ZRSR2 c.983_984insGAA, p.(Phe328delinsLeuAsn)

Reasoning

The variant is absent from the queried gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0 population datasets, supporting PM2 at supporting strength under the generic ACMG/AMP fallback.

Rule

Generic ACMG/AMP PM2: absent from controls in population databases; applied at supporting strength.

Evidence considered

• gnomAD v2.1 queried X-15840899-T-TGAA and found it absent.
• gnomAD v4.1 and gnomAD-Canada v1.0 queried X-15822776-T-TGAA and found it absent.
• No ZRSR2 ClinGen VCEP or gene-specific population-frequency specification was retrieved for this case; the case framework specifies generic ACMG/AMP fallback.

Still missing

• No gene-specific PM2 specification was available; the generic ACMG/AMP framework was used.

Sources

Reasoning

NM_005089.3:c.983_984insGAA is a 3-nucleotide insertion normalized by both Mutalyzer and VariantValidator to an in-frame consequence predicting NP_005080.1:p.(Phe328delinsLeuAsn) (Phe328 deleted, Leu-Asn inserted; net protein length change of +1 residue, 483 to 484 aa). The variant lies in exon 11, the terminal exon of NM_005089.3, so no NMD applies and the protein-level change is expressed. SpliceAI max delta score 0.05 (<0.1) predicts no significant splice impact, consistent with the protein-level in-frame consequence being operative. The variant is NOT in a repeat region: reference protein sequence around p.328 (...CPRGKHCNFLHVFRN...) corresponds to the C-terminal portion of the conserved CCCH-type zinc finger of ZRSR2, while the SR/RS dipeptide repeat region lies C-terminal (approximately p.355 onward). An in-frame insertion causing a protein length change in a non-repeat region meets PM4 at default moderate strength under the generic ACMG/AMP 2015 framework (Richards et al., PMID 25741868), which governs because no ZRSR2 CSPEC/VCEP exists.

Rule

PM4 (Richards et al. 2015, PMID 25741868): Protein length changes as a result of in-frame deletions/insertions in a non-repeat region or stop-loss variants. Default strength: moderate.

Evidence considered

• Variant normalization (Mutalyzer and VariantValidator, recorded in prefetch/case_summary): NM_005089.3:c.983_984insGAA is an in-frame 3-nucleotide insertion predicting NP_005080.1:p.(Phe328delinsLeuAsn), i.e., Phe328 replaced by Leu-Asn with a net protein length change of +1 residue (reference 483 aa to predicted 484 aa).
• Exon context: the variant maps to exon 11, the terminal exon of NM_005089.3 (VariantValidator variant_exonic_positions), so the in-frame transcript is not subject to NMD and the predicted protein-level change is expressed.
• SpliceAI Lookup reports max delta score 0.05 for this variant, predicting no significant splice impact; this supports the in-frame protein length change as the operative consequence rather than aberrant splicing.
• Protein sequence context from the NP_005080.1 reference: the region around p.328 is ...CPRGKHCNFLHVFRN..., the C-terminal portion of the conserved CCCH-type zinc finger; the SR/RS dipeptide repeat region of ZRSR2 is C-terminal (approximately p.355 onward), so p.328 is in a non-repeat, functionally relevant domain.
• Governing framework: no ZRSR2 CSPEC/VCEP was found (vcep_materials: 'no cspec url available'); the generic ACMG/AMP 2015 rules (Richards et al., PMID 25741868, as recorded in output/generic_acmg_classification_rules.md) were applied. PM4 is defined as protein length changes from in-frame deletions/insertions in a non-repeat region or stop-loss variants.
• pvs1_variant_assessment independently records consequence_class 'in_frame_indel_unknown' and variant_bucket 'other' (not a default null-variant bucket), consistent with an in-frame, non-truncating protein-level change.

Still missing

• No curated domain annotation (UniProt/Pfam) was retrieved in this case; the zinc-finger location of p.328 is inferred from the reference protein sequence (conserved CCCH motif CPRGKHCNFLH).
• No variant-specific functional or clinical literature exists (literature_pass found 0 PMIDs), so PM4 rests on sequence-level consequence evidence only.

