ClinVar submission audit for this variant records expert_panel_submissions: 0 and expert_panels: {} (evidence.json), and no submission asserts Benign/Likely benign (all 6 submissions are Pathogenic or Likely pathogenic). The BP6 trigger condition is absent; the criterion is not met.

Rule

Framework-restricted BP6: exact-variant ClinVar expert-panel (3-star) Benign/Likely benign classification required; lab-only and aggregate assertions excluded (ACMG/AMP 2015 BP6, PMID:25741868)

Evidence considered

• clinvar

Sources

Reasoning

PS1 (ACMG/AMP 2015) requires a DIFFERENT nucleotide change producing the same amino acid change (p.Cys253Tyr) that is already established pathogenic. The case bundle documents only the change under evaluation, NM_004562.2:c.758G>A (p.Cys253Tyr) itself: ClinVar VCV000645725 (variant title 'NM_004562.3(PRKN):c.758G>A (p.Cys253Tyr)'), PMID:16769863 (GenePD, Table 1 Family 4: c.758G>A / p.Cys253Tyr), and PMID:18519021 (homozygous p.C253Y patient). No alternate nucleotide substitution at c.758 (e.g., c.758G>T or c.758G>C producing the same p.Cys253Tyr) is documented in any retrieved source, and no PS1-specific comparator search was executed by the pipeline (prefetch ClinVar queries returned only the exact variant). In addition, ClinVar candidate variation ID 903817 was listed in prefetch but its content was never retrieved or described in the bundle, so it could not be evaluated as a potential same-amino-acid alternate-nucleotide record. Because meeting PS1 requires positive documentation of such an alternate-nucleotide change and none was searched for or found, PS1 is left not assessed rather than affirmatively excluded.

Rule

ACMG/AMP 2015 (Richards et al., PMID:25741868) PS1: 'Same amino acid change... a different nucleotide change that results in the same amino acid [e.g., c.34G>C (p.Val12Leu) and c.34G>T (p.Val12Leu)] can also be assumed to be pathogenic' — requires documented alternate nucleotide change producing p.Cys253Tyr; none available in the bundle.

Evidence considered

• clinvar - VCV000645725 documents only NM_004562.3(PRKN):c.758G>A (p.Cys253Tyr); no alternate nucleotide change at this position retrieved in the record
• PMID:16769863 - c.758G>A (p.Cys253Tyr) is the only change at codon 253 reported in the GenePD cohort (Table 1 Family 4)
• PMID:18519021 - p.C253Y is the only amino acid change at residue 253 reported; no alternate-nucleotide/alternate-AA variant at the codon
• PMID:25741868 - PS1 definition consulted (see paper_findings)

Still missing

• No dedicated search for alternate nucleotide changes at c.758 producing p.Cys253Tyr (e.g., c.758G>T, c.758G>C) was run against ClinVar/PubMed in this case
• ClinVar variation ID 903817 listed as candidate variation in prefetch but uncharacterized in the bundle; could not be evaluated as a potential PS1 comparator

Sources

Reasoning

Generic ACMG/AMP 2015 (PMID:25741868) governs PS3 because no ClinGen CSPEC/VCEP specification exists for PRKN (cspec lookup failed; vcep_materials empty; final_classification_framework fell back to generic_acmg, precedence rank 3). PS3 requires well-established functional studies showing a deleterious effect, with assay validity judged on biological relevance, validation, reproducibility, and controls. No variant-specific functional assay for p.Cys253Tyr was identified in any reviewed source: (1) OncoKB explicitly reports no variant-specific reviewed functional evidence for PRKN C253Y (is_variant_specific=false, not_reviewed=true); (2) PMID:18519021 (Sironi 2008) reports one homozygous early-onset PD patient with p.C253Y, listing it among 'previously reported... known as pathogenic changes', but the study performs no functional assay - its in vitro solubility/cysteine-substitution discussion concerns the distinct residue Cys238, not Cys253; (3) PMID:16769863 (Sun 2006, GenePD) reports c.758G>A (p.Cys253Tyr) in compound heterozygosity with an exon 2-4 deletion in a familial PD patient - a clinical genetics observation only. In silico predictors (REVEL 0.818, BayesDel 0.395, SpliceAI max delta 0.00) are computational predictions, not functional studies, and per ACMG 2015 cannot satisfy PS3 (they are evaluated under PP3/BP4 by another group). p.Cys253Tyr lies within the RING1 zinc-binding domain of the parkin TRIAD module, which PMID:18519021 describes as the critical zone for E3 ubiquitin ligase activity, but domain location alone is not PS3-grade evidence (domain context is PM1 territory). Evidence is therefore insufficient to assess PS3: the functional impact of p.Cys253Tyr is uncharacterized by any validated assay in the reviewed corpus.

