PRPF8 c.4639-17T>G, p.?

Reasoning

BA1 is met under the generic ACMG/AMP framework: the well-powered gnomAD v4.1 African/African American group has AF 14.9547% (11,219/75,020 alleles), with 875 homozygotes and grpmax FAF 14.7231%, far exceeding the generic BA1 >5% allele-frequency threshold. The same marked ancestry-specific frequency and numerous homozygotes are reproduced in gnomAD v2.1 and gnomAD-Canada, arguing against a sampling or threshold artifact.

Rule

Generic ACMG/AMP BA1: population allele frequency >5%; no PRPF8 VCEP specification was available.

Evidence considered

• gnomAD v4.1: total AF 1.17106% (18,903/1,614,178 alleles), African/African American AF 14.9547% (11,219/75,020), 875 homozygotes, and grpmax FAF 14.7231%.
• gnomAD v2.1 independently reports African/African American AF 14.6743% (3,663/24,962), 272 homozygotes, and grpmax FAF 14.4349%.
• gnomAD-Canada reports African AF 15.7171% (160/1,018), 11 homozygotes, and FAF95 13.7306%.
• The generic ACMG/AMP BA1 rule from Richards et al. (PMID:25741868) applies when allele frequency is greater than 5%. No PRPF8-specific ClinGen specification was retrieved for this case.

Sources

Reasoning

PM2 is not met because the variant is common in multiple population datasets, including AF 14.9547% and FAF 14.7231% in the gnomAD v4.1 African/African American group.

Rule

Generic ACMG/AMP PM2: absent from controls or present at an extremely low frequency; not satisfied.

Evidence considered

• gnomAD v4.1 reports 18,903 variant alleles among 1,614,178 total alleles (AF 1.17106%) and African/African American AF 14.9547% (11,219/75,020); gnomAD v2.1 and gnomAD-Canada show consistent high frequencies.
• A variant observed at these frequencies is neither absent nor rare in population databases.

Sources

Reasoning

NM_006445.3:c.4639-17T>G is an intronic variant (not missense, not synonymous), so the applicable computational path is splice-impact prediction rather than missense in-silico or aGVGD. No official ClinGen VCEP (no CSPEC found; framework_mode=generic_acmg) or gene-specific PP3/BP4 lookup spreadsheet was available for PRPF8 in this case's materials, so generic ACMG/SpliceAI-based assessment applies. SpliceAI predicts max delta score = 0.135 (DS_AG=0.135, DS_AL=0.006, DS_DG=0.003, DS_DL=0.002), which falls in the intermediate/inconclusive zone (0.1-0.2) per the ClinGen SVI Splicing Subgroup recommended thresholds, below the >=0.2 threshold generally used to support a splice-altering prediction. Therefore PP3 is not met on computational evidence alone. REVEL and BayesDel scores are not applicable/not available (variant is not missense; both returned null/not found in prefetch), so the missense in-silico sub-path does not apply to this variant.

Rule

Generic ACMG PP3 (computational splice-impact prediction); SpliceAI max delta 0.135 does not clear the >=0.2 supporting threshold

Evidence considered

• SpliceAI Lookup prediction for NM_006445.3:c.4639-17T>G: max delta score 0.135 (DS_AG=0.135, DS_AL=0.006, DS_DG=0.003, DS_DL=0.002), interpreted by the pipeline as 'no significant splice impact'.
• No gene-specific CSPEC/VCEP splicing evidence-code lookup spreadsheet was found for PRPF8 in vcep_materials.json (files_found empty, gene_vcep_dir null).
• REVEL score not found/null and BayesDel score not found/null in prefetch.json; not applicable because the variant is intronic, not missense.
• General SpliceAI delta-score calibration bands (>=0.2 supporting splice-altering / <0.1 supporting no splice effect / 0.1-0.2 inconclusive) reflect the widely used ClinGen SVI Splicing Subgroup recommendations (Walker LC et al., 2023, PMID 37652022); this publication is not part of the case's source_registry so it is cited here as background calibration knowledge only, not as an evidence_refs key.