Sources

Reasoning

No ZRSR2 VCEP CSPEC or gene-specific PP3/BP4 lookup spreadsheet is available (cspec.found=false; vcep_materials/final_classification_framework indicate generic_acmg_fallback with no gene-specific rule payload), so generic ACMG/AMP applies. SpliceAI predicts no meaningful splice impact for NM_005089.3:c.983_984insGAA: max delta score = 0.05 (DS_AG 0.05, DS_AL 0.016, DS_DG 0.0, DS_DL 0.008), well below the 0.2 high-recall threshold established by the SpliceAI developers (Jaganathan et al. 2019, Cell, PMID 30661751) for a splice-altering effect. This variant is a small in-frame insertion/delins (p.(Phe328delinsLeuAsn)) rather than a missense SNV, so REVEL/BayesDel are not applicable (both explicitly skipped per prefetch.json as 'not snv or no coords') and cannot contribute additional computational evidence in either direction. The single calibrated computational tool that is applicable here (SpliceAI) is concordant in predicting no deleterious splicing effect, supporting BP4 at supporting strength.

Rule

generic_acmg_BP4_computational_no_gene_specific_cspec

Evidence considered

• SpliceAI evidence_sentence: 'SpliceAI predicts no significant splice impact for this variant (max delta score = 0.05).' (source: spliceai)
• SpliceAI raw scores: DS_AG=0.05, DS_AL=0.016, DS_DG=0.0, DS_DL=0.008, max_delta_score=0.05, well below the 0.2 pathogenic-supporting cutoff (source: spliceai)
• Jaganathan KK et al., 'Predicting Splicing from Primary Sequence with Deep Learning', Cell 2019, PMID 30661751 -- source of the 0.2/0.5/0.8 SpliceAI delta-score calibration thresholds used to interpret the max delta of 0.05 as non-significant
• prefetch.json: revel/bayesdel steps skipped ('not snv or no coords') confirming missense predictors are not applicable to this in-frame insertion, so BP4 rests solely on the SpliceAI negative prediction

Still missing

• No gene-specific ZRSR2 VCEP threshold override available; relying on the generic SpliceAI developer-calibrated cutoff
• No missense in-silico predictor (REVEL/BayesDel) evidence exists for this indel to independently corroborate a benign computational signal

Sources

Reasoning

No ClinGen VCEP CSPEC or gene-specific PP3/BP4 lookup material is available for ZRSR2 (cspec.found=false, vcep_materials empty, final_classification_framework source=generic_acmg_fallback), so generic ACMG/AMP computational-evidence assessment applies. This variant, NM_005089.3:c.983_984insGAA (NP_005080.1:p.(Phe328delinsLeuAsn)), is a 3-bp in-frame insertion/delins, not a missense SNV, so REVEL and BayesDel were not computed (prefetch.revel/bayesdel both report skipped: 'not snv or no coords') and the missense in-silico path is not applicable here. The only calibrated computational tool available for this variant is SpliceAI, and it predicts no meaningful splice disruption: max delta score = 0.05 (DS_AG 0.05, DS_AL 0.016, DS_DG 0.0, DS_DL 0.008), well below the SpliceAI high-recall pathogenic-supporting cutoff of 0.2 established in the tool's original validation (Jaganathan et al. 2019, Cell, PMID 30661751). There is no computational signal supporting a deleterious effect by either the splice or missense in-silico path, so PP3 is not met.

Rule

generic_acmg_PP3_computational_no_gene_specific_cspec

Evidence considered

• SpliceAI evidence_sentence: 'SpliceAI predicts no significant splice impact for this variant (max delta score = 0.05).' (source: spliceai)
• SpliceAI raw scores: DS_AG=0.05, DS_AL=0.016, DS_DG=0.0, DS_DL=0.008, max_delta_score=0.05 (source: spliceai)
• SpliceAI original calibration cutoffs (0.2 high-recall / 0.5 recommended / 0.8 high-precision) from Jaganathan KK et al., 'Predicting Splicing from Primary Sequence with Deep Learning', Cell 2019, PMID 30661751; observed max delta of 0.05 falls below even the lowest (0.2) pathogenic-supporting cutoff
• prefetch.json: revel and bayesdel steps both marked skipped, reason 'not snv or no coords', confirming this in-frame insertion is outside the scope of missense in-silico predictors

Still missing

• No gene-specific ZRSR2 VCEP PP3/BP4 lookup table available to override generic SpliceAI threshold interpretation
• REVEL/BayesDel not applicable/not computed for this in-frame insertion, so no orthogonal missense predictor signal exists

Sources

Reasoning

PP5 is not met because the exact variant is absent from ClinVar; consequently, there is no exact-variant Pathogenic/Likely pathogenic ClinVar expert-panel assertion.

