PRPF8 c.4792G>A, p.(Asp1598Asn)

Reasoning

Generic ACMG/AMP fallback applies because no PRPF8-specific VCEP specification was available. PMID:24781015 identified recurrent PRPF8 D1598 mutations and tested the homologous human D1598N-equivalent Prp8-D1670N substitution in a Saccharomyces cerevisiae ACT1-CUP1 splicing reporter. The D1670N substitution altered spliceosomal function by improving splicing of an intron reporter defective in the second step of splicing, while the comparator D1670H did not show the suppressor effect. This is a biologically relevant, controlled functional assay supporting an abnormal effect for the exact amino-acid substitution, but it is a heterologous homolog assay rather than direct testing of the patient's nucleotide change in a human system; therefore PS3 is limited to supporting strength.

Rule

Generic ACMG/AMP PS3: well-established functional studies supportive of a damaging effect can meet PS3; strength is limited here because the evidence is from a homologous yeast assay rather than direct human-cell testing.

Evidence considered

• PMID:24781015 reports recurrent PRPF8 mutations involving D1598, with a D1598 mutation found in four patient samples.
• In the homologous Saccharomyces cerevisiae Prp8 model, D1670N corresponds to human PRPF8 D1598N and improved splicing of an intron mutant defective in the second step of splicing.
• The same assay showed no suppressor effect for the D1670H comparator, providing an internal amino-acid-specific control.
• The assay used an ACT1-CUP1 copper-resistance splicing reporter, providing a biologically relevant readout of spliceosomal function.

Still missing

• No direct human-cell assay testing PRPF8 c.4792G>A/p.Asp1598Asn was identified.
• The extracted report does not provide quantitative assay results or a complete description of all assay controls.

Sources

Reasoning

No governing PRPF8-specific VCEP framework was available, so the supplied generic PM2 threshold of AF <=0.0001 applies. The variant is rare but observed in gnomAD v4.1 at AF 2.29251e-05 overall (37/1,613,954 alleles; 0 homozygotes), with a maximum subpopulation AF of 6.75493e-05, and in gnomAD v2.1 at AF 3.18137e-05 overall (8/251,464 alleles; 0 homozygotes). The non-cancer subsets are also below threshold: gnomAD v2.1 non-cancer exomes AF 3.37655e-05 and gnomAD v3.1 non-cancer genomes AF 3.37984e-05. These frequencies meet PM2 at supporting strength.

Rule

Generic ACMG/AMP PM2: absence or very low frequency in population databases, operationalized here as AF <=0.0001, supports PM2 at supporting strength; no applicable PRPF8-specific VCEP specification was available.

Evidence considered

• gnomAD v4.1 reports AF 2.29251e-05 overall (37/1,613,954 alleles; 0 homozygotes), maximum subpopulation AF 6.75493e-05, and grpmax FAF 1.995e-05.
• gnomAD v2.1 reports AF 3.18137e-05 overall (8/251,464 alleles; 0 homozygotes) and grpmax FAF 2.293e-05.
• The non-cancer subsets report AF 3.37655e-05 in gnomAD v2.1 non-cancer exomes and AF 3.37984e-05 in gnomAD v3.1 non-cancer genomes, both with 0 homozygotes.
• The supplied generic PM2 threshold is AF <=0.0001 at supporting strength, based on the ClinGen SVI recommendation identified with PMID:25741868; all reported frequencies are below that threshold.

Sources

Reasoning

No applicable PRPF8-specific VCEP framework was identified, so generic ACMG/AMP PS1 was considered. The variant is NM_006445.3:c.4792G>A, p.Asp1598Asn. PMID:24781015 reports recurrent PRPF8 mutation at D1598 and a yeast assay of the homologous D1670N substitution, corresponding to human D1598N, but it does not establish a pathogenic alternative-nucleotide variant producing p.Asp1598Asn. The available report therefore does not satisfy PS1's requirement for an established pathogenic same-amino-acid change caused by a different nucleotide substitution.