Still missing

• SpliceAI max delta (0.135) falls in the inconclusive 0.1-0.2 band, not clearly meeting or excluding a splice-altering prediction
• No gene-specific (PRPF8) VCEP PP3/BP4 splicing threshold or lookup table is available in this case's vcep_materials to override the generic threshold
• No RNA/splicing functional assay evidence considered here (that is PS3/BS3 territory, outside this group's scope)

Sources

Reasoning

The exact-variant ClinVar audit identified no pathogenic or likely pathogenic expert-panel assertion; therefore PP5 cannot be applied.

Rule

Apply PP5 only for an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification.

Evidence considered

• The exact-match ClinVar submission audit reports zero expert-panel submissions. The available classified submissions are benign single-submitter assertions, which cannot support PP5.

Sources

Reasoning

Using the same SpliceAI evidence as PP3: max delta score = 0.135 falls in the intermediate/inconclusive band (0.1-0.2), above the <0.1 threshold generally used to support a benign (no splice effect) prediction. Therefore BP4 is also not met - the score is not low enough to confidently predict no splicing impact. No gene-specific PRPF8 VCEP PP3/BP4 spreadsheet was found to override this generic assessment. Missense in-silico predictors (REVEL/BayesDel) are not applicable since the variant is intronic, not missense.

Rule

Generic ACMG BP4 (computational splice-impact prediction); SpliceAI max delta 0.135 does not clear the <0.1 benign-supporting threshold

Evidence considered

• SpliceAI Lookup prediction for NM_006445.3:c.4639-17T>G: max delta score 0.135, in the intermediate/inconclusive band relative to the <0.1 threshold typically used for BP4-supporting evidence.
• No gene-specific CSPEC/VCEP splicing evidence-code lookup spreadsheet was found for PRPF8 in vcep_materials.json.
• REVEL and BayesDel scores not available/not applicable (intronic variant, not missense).
• General SpliceAI delta-score calibration bands reflect the ClinGen SVI Splicing Subgroup recommendations (Walker LC et al., 2023, PMID 37652022), cited here as background calibration knowledge only since it is not in this case's source_registry.

Still missing

• SpliceAI max delta (0.135) is above the <0.1 benign-supporting cutoff but below the 0.2 pathogenic-supporting cutoff, leaving the call indeterminate rather than clearly benign
• No gene-specific PRPF8 VCEP splicing evidence-code lookup available to resolve the intermediate score

Sources

Reasoning

The exact-variant ClinVar audit identified no benign or likely benign expert-panel assertion; therefore BP6 cannot be applied.

Rule

Apply BP6 only for an exact-variant ClinVar expert-panel Benign or Likely benign classification.

Evidence considered

• The exact-match ClinVar submission audit reports zero expert-panel submissions. The aggregate benign label and benign single-submitter assertions are not expert-panel assertions and cannot trigger BP6.

Sources

Reasoning

No proband-level phenotype or parental testing data are present to establish a confirmed de novo occurrence.

Rule

Generic ACMG/AMP: PS2 requires a confirmed de novo occurrence in an affected individual with an appropriate phenotype and confirmed maternity and paternity.

Evidence considered

• No applicable PRPF8 VCEP/CSpec was retrieved; the generic ACMG/AMP framework applies.
• The consolidated case-evidence bundle contains no individual-level proband, parental genotype, parentage, or de novo-testing information.

Still missing

• Obtain proband phenotype and a laboratory report confirming the variant in the proband.
• Obtain parental genotypes with confirmed biological parentage; assess parental phenotype where relevant.

Reasoning

No functional assay data (e.g., RT-PCR/minigene splicing assay, cDNA analysis, or other validated well-established functional study) for NM_006445.3:c.4639-17T>G was found anywhere in the case-evidence bundle. The gene has no applicable VCEP/CSPEC framework (framework_mode=generic_acmg, cspec.found=false), so generic ACMG/AMP PS3 criteria apply, which require a validated functional study demonstrating a damaging effect on protein/splicing function. Only an in-silico SpliceAI prediction is present (max delta score 0.14, predicting no significant splice impact), which is not functional assay evidence and cannot support PS3 (that belongs to PP3/BP4 evaluation). No publication in the bundle (PMID:25741868 general ACMG guidelines, PMID:31275557 pan-cancer cryptic splicing mutation repository, PMID:28492532 Sherloc methodology) reports a functional/RNA study of this specific variant.