Rule

Apply PP5 only for an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification; ordinary laboratory submissions, aggregate labels, and non-expert assertions do not qualify.

Evidence considered

• ClinVar exact-variant search returned no result for NM_005089.3:c.983_984insGAA.

Sources

Reasoning

The variant is absent from the queried gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0 datasets; therefore it does not meet the stand-alone benign population-frequency criterion.

Rule

Generic ACMG/AMP BA1: allele frequency is sufficiently high for a stand-alone benign classification.

Evidence considered

• No ZRSR2 c.983_984insGAA / X-15840899-T-TGAA (GRCh37) or X-15822776-T-TGAA (GRCh38) record was returned by the queried gnomAD datasets.

Sources

Reasoning

The variant is absent from the queried population datasets and consequently has no allele frequency exceeding that expected for a pathogenic ZRSR2-associated condition.

Rule

Generic ACMG/AMP BS1: allele frequency is greater than expected for the disorder.

Evidence considered

• gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0 each reported the queried allele as absent.

Still missing

• No ZRSR2-specific BS1 threshold or disease prevalence/penetrance calibration was available; however, this does not affect the conclusion because the variant is absent.

Sources

Reasoning

BP3 requires an in-frame indel located in a repeat region without known function. The variant p.(Phe328delinsLeuAsn) is not in a repeat region: the local reference sequence (...CPRGKHCNFLHVFRN...) is the C-terminal portion of the conserved CCCH-type zinc finger of ZRSR2, a functional RNA-binding domain, and shows no tandem or low-complexity repeat structure. The SR/RS dipeptide repeat-rich region of ZRSR2 lies C-terminal (approximately p.355 onward) and does not include p.328. Because the variant is an in-frame insertion in a non-repeat, functionally relevant domain rather than a repeat without known function, BP3 is not met.

Rule

BP3 (Richards et al. 2015, PMID 25741868): In-frame indels in a repeat without known function. Default strength: supporting (benign).

Evidence considered

• Governing framework: no ZRSR2 CSPEC/VCEP exists; generic ACMG/AMP 2015 rules (Richards et al., PMID 25741868, as recorded in output/generic_acmg_classification_rules.md) define BP3 as in-frame indels in a repeat without known function.
• Protein sequence context from the NP_005080.1 reference: the sequence around p.328 (...CPRGKHCNFLHVFRN...) belongs to the C-terminal portion of the conserved CCCH-type zinc finger and is not a tandem/low-complexity repeat; the SR/RS dipeptide repeat region of ZRSR2 begins C-terminal (approximately p.355 onward).
• Variant normalization (Mutalyzer/VariantValidator) confirms the variant is an in-frame 3-nucleotide insertion (p.(Phe328delinsLeuAsn)) rather than a repeat-tract duplication, and pvs1_variant_assessment records consequence_class 'in_frame_indel_unknown' in exon 11.

Still missing

• Repeat-region status is based on inspection of the reference protein sequence; no repeat-masking/annotation tool output (e.g., UniProt feature table or RepeatMasker) was present in the evidence bundle.

Sources

Reasoning

BP6 is not met because the exact variant is absent from ClinVar; consequently, there is no exact-variant Benign/Likely benign ClinVar expert-panel assertion.

Rule

Apply BP6 only for an exact-variant ClinVar expert-panel Benign or Likely benign classification; ordinary laboratory submissions, aggregate labels, and non-expert assertions do not qualify.

Evidence considered

• ClinVar exact-variant search returned no result for NM_005089.3:c.983_984insGAA.

Sources

Reasoning

No proband phenotype or parental testing data are provided; de novo occurrence cannot be established.

Rule

Generic ACMG/AMP fallback: PS2 requires a confirmed de novo occurrence in an affected proband with compatible phenotype; the required case and parental evidence is absent.

Evidence considered

• No ZRSR2 CSPEC/VCEP specification was retrieved; the case is designated for generic ACMG/AMP fallback.
• The consolidated evidence bundle contains no proband clinical information, parental results, pedigree, or de novo report for this variant.