Rule

Generic ACMG/AMP PS1: same amino-acid change as a previously established pathogenic variant caused by a different nucleotide change.

Evidence considered

• PMID:24781015 states that mutation at PRPF8 D1598 was found in four patient samples and evaluates homologous yeast substitutions D1670Y and D1670N, corresponding to human D1598Y/N.
• The paper does not classify a different-nucleotide p.Asp1598Asn variant as pathogenic, so the generic PS1 requirement is not fulfilled.

Still missing

• An established pathogenic PRPF8 variant from a different nucleotide substitution that produces p.Asp1598Asn is not documented.

Sources

Reasoning

No applicable PRPF8-specific VCEP framework was identified, so generic ACMG/AMP PM5 was considered. PMID:24781015 reports substitutions at D1598, including the D1598N-equivalent yeast assay and D1598Y-equivalent substitution, but does not establish an alternate missense variant at residue 1598 as pathogenic for the relevant germline disorder. Recurrent somatic occurrence and an experimental effect alone do not meet PM5's requirement for a previously established pathogenic missense change at the same residue.

Rule

Generic ACMG/AMP PM5: novel missense change at a residue where a different missense change has previously been established as pathogenic.

Evidence considered

• PMID:24781015 reports recurrent mutation at D1598 and functional testing of homologous D1670Y and D1670N substitutions.
• The publication does not provide an established pathogenic classification for an alternate PRPF8 missense variant at residue 1598 in the relevant germline disease context.

Still missing

• A previously established pathogenic alternate missense variant at PRPF8 residue 1598 is not documented.

Sources

Reasoning

The variant is a missense substitution, NM_006445.3:c.4792G>A (p.Asp1598Asn), so PP3 is evaluated using REVEL only. The REVEL score is 0.638, below the generic ClinGen SVI supporting threshold of >=0.644 from the supplied Pejaver et al. 2022 calibration (PMID:36413997); therefore PP3 is not met. SpliceAI is not used for this missense PP3 assessment, and BayesDel has no applicable generic strength calibration.

Rule

Missense variant: apply REVEL only; PP3 supporting requires REVEL >=0.644 under the supplied generic calibration.

Evidence considered

• The variant is a missense substitution, p.Asp1598Asn, so the missense REVEL path is the sole applicable computational path.
• REVEL score: 0.638; the supplied ClinGen SVI supporting PP3 threshold is >=0.644 (Pejaver et al. 2022, PMID:36413997).
• SpliceAI was not used because splice predictions cannot support PP3 for a missense variant; BayesDel was not used because no generic ACMG-strength calibration is established.

Sources

Reasoning

PP5 is not met because the exact ClinVar record for NM_006445.3:c.4792G>A (p.Asp1598Asn) has two laboratory submissions, both classified as Uncertain significance, with zero expert-panel submissions. The available assertions are ordinary laboratory submissions rather than an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification. The cited literature does not replace the required expert-panel assertion.

Evidence considered

• The exact ClinVar variation record has two submissions, both Uncertain significance, review status criteria provided by single submitters, and expert_panel_submissions equal to 0.
• Under the requested rule, ordinary laboratory submissions, aggregate labels, and non-expert assertions cannot trigger PP5; no exact-variant expert-panel Pathogenic or Likely pathogenic classification is present.

Sources

Reasoning

No governing PRPF8-specific VCEP framework was available, so the generic ACMG/AMP BA1 threshold of allele frequency >=0.05 (Richards et al. 2015, PMID:25741868) applies. The variant's highest reported all-comers population frequency is 6.75493e-05 in gnomAD v4.1 Ashkenazi Jewish participants, while the overall gnomAD v4.1 frequency is 2.29251e-05 (37/1,613,954 alleles); both are far below 0.05. Therefore BA1 is not met.

Rule

Generic ACMG/AMP BA1: allele frequency >=0.05 is stand-alone benign evidence; no applicable PRPF8-specific VCEP specification was available.