Evidence considered

• SpliceAI predicts no significant splice impact for this variant (max delta score = 0.14) per case_summary.json/prefetch.json spliceai section — an in-silico prediction, not functional assay evidence, and thus not usable for PS3.
• review_manifest.json / case_summary.json confirm framework_mode is generic_acmg with cspec.found=false, so no PRPF8-specific VCEP PS3 assay-strength calibration table is available.
• literature_pass.json high_priority_papers list only PMID:25741868 (ACMG/AMP guidelines paper, no variant-specific functional data) and PMID:31275557 (pan-cancer cryptic splicing repository abstract, no confirmation this variant is in the validated dataset).

Still missing

• No RT-PCR, minigene, cDNA/RNA-seq, or other validated functional splicing assay for this variant was located in the bundle or literature pass.
• PMID:31275557 abstract concerns a pan-cancer repository of validated splicing mutations but no evidence in the bundle confirms this specific variant is included in that dataset.

Reasoning

No case-control or statistically evaluated case-series evidence for enrichment of NM_006445.3:c.4639-17T>G in a PRPF8-related phenotype was retrieved.

Rule

Generic ACMG/AMP PS4: prevalence in affected individuals must be significantly increased compared with controls.

Evidence considered

• No retrieved literature item was identified as containing case-control, affected-cohort, or enrichment evidence for this exact variant.
• The available population records establish that the variant is observed in population datasets but do not constitute affected-versus-control enrichment evidence.

Still missing

• An ancestry-matched case-control study or a well-ascertained affected cohort reporting this exact variant, with denominator data and a statistical enrichment analysis, is required.

Reasoning

No affected individual with this PRPF8 variant and a pathogenic/likely pathogenic PRPF8 variant in trans is documented in the consolidated evidence bundle. Therefore, PM3 cannot be assessed.

Rule

Generic ACMG/AMP 2015 PM3: apply only when the variant is observed in trans with a pathogenic variant in an affected individual with a recessive disorder.

Evidence considered

• The case bundle identifies no PRPF8 CSPEC or VCEP material and specifies the generic ACMG fallback.
• Generic ACMG/AMP guidance states that, for a recessive disorder, a heterozygous variant in trans with a known pathogenic variant can provide moderate pathogenic evidence (PM3).

Still missing

• Affected-proband genotype data identifying a second PRPF8 pathogenic/likely pathogenic variant are not available.
• Phase data establishing that this variant is in trans with such a variant are not available.
• No applicable PRPF8 VCEP/gene-specific specification was retrieved; generic ACMG/AMP guidance was used.

Sources

Reasoning

No reported de novo occurrence is available, including an occurrence lacking confirmed parental identity.

Rule

Generic ACMG/AMP: PM6 requires a de novo occurrence in an affected individual with an appropriate phenotype when maternity and paternity are not confirmed.

Evidence considered

• No applicable PRPF8 VCEP/CSpec was retrieved; the generic ACMG/AMP framework applies.
• The consolidated case-evidence bundle contains no individual-level proband, parental genotype, parentage, or de novo-testing information.

Still missing

• Obtain proband and parental genotypes or a documented de novo report.
• If parentage cannot be confirmed, document the affected proband phenotype and the testing method.

Reasoning

No pedigree, affected relatives, genotype-phenotype co-segregation data, or informative meioses are available.

Rule

Generic ACMG/AMP: PP1 requires co-segregation of the variant with disease in multiple affected family members; strength depends on the amount and quality of segregation evidence.

Evidence considered

• No applicable PRPF8 VCEP/CSpec was retrieved; the generic ACMG/AMP framework applies.
• The consolidated case-evidence bundle contains no family-level segregation observations or meiosis counts.

Still missing

• Obtain a pedigree with phenotype status and variant testing for informative affected and unaffected relatives.
• Document the number of informative meioses and any phenocopy, age-dependent penetrance, or variable-expression considerations.

Reasoning

No individual-level phenotype, diagnostic work-up, or evidence that the observed phenotype is highly specific for a PRPF8-related disorder was retrieved for this exact variant.

Rule

Generic ACMG/AMP PP4: the patient's phenotype or family history must be highly specific for a disease with a single genetic etiology.

Evidence considered

• The exact-variant ClinVar record contains broad condition labels and not-provided entries, without an individual-level phenotype supporting phenotype specificity.

Still missing

• A clinically characterized individual carrying the exact variant and a demonstration that the phenotype is highly specific for a PRPF8-related disorder are required.