Rule

ACMG/AMP 2015 (PMID:25741868) PS3: well-established functional studies showing a deleterious effect (strong; strength calibrated by assay validation, biological relevance, and controls). Applied under generic ACMG fallback because no PRKN ClinGen CSPEC/VCEP exists.

Evidence considered

• PMID:25741868
• PMID:18519021
• PMID:16769863
• oncokb

Still missing

• No validated parkin functional assay data for p.Cys253Tyr in the evidence bundle (e.g., E3 ubiquitin ligase/ubiquitination activity assays, parkin protein stability/solubility assays, mitophagy or mitochondrial-translocation assays, mRNA/RNA assays). A well-established variant-specific functional study showing a deleterious effect would be required to meet PS3.
• Only in silico predictions available (REVEL 0.818, BayesDel 0.395, SpliceAI 0.00); these are explicitly excluded from PS3/BS3 by ACMG 2015.

Sources

Reasoning

PP2 (ACMG/AMP 2015) has two limbs: (a) the gene has a low rate of benign missense variation, and (b) missense variants are a common mechanism of disease. Limb (b) is partially supported for PRKN: multiple missense changes are established pathogenic (p.R42P, p.M192L, p.T240M, p.C253Y, p.R275W, p.R402C described as 'known pathogenic changes' in PMID:18519021; the c.758G>A missense is found in trans with an exon-3 deletion in an affected individual, PMID:16769863 Family 4). Limb (a) cannot be evaluated from the bundle: no gene-level missense constraint metric (e.g., gnomAD missense Z-score or observed/expected missense ratio) or benign-missense-rate data for PRKN are present. Additionally, PRKN-associated disease is autosomal recessive with a strong loss-of-function mechanism (exon rearrangements and truncations are frequent; pvs1_gene_context: lof_mechanism_supported=true), and PP2's applicability to recessive, LOF-dominant genes is a recognized judgment call that no ClinGen CSPEC/VCEP guidance in this case (vcep_materials empty) resolves. PP2 is therefore left not assessed, with the missing constraint data identified as the gap.

Rule

ACMG/AMP 2015 (Richards et al., 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' (supporting). Limb (b) partially supported; limb (a) unevaluable — no constraint data in bundle.

Evidence considered

• PMID:18519021 - Results 3.1: six established pathogenic missense changes listed (incl. p.C253Y) — missense is a documented disease mechanism in PRKN
• PMID:16769863 - Table 1 Family 4: c.758G>A (p.Cys253Tyr) missense in trans with exon-3 deletion in affected individuals — missense contributes to biallelic disease
• pvs1_gene_context - output/pvs1_gene_context.json: lof_mechanism_supported=true (germline PRKN loss of function is a disease mechanism) — context for LOF-dominant recessive gene

Still missing

• No gene-level missense constraint (gnomAD missense Z-score / observed-expected ratio) or benign-missense-rate data for PRKN available in the case bundle
• PP2 applicability to an autosomal-recessive, LOF-dominant gene unresolved — no ClinGen CSPEC/VCEP guidance retrieved (vcep_materials.json empty; generic_acmg fallback)

Sources

Reasoning

Generic ACMG/AMP 2015 PP4 requires the proband's phenotype or family history to be highly specific for a disease with a single genetic etiology (PRKN-related early-onset/autosomal recessive juvenile Parkinson disease). The consolidated case-evidence bundle contains no proband-level clinical phenotype data (no clinical/phenotype JSON is included among review_manifest.json, case_summary.json, source_registry.json, prefetch.json, evidence.json, literature_pass.json, vcep_materials.json, fulltext_map.json). Without the proband's phenotype, family history, onset age, or inheritance pattern, PP4 cannot be assessed. The gene-disease association (PRKN biallelic loss-of-function causing early-onset PD) is documented in literature, but phenotype specificity must be evaluated against the actual case.