Evidence considered

• gnomAD v4.1 reports an overall AF of 2.29251e-05 (37/1,613,954 alleles; 0 homozygotes) and a highest subpopulation AF of 6.75493e-05 in Ashkenazi Jewish participants.
• gnomAD v2.1 reports an overall AF of 3.18137e-05 (8/251,464 alleles; 0 homozygotes) and a highest subpopulation AF of 5.43656e-05 in East Asian participants.
• The generic ACMG/AMP BA1 threshold is AF >=0.05, from Richards et al. 2015 (PMID:25741868), and neither observed frequency approaches it.

Sources

Reasoning

No governing PRPF8-specific VCEP framework or gene-specific benign frequency threshold was available, so the supplied generic BS1 threshold of AF >=0.01 applies. The highest observed frequency is 6.75493e-05 in gnomAD v4.1 Ashkenazi Jewish participants, and the overall gnomAD v4.1 AF is 2.29251e-05; both are below 0.01. The gnomAD v2.1 overall AF is likewise only 3.18137e-05. Therefore BS1 is not met.

Rule

Generic ACMG/AMP BS1: allele frequency >=0.01 is strong benign evidence; no applicable PRPF8-specific VCEP specification was available.

Evidence considered

• gnomAD v4.1 reports an overall AF of 2.29251e-05 and a highest subpopulation AF of 6.75493e-05 in Ashkenazi Jewish participants, with 0 homozygotes.
• gnomAD v2.1 reports an overall AF of 3.18137e-05 and a highest subpopulation AF of 5.43656e-05 in East Asian participants, with 0 homozygotes.
• The supplied generic BS1 threshold is AF >=0.01; no PRPF8-specific population threshold was available.

Sources

Reasoning

No governing PRPF8-specific VCEP framework was available. The available population data show zero observed homozygotes in gnomAD v4.1 (0 among 1,613,954 alleles), gnomAD v2.1 (0 among 251,464 alleles), gnomAD v2.1 non-cancer exomes (0 among 236,928 alleles), and gnomAD v3.1 non-cancer genomes (0 among 147,936 alleles). Because BS2 requires observation of the variant in healthy adults at a frequency inconsistent with the disease model, absence of homozygotes does not establish BS2; the criterion is not met.

Rule

Generic ACMG/AMP BS2: observation of the variant in healthy adults, including homozygotes where relevant to the disease model, supports benign evidence; no such homozygote observation is present.

Evidence considered

• gnomAD v4.1 reports 0 homozygotes with 37 observed alleles among 1,613,954 total alleles.
• gnomAD v2.1 reports 0 homozygotes with 8 observed alleles among 251,464 total alleles.
• The available non-cancer subsets also report 0 homozygotes: gnomAD v2.1 non-cancer exomes, 8/236,928 alleles; gnomAD v3.1 non-cancer genomes, 5/147,936 alleles.
• BS2 requires positive observation of the variant in healthy adults, typically including homozygotes for a dominant disease model; zero homozygotes is absence of supporting evidence rather than fulfillment of BS2.

Sources

Reasoning

The variant is a missense substitution, NM_006445.3:c.4792G>A (p.Asp1598Asn), so BP4 is evaluated using REVEL only. The REVEL score is 0.638, above the generic ClinGen SVI supporting threshold of <=0.29 from the supplied Pejaver et al. 2022 calibration (PMID:36413997); therefore BP4 is not met. SpliceAI is not used for this missense BP4 assessment, and BayesDel has no applicable generic strength calibration.

Rule

Missense variant: apply REVEL only; BP4 supporting requires REVEL <=0.29 under the supplied generic calibration.

Evidence considered

• The variant is a missense substitution, p.Asp1598Asn, so the missense REVEL path is the sole applicable computational path.
• REVEL score: 0.638; the supplied ClinGen SVI supporting BP4 threshold is <=0.29 (Pejaver et al. 2022, PMID:36413997).
• SpliceAI was not used because splice predictions cannot support BP4 for a missense variant; BayesDel was not used because no generic ACMG-strength calibration is established.