Sources

Reasoning

Although many homozygotes are observed in population datasets, BS2 requires observation in healthy adult individuals for a fully penetrant early-onset disorder. The available population records do not provide individual phenotype, age, or PRPF8-associated disease penetrance data; therefore, homozygote counts alone are insufficient for BS2.

Rule

Generic ACMG/AMP BS2: requires unaffected adult observation in the context of fully penetrant early-onset disease; population homozygote counts without phenotype are insufficient.

Evidence considered

• Homozygotes are present in gnomAD v4.1 (1,029 total; 875 African/African American), gnomAD v2.1 (339 total; 272 African/African American), and gnomAD-Canada (17 total; 11 African), but these datasets do not establish that the individuals are unaffected adults.
• The generic ACMG/AMP BS2 concept in Richards et al. (PMID:25741868) requires observation in healthy adults for a disorder expected to be fully penetrant early in life.

Still missing

• Phenotyped adult homozygote data and disease-specific age-of-onset and penetrance information are not available in the case evidence bundle.

Sources

Reasoning

No functional assay data demonstrating normal/no-impact splicing or protein function for NM_006445.3:c.4639-17T>G was found in the bundle. As with PS3, only an in-silico SpliceAI prediction (max delta score 0.14, predicting no significant splice impact) is available; SpliceAI is a computational predictor, not a validated functional assay, so it cannot be used to satisfy BS3 under generic ACMG/AMP or any PRPF8-specific framework (none exists here). No literature-pass publication documents a well-established in vitro/in vivo functional study (e.g., minigene splicing assay, cDNA analysis) of this exact variant showing no damaging effect.

Evidence considered

• SpliceAI predicts no significant splice impact for this variant (max delta score = 0.14) per case_summary.json/prefetch.json — computational, not a functional assay, so not usable for BS3.
• No functional/RNA study publications specific to this variant were identified in literature_pass.json or source_registry.json.

Still missing

• No validated splicing/functional assay (RT-PCR, minigene, RNA-seq of patient tissue) results for this variant were located.
• SpliceAI in-silico score alone cannot substitute for BS3 functional-assay evidence.

Reasoning

No informative family data demonstrate lack of segregation between the variant and disease.

Rule

Generic ACMG/AMP: BS4 requires lack of segregation between the variant and disease in affected family members, with appropriate consideration of phenocopies and penetrance.

Evidence considered

• No applicable PRPF8 VCEP/CSpec was retrieved; the generic ACMG/AMP framework applies.
• The consolidated case-evidence bundle contains no family-level genotype-phenotype data demonstrating non-segregation.

Still missing

• Obtain a pedigree with phenotype status and variant testing for informative relatives, including affected non-carriers or unaffected carriers where clinically interpretable.
• Assess phenocopy, penetrance, age of onset, and variable expressivity before using non-segregation evidence.

Reasoning

No case-level observation establishes this variant in cis with a pathogenic/likely pathogenic PRPF8 variant in a recessive disease context, or in trans with a pathogenic/likely pathogenic PRPF8 variant in a dominant disease context. Therefore, BP2 cannot be assessed.

Rule

Generic ACMG/AMP 2015 BP2: apply only to a documented cis/trans observation with a pathogenic variant in the appropriate disease-inheritance context.

Evidence considered

• The case bundle identifies no PRPF8 CSPEC or VCEP material and specifies the generic ACMG fallback.
• Generic ACMG/AMP guidance describes a second variant in cis in a recessive gene as supporting benign evidence (BP2), and a variant in trans with a pathogenic variant in a dominant disorder as supporting benign evidence (BP2).

Still missing

• Case-level co-occurring PRPF8 variant data and phase are not available.
• The inheritance model and phenotype of any affected proband carrying a relevant co-occurring PRPF8 variant are not available.
• No applicable PRPF8 VCEP/gene-specific specification was retrieved; generic ACMG/AMP guidance was used.

Sources

Reasoning

No report was retrieved of an individual carrying this exact variant who has a phenotype fully explained by an alternative molecular diagnosis.

Rule

Generic ACMG/AMP BP5: a variant is found in a case with an alternate molecular basis for disease.

Evidence considered

• No retrieved evidence provides an individual-level alternative molecular diagnosis for a carrier of this exact variant.

Still missing

• An individual-level report documenting the exact variant, the relevant phenotype, and an alternative molecular diagnosis that fully explains that phenotype is required.