Still missing

• Proband phenotype and diagnostic indication.
• Maternal and paternal genotypes for NM_005089.3:c.983_984insGAA, including confirmation of biological parentage and testing method.
• Assessment of parental mosaicism if the variant is absent from parental blood testing.

Sources

Reasoning

No ClinGen VCEP framework exists for ZRSR2 (cspec.found=false; framework_mode=generic_acmg). No functional/validated assay data (minigene splicing assay, RNA-seq splicing analysis, colony-forming assay, or other in vitro/in vivo functional study) for this specific variant (NM_005089.3:c.983_984insGAA / p.(Phe328delinsLeuAsn)) was identified in the case bundle. Literature search for germline ZRSR2 disease association (pvs1_gene_context) returned only gene-level/mechanism papers (aplastic anemia case report, oral-facial-digital syndrome paper, minor spliceosome review, MDS pathogenesis review, unrelated NF1 case) with no variant-specific functional assay results reported for this insertion/delins. OncoKB explicitly notes no variant-specific reviewed functional evidence. SpliceAI predicts no significant splice impact (max delta 0.05), which is a predictive in silico tool, not a validated functional assay, and is out of scope for PS3/BS3.

Still missing

• No validated functional assay (splicing minigene, RNA-seq, colony assay, or other) evaluating this specific variant was found in the available literature or VCEP materials.
• No ClinGen ZRSR2 VCEP-specific PS3/BS3 assay calibration exists to apply.

Reasoning

No affected-case series, case-control data, or statistically evaluable enrichment evidence for this exact variant was retrieved.

Rule

Generic ACMG/AMP PS4: prevalence in affected individuals must be significantly increased compared with controls; no ZRSR2-specific VCEP specification was available.

Evidence considered

• No variant-specific literature citations or phenotype/case data were present in the consolidated evidence bundle.

Still missing

• Variant-specific affected and control allele counts or a published case-control enrichment analysis are required.

Reasoning

No applicable ZRSR2 VCEP/gene-specific specification was retrieved, so generic ACMG/AMP criteria govern. The consolidated case evidence contains no affected-proband observation with this variant and a pathogenic/likely pathogenic variant at another locus, no phase result, and no evidence of a recessive disease context. PM3 cannot be assessed.

Rule

Generic ACMG/AMP PM3; no qualifying in-trans observation is available.

Evidence considered

• The governing-material records state that no ZRSR2 CSPEC/VCEP or local gene-specific framework was available and identify generic ACMG/AMP fallback.
• The case bundle supplies variant normalization and database observations but no proband-level genotype, second-allele, phase, segregation, or family data relevant to PM3.

Still missing

• Affected-proband genotype(s) containing NM_005089.3:c.983_984insGAA and a confirmed pathogenic/likely pathogenic variant in the relevant gene.
• Validated phase evidence demonstrating that the two variants are in trans.
• A defined ZRSR2 disease association and inheritance mode supporting use of recessive allelic evidence.

Sources

Reasoning

No proband phenotype or parental testing data are provided; an assumed de novo occurrence cannot be established.

Rule

Generic ACMG/AMP fallback: PM6 requires an affected proband with a de novo occurrence assumed because parentage is not confirmed; no such occurrence is documented.

Evidence considered

• No ZRSR2 CSPEC/VCEP specification was retrieved; the case is designated for generic ACMG/AMP fallback.
• The consolidated evidence bundle contains no parental testing result or report of a de novo occurrence for this variant.

Still missing

• Proband phenotype and diagnostic indication.
• Maternal and paternal genotypes for NM_005089.3:c.983_984insGAA.
• Whether any reported de novo occurrence lacks parentage confirmation.

Sources

Reasoning

No pedigree, affected relatives, or relative genotypes are provided; cosegregation cannot be evaluated.

Rule

Generic ACMG/AMP fallback: PP1 requires evidence that the variant cosegregates with disease in multiple affected family members; no segregation evidence is available.

Evidence considered

• No ZRSR2 CSPEC/VCEP specification was retrieved; the case is designated for generic ACMG/AMP fallback.
• The consolidated evidence bundle provides no family segregation data, pedigree, or meiosis count for this variant.