Sources

Reasoning

BP6 is not met because the exact ClinVar record for NM_006445.3:c.4792G>A (p.Asp1598Asn) has no expert-panel submission. The two available laboratory submissions classify the variant as Uncertain significance, not Benign or Likely benign. No exact-variant ClinVar expert-panel Benign or Likely benign assertion is present.

Evidence considered

• The exact ClinVar variation record reports two laboratory submissions, both Uncertain significance, and expert_panel_submissions equal to 0.
• Under the requested rule, ordinary laboratory submissions and non-expert assertions cannot trigger BP6; no exact-variant expert-panel Benign or Likely benign classification is present.

Sources

Reasoning

No case-level evidence documents that NM_006445.3:c.4792G>A (p.Asp1598Asn) arose de novo in a confirmed proband, and no parental genotypes or maternity/paternity information are provided. The relevant published report identifies recurrent PRPF8 D1598 mutations in patient samples but does not provide individual-level de novo testing for this variant.

Rule

Generic ACMG/AMP PS2 requires a confirmed de novo variant in an affected individual with confirmed maternity and paternity; stronger application requires additional de novo observations or supporting evidence.

Evidence considered

• The available case evidence contains no proband-level pedigree, parental genotype, trio testing, or validated de novo statement for NM_006445.3:c.4792G>A.
• PMID:24781015 reports PRPF8 D1598 mutations in a patient cohort but explicitly provides no individual patient-level de novo or segregation evidence for p.D1598N.

Still missing

• Confirmed proband phenotype and molecular diagnosis
• Maternal and paternal testing results for c.4792G>A
• Maternity and paternity confirmation
• Assessment for parental mosaicism

Sources

Reasoning

No qualifying germline case-control enrichment analysis or statistically significant excess of this exact variant in affected individuals versus controls was identified. PMID:24781015 reports recurrent somatic PRPF8 D1598 mutations in a myeloid-neoplasm cohort, including four patient samples, but it does not establish germline disease prevalence, an appropriate control comparison, or variant-specific clinical enrichment for NM_006445.3:c.4792G>A (p.Asp1598Asn).

Evidence considered

• PMID:24781015 reports that a PRPF8 mutation at D1598 was found in four patient samples in a 447-patient myeloid-neoplasm cohort, but the source is somatic and provides no germline case-control enrichment analysis for p.D1598N.
• Because no applicable PRPF8 VCEP specification was retrieved, generic ACMG/AMP interpretation applies; generic PS4 requires phenotype prevalence significantly increased in affected individuals compared with controls, which cannot be evaluated from the available evidence.

Still missing

• A qualifying germline affected-case and control comparison for the exact variant is unavailable.
• No odds ratio, likelihood ratio, confidence interval, or other governing PS4 threshold comparison is available.

Sources

Reasoning

No applicable PRPF8-specific VCEP framework or domain table was identified. Generic PM1 requires a well-established critical functional domain or hotspot in which pathogenic missense variants cluster and benign variation is uncommon. The available evidence documents no approved critical domain encompassing residue 1598, and the hotspot source reports no statistically significant hotspot for PRPF8 D1598. Without an authoritative domain definition or equivalent validated hotspot evidence, PM1 cannot be assessed reliably.

Rule

Generic ACMG/AMP PM1: variant located in a mutational hotspot and/or critical, well-established functional domain without benign variation.

Evidence considered

• vcep_materials.json contains no PRPF8 framework and its domain_tables object is empty.
• The hotspot assessment reports no statistically significant hotspot for PRPF8 D1598.

Still missing

• An authoritative PRPF8 domain table or validated critical-domain annotation covering residue 1598 is unavailable.
• No validated hotspot evidence supporting PM1 is available.

Reasoning

No affected proband with a confirmed recessive phenotype, second pathogenic or likely pathogenic allele, or validated trans phase was documented for NM_006445.3:c.4792G>A (p.Asp1598Asn). The available PRPF8 publication reports recurrent D1598 mutations in somatic myeloid malignancy samples but provides no germline proband-level allelic phase or compound-heterozygous evidence, so PM3 cannot be assessed.