Rule

ACMG/AMP 2015 PP4 (PMID:25741868) applied under generic fallback: phenotype or family history highly specific for a disease with a single genetic etiology

Still missing

• No proband phenotype/clinical data (onset age, motor signs, levodopa response, family history, inheritance) in the evidence bundle
• Cannot confirm the case phenotype matches PRKN-related early-onset Parkinson disease with high specificity

Reasoning

Generic ACMG/AMP 2015 (PMID:25741868) governs BS3 (no PRKN ClinGen CSPEC/VCEP available). BS3 requires well-established functional studies demonstrating NO deleterious effect (preserved/normal function). No functional study of p.Cys253Tyr exists in the reviewed corpus: OncoKB reports no variant-specific functional evidence; PMID:18519021 and PMID:16769863 report the variant in affected patients (homozygous and compound heterozygous, respectively) without any functional assay; the variant is present at very low frequency in gnomAD (v2.1 AF 1.59e-05, 0 homozygotes; v4.1 AF 6.81e-06, 0 homozygotes), which is population evidence for other groups (PM2/BS1/BA1) and is not functional evidence. Absence of functional characterization cannot be used as benign evidence. Evidence is therefore insufficient to assess BS3: no assay demonstrates that p.Cys253Tyr preserves parkin function.

Rule

ACMG/AMP 2015 (PMID:25741868) BS3: well-established functional studies showing no deleterious effect (strong). Applied under generic ACMG fallback because no PRKN ClinGen CSPEC/VCEP exists.

Evidence considered

• PMID:25741868
• PMID:18519021
• PMID:16769863
• oncokb

Still missing

• No functional study demonstrating preserved/normal parkin function for p.Cys253Tyr (e.g., E3 ubiquitin ligase activity, protein stability, or mitophagy assays showing wild-type-like activity). Such a validated assay would be required to meet BS3.
• No RNA/splicing functional data exists, though SpliceAI predicts no splice impact (max delta 0.00) - an in silico prediction that cannot substitute for a functional assay.

Sources

Reasoning

No non-segregation evidence is available for c.758G>A. No source reports an affected family member who lacks the variant, and no genotype data exist for unaffected or affected relatives beyond the GenePD Family 3 observation (which shows co-segregation, not non-segregation). The GenePD study does describe an affected sibling without the family's mutations (Family 8), but that family carries different variants (exon 1/12 duplications), not c.758G>A, so it is not relevant to this variant. BS4 requires positive evidence of non-segregation with disease in affected family members; with no such data, the criterion cannot be assessed. Generic ACMG/AMP 2015 (PMID:25741868) fallback governs.

Rule

BS4: Non-segregation with disease in affected family members - Strong (benign) (Richards et al. 2015, PMID:25741868; generic ACMG fallback - no PRKN VCEP/CSPEC available)

Evidence considered

• GenePD study (PMID:16769863): Family 8 affected sibling lacked the family mutations (exon 1/12 duplications) - not this variant, so not informative for BS4; no non-segregation data for c.758G>A anywhere in the case bundle

Still missing

• No affected family members reported to lack c.758G>A (no non-segregation observations)
• No genotype data for additional family members beyond GenePD Family 3
• Targeted segregation testing of family members would be needed to assess BS4

Sources

Reasoning

Generic ACMG/AMP 2015 BP5 requires the variant to be found in a case with an alternate molecular basis for disease (e.g., a second, clearly causative genetic or non-genetic etiology identified in the proband). The consolidated case-evidence bundle contains no proband-level clinical or molecular data (no case/phenotype JSON, no alternate-diagnosis workup), so it cannot be determined whether an alternate molecular basis exists. BP5 cannot be assessed from the available evidence.

Rule

ACMG/AMP 2015 BP5 (PMID:25741868) applied under generic fallback: variant found in a case with an alternate molecular basis for disease

Still missing

• No proband-level data on presence or absence of an alternate molecular basis for disease
• No case report/case summary with diagnostic workup in the evidence bundle