Still missing

• Pedigree with phenotype status for informative relatives.
• Variant genotypes for affected and unaffected informative relatives.
• Number of informative meioses and penetrance/phenocopy considerations for the relevant ZRSR2-associated phenotype.

Sources

Reasoning

The case supplies no proband phenotype, disease indication, or phenotypic testing results. Therefore, phenotype specificity for a ZRSR2-associated disorder cannot be assessed.

Rule

Generic ACMG/AMP PP4: the patient's phenotype or family history must be highly specific for a disease with a single genetic etiology; no ZRSR2-specific VCEP specification was available.

Evidence considered

• No variant-specific literature citations or phenotype data were present in the consolidated evidence bundle.

Still missing

• Provide the proband phenotype, clinical indication, and relevant differential diagnoses, with evidence that the phenotype is highly specific for a ZRSR2-associated condition.

Reasoning

The queried population datasets report the variant as absent, so they provide no observations in unaffected adults or, for this X-linked locus, no relevant hemizygous/homozygous observations from which BS2 could be assessed.

Rule

Generic ACMG/AMP BS2: variant observed in a healthy adult individual for a disorder expected to be fully penetrant at an early age.

Evidence considered

• The queried gnomAD datasets contain no record for the allele, precluding assessment of unaffected-adult observations or X-chromosome hemizygote observations.

Still missing

• Reliable observations of the variant in well-phenotyped unaffected adult individuals, with sex and phenotype context appropriate to the relevant ZRSR2-associated condition, are unavailable.

Sources

Reasoning

No ClinGen VCEP framework exists for ZRSR2 (cspec.found=false; framework_mode=generic_acmg). No functional assay evidence (validated or otherwise) showing normal/wild-type-like function for this specific variant was identified in the case bundle. No variant-specific functional data was reported by any source (OncoKB, literature search, ClinVar absent).

Still missing

• No functional assay data of any kind (supportive of normal or abnormal function) was located for this specific variant.

Reasoning

No informative affected non-carriers or unaffected carriers are documented; lack of segregation cannot be evaluated.

Rule

Generic ACMG/AMP fallback: BS4 requires robust evidence that the variant does not segregate with disease in an informative family; no such evidence is available.

Evidence considered

• No ZRSR2 CSPEC/VCEP specification was retrieved; the case is designated for generic ACMG/AMP fallback.
• The consolidated evidence bundle provides no family segregation or non-segregation observations for this variant.

Still missing

• Pedigree with phenotype status and age at assessment for informative relatives.
• Variant genotypes in affected and unaffected relatives.
• Assessment of phenocopies, variable expressivity, age-related penetrance, and genetic heterogeneity before using non-segregation as benign evidence.

Sources

Reasoning

No applicable ZRSR2 VCEP/gene-specific specification was retrieved, so generic ACMG/AMP criteria govern. The consolidated case evidence contains no confirmed co-occurrence of this variant with a pathogenic/likely pathogenic variant in cis, and no observation of the combination in a healthy individual for a fully penetrant dominant disorder. BP2 cannot be assessed.

Rule

Generic ACMG/AMP BP2; no qualifying cis co-occurrence or unaffected-carrier observation is available.

Evidence considered

• The governing-material records state that no ZRSR2 CSPEC/VCEP or local gene-specific framework was available and identify generic ACMG/AMP fallback.
• The case bundle supplies no co-occurring pathogenic/likely pathogenic variant, phase data, unaffected-carrier observation, penetrance information, or family data relevant to BP2.

Still missing

• Validated phase evidence showing cis co-occurrence with a pathogenic/likely pathogenic variant in the relevant gene, if such a variant exists.
• If evaluating the alternate generic BP2 scenario, documentation that the individual carrying the relevant variant combination is unaffected and that the applicable dominant disorder is fully penetrant.

Sources

Reasoning

No proband-level molecular testing results or clinical diagnosis was supplied to determine whether an alternate molecular cause fully explains the phenotype.

Rule

Generic ACMG/AMP BP5: use only when an alternate molecular basis fully explains the patient's phenotype; no ZRSR2-specific VCEP specification was available.

Evidence considered

• No proband phenotype or alternate molecular diagnosis was present in the consolidated evidence bundle.

Still missing

• Provide the proband phenotype and comprehensive molecular testing results, including any independently causal alternate diagnosis.