Evidence considered

• No source in the reviewed case evidence documents this variant in an affected germline proband with a second pathogenic or likely pathogenic variant in trans.
• PMID:24781015 reports D1598 mutations in somatic myeloid malignancy samples and does not provide germline compound-heterozygous, phase, or proband-level evidence for c.4792G>A/p.Asp1598Asn.

Still missing

• Affected-proband clinical observations linked to this variant are unavailable.
• A pathogenic or likely pathogenic second allele and its phase relative to this variant are unavailable.
• The inheritance mode relevant to this individual is unavailable.

Reasoning

No case-level or publication-level evidence documents NM_006445.3:c.4792G>A as apparently de novo in an affected individual without parental testing. The available D1598 report does not establish an apparently de novo origin for this specific variant.

Rule

Generic ACMG/AMP PM6 may be considered for an apparently de novo variant in an affected individual when parental testing is unavailable, but the phenotype and variant-specific observation must be documented.

Evidence considered

• No available evidence identifies this exact variant as apparently de novo in an affected individual.
• PMID:24781015 reports recurrent D1598 mutations and functional testing but does not report apparent de novo status for c.4792G>A/p.D1598N.

Still missing

• A published or clinical report of the variant in an affected proband
• Explicit parental testing status or statement that testing was unavailable
• Confirmed phenotype consistent with PRPF8-related disease

Sources

Reasoning

No affected relatives, unaffected relatives, family pedigree, informative meioses, or variant-genotype/phenotype transmission data are available for NM_006445.3:c.4792G>A. Therefore, there is no basis to evaluate cosegregation or assign PP1 at supporting, moderate, or strong strength.

Rule

Generic ACMG/AMP PP1 requires cosegregation of the variant with disease in affected family members; strength depends on the number and informativeness of meioses and the phenotype assessment.

Evidence considered

• The available case evidence contains no family pedigree, affected-relative testing, unaffected-relative testing, or segregation counts for this variant.
• The relevant D1598 publication describes a patient cohort and functional assay, not family-based cosegregation or informative meioses.

Still missing

• Pedigree with clinically characterized relatives
• Genotypes of affected and unaffected relatives
• Number of informative meioses
• Phenotype consistency and penetrance assessment

Sources

Reasoning

Generic PP2 requires gene-level evidence that missense variation is a common disease mechanism and that benign missense variation is uncommon. The case materials provide no validated PRPF8 missense constraint or gene-specific benign-versus-pathogenic missense distribution sufficient for this criterion. The presence of PRPF8 loss-of-function literature does not by itself establish the PP2 missense-pattern requirement.

Rule

Generic ACMG/AMP PP2: missense variation is a common mechanism of disease in the gene and benign missense variation is uncommon.

Evidence considered

• The case materials contain no applicable PRPF8-specific missense constraint or calibrated missense mechanism analysis.
• PRPF8 loss-of-function context alone is insufficient to apply PP2.

Still missing

• Gene-specific evidence demonstrating a low rate of benign PRPF8 missense variation and a high rate of pathogenic missense variation is unavailable.

Reasoning

The available evidence does not provide a patient-level phenotype for the individual carrying NM_006445.3:c.4792G>A, nor does it establish that the variant is associated with a highly specific phenotype or a well-defined disease spectrum. The ClinVar record lists retinitis pigmentosa as a condition but contains only uncertain-significance laboratory submissions and no qualifying case phenotype description.

Evidence considered

• The exact ClinVar record lists retinitis pigmentosa among condition labels, but both submissions classify the variant as uncertain significance and provide no patient-level phenotype details.
• No retrieved source establishes a highly specific phenotype that would support PP4.

Still missing

• Patient-level clinical findings and phenotype specificity are unavailable.
• A well-established, highly specific phenotype-to-disease correlation for this exact assessment is not documented.

Sources

Reasoning

No validated functional study demonstrating that PRPF8 c.4792G>A/p.Asp1598Asn has normal function was identified. The available D1598N-equivalent yeast assay reports an altered spliceosomal phenotype rather than preserved normal function, so it cannot support BS3.

Rule

Generic ACMG/AMP BS3: well-established functional studies showing no damaging effect can support BS3; no such result is available for this variant.

Evidence considered

• PMID:24781015 describes altered spliceosomal function for the D1598N-equivalent substitution, not a normal-function result.
• No source in the reviewed functional evidence provides a validated benign assay result for this variant.

Still missing

• A well-controlled functional assay showing normal PRPF8-D1598N activity relative to wild type is unavailable.

Sources

Reasoning

No clinically characterized relatives carrying or lacking NM_006445.3:c.4792G>A are reported, so absence of segregation with disease cannot be demonstrated. Population frequency and functional findings do not constitute BS4 non-segregation evidence.

Rule

Generic ACMG/AMP BS4 requires well-documented affected relatives who do not carry the variant, with consideration of phenocopies, reduced penetrance, age dependence, and phenotype specificity.

Evidence considered

• No family-level genotype and phenotype data show that affected relatives do not carry the variant.
• The available population and functional records cannot establish non-segregation within a family.

Still missing

• Affected relatives with reliable clinical phenotyping
• Variant testing results in affected relatives
• Assessment of phenocopies and age-dependent penetrance

Sources

Reasoning

Generic BP1 requires a gene in which primarily truncating variants cause disease and missense variants are not an established disease mechanism. Although the case materials contain literature supporting PRPF8 loss-of-function disease biology, they do not establish that PRPF8 disease is primarily truncating or that missense variation is generally nonpathogenic. The available recurrent D1598 missense observations further prevent a reliable BP1 conclusion from the supplied evidence.

Rule

Generic ACMG/AMP BP1: missense variants are not a common mechanism of disease in a gene where truncating variants are the established disease mechanism.

Evidence considered

• The case materials describe PRPF8 loss-of-function biology but provide no validated missense-exclusion framework for BP1.
• PMID:24781015 reports recurrent PRPF8 missense mutations at D1598, so the supplied evidence does not support treating missense variation as categorically nonpathogenic.

Still missing

• A validated gene-level determination that PRPF8 disease is primarily caused by truncating variants and not missense variants is unavailable.

Reasoning

No individual or family-level observation was documented showing NM_006445.3:c.4792G>A (p.Asp1598Asn) in cis with a pathogenic or likely pathogenic variant, or in trans with a pathogenic variant in a dominant disorder context where BP2 would apply. Population presence and a somatic recurrence report do not establish allelic phase or benignity, so BP2 cannot be assessed.

Evidence considered

• No source in the reviewed case evidence documents this variant in cis with a pathogenic or likely pathogenic allele or in trans with a pathogenic allele in an appropriate clinical context.
• The available population observations and somatic cancer report do not provide allelic phase and therefore cannot establish BP2.

Still missing

• Allele-level phasing with any pathogenic or likely pathogenic variant is unavailable.
• Family or proband genotype data sufficient to evaluate BP2 are unavailable.
• The applicable PRPF8 inheritance model for this observation is unavailable.

Reasoning

No qualifying evidence shows that this variant is present in an affected individual whose phenotype is better explained by a pathogenic variant in another gene or by another molecular diagnosis. The available PRPF8 D1598 literature concerns recurrent somatic mutations and functional assays, not an alternate molecular diagnosis for a germline affected individual. Therefore BP5 cannot be assessed from the retrieved evidence.

Evidence considered

• PMID:24781015 describes somatic PRPF8 mutations in myeloid malignancies and a homologous yeast assay, but does not report an affected germline individual with an alternate molecular diagnosis.
• No other retrieved source supplies the alternate-cause evidence required for BP5.

Still missing

• An affected individual with this exact variant and a confirmed alternate molecular cause has not been documented.
• No applicable gene-specific BP5 rule or numeric likelihood-ratio threshold is available.

Sources