IDH2 c.374-10G>A, p.?

Reasoning

Same generic fallback rationale as PP3: no IDH2 VCEP/CSPEC framework available, so SpliceAI governs for this non-missense (intronic) variant. SpliceAI max delta score of 0.003 is well below the 0.1 threshold for BP4 (supporting), so BP4 is met.

Rule

generic_acmg_fallback: SpliceAI max delta < 0.1 -> BP4 supporting (intronic/non-missense variant path)

Evidence considered

• SpliceAI Lookup max delta score = 0.003 for NM_002168.2:c.374-10G>A, below the 0.1 generic-fallback BP4 threshold.
• generic_acmg_classification_rules.md: SpliceAI max delta < 0.1 -> BP4 (supporting) for intronic/non-canonical-splice-position variants under the generic non-VCEP fallback.

Sources

Reasoning

PM2 is not met because the variant is present in population databases. It is observed at 5/1,613,834 alleles in gnomAD v4.1 and 1/251,344 alleles in gnomAD v2.1, with zero homozygotes in both datasets; it is absent from gnomAD-Canada v1.0, but absence in one database does not override confirmed presence in large general-population datasets.

Rule

Generic ACMG/AMP 2015 PM2: absence from a large general population or race-matched control cohort supports moderate pathogenic evidence; confirmed population presence means PM2 is not met.

Evidence considered

• gnomAD v4.1 reports the variant present at 5/1,613,834 alleles (AF 3.09821e-06), with zero homozygotes.
• gnomAD v2.1 reports the variant present at 1/251,344 alleles (AF 3.97861e-06), with zero homozygotes.
• gnomAD-Canada v1.0 reports the variant absent, but this single-database absence is insufficient for PM2 when the variant is present in gnomAD v2.1 and v4.1.
• Richards et al. 2015 (PMID:25741868) describes PM2 as absence from a large general population or race-matched control cohort; the confirmed observations in gnomAD v2.1 and v4.1 fail that requirement.

Sources

Reasoning

No ClinGen VCEP/CSPEC framework was found for IDH2 (vcep_materials.json: found=false, no cspec url; case_summary.json cspec.found=false), so the generic ACMG/AMP in-silico calibration applies. The variant NM_002168.2:c.374-10G>A is an intronic, non-canonical splice-region substitution (not a missense change), so the applicable predictor is SpliceAI, not REVEL/BayesDel. SpliceAI Lookup reports all four delta scores (DS_AG=0.002, DS_AL=0.003, DS_DG=0.00, DS_DL=0.00) with a maximum delta score of 0.003, far below the 0.2 threshold required to invoke PP3 under the generic fallback rule. PP3 is not met.

Rule

generic_acmg_fallback: SpliceAI max delta > 0.2 -> PP3 supporting (intronic/non-missense variant path)

Evidence considered

• SpliceAI Lookup (spliceailookup.broadinstitute.org) for NM_002168.2:c.374-10G>A reports max delta score = 0.003 (DS_AG=0.002, DS_AL=0.003, DS_DG=0.0, DS_DL=0.0), well below the 0.2 supporting-pathogenic threshold.
• generic_acmg_classification_rules.md specifies SpliceAI max delta > 0.2 -> PP3 (supporting) for intronic/non-canonical-splice-position variants when no VCEP-specific PP3/BP4 rule governs; this variant's own REVEL/BayesDel scores are not applicable since it is not a missense substitution.

Sources

Reasoning

The exact ClinVar record for NM_002168.2:c.374-10G>A has one single-submitter clinical laboratory assertion of uncertain significance with criteria provided. It has zero expert-panel submissions. Under the governing rule, a non-expert laboratory assertion and an uncertain classification cannot trigger PP5.

Rule

Apply PP5 only when the exact variant has a ClinVar expert-panel Pathogenic or Likely pathogenic classification; ordinary laboratory submissions, aggregate labels, non-expert assertions, and uncertain classifications do not qualify.

Evidence considered

• ClinVar variation 158666 is an exact variant match and reports Uncertain significance, criteria provided, single submitter; the submission is from Genetic Services Laboratory, University of Chicago.
• The ClinVar submission audit reports expert_panel_submissions = 0 and identifies the sole submission as non-expert clinical testing.
• PP5 is restricted here to an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification; that qualifying assertion is absent.

Sources

Reasoning

BA1 is not met. The variant is observed in gnomAD v4.1 at an overall allele frequency of 3.09821e-06 (5/1,613,834 alleles), with the highest listed population frequency of 4.00662e-05 in African/African American individuals; these frequencies are far below the generic ACMG/AMP stand-alone benign threshold of greater than 5%.

Rule

Generic ACMG/AMP 2015 BA1: allele frequency over 5% in a control population supports stand-alone benign evidence.

Evidence considered

• gnomAD v2.1 reports 1/251,344 alleles, overall AF 3.97861e-06, no homozygotes, and a highest South Asian AF of 3.26648e-05.
• gnomAD v4.1 reports 5/1,613,834 alleles, overall AF 3.09821e-06, no homozygotes, grpmax FAF 1.064e-05, and a highest African/African American AF of 4.00662e-05.
• Richards et al. 2015 (PMID:25741868) states that an allele frequency over 5% provides stand-alone benign support (BA1); the observed frequencies are well below this threshold.

Sources

Reasoning

The exact ClinVar record has only a non-expert single-submitter uncertain-significance assertion and no expert-panel submissions. It does not contain an expert-panel Benign or Likely benign classification, so BP6 is not met.

Rule

Apply BP6 only when the exact variant has a ClinVar expert-panel Benign or Likely benign classification; ordinary laboratory submissions, aggregate labels, non-expert assertions, and uncertain classifications do not qualify.

Evidence considered

• ClinVar variation 158666 is an exact variant match with an Uncertain significance classification from one clinical laboratory submission.
• The ClinVar submission audit reports expert_panel_submissions = 0; ordinary laboratory submissions and non-expert assertions are explicitly excluded from BP6.
• BP6 is restricted here to an exact-variant ClinVar expert-panel Benign or Likely benign classification, which is absent.

Sources

Reasoning

No case-level evidence documents a de novo occurrence of NM_002168.2:c.374-10G>A, including confirmed parental genotypes, maternity/paternity, or a phenotype-consistent proband. The evidence bundle contains no proband description, parental testing result, meiosis information, or relevant variant-specific publication. PS2 cannot be assessed.

Evidence considered

• The case bundle identifies the variant as NM_002168.2:c.374-10G>A in IDH2 but provides no proband-level clinical or family segregation data.
• The case bundle reports no applicable IDH2 ClinGen/VCEP specification; generic ACMG/AMP fallback is therefore used, but generic criteria definitions do not supply case-specific de novo evidence.
• The only indexed publication is a general ACMG/AMP guideline and does not report this variant or a de novo family; it cannot establish PS2 for this case.

Still missing

• Proband phenotype and diagnostic context
• Testing results for both biological parents
• Confirmation of maternity and paternity
• Evidence that the variant is absent in both parents and arose de novo
• Independent confirmation of the variant in the proband

Reasoning

No IDH2-specific ClinGen VCEP specification is available for this case (cspec.found=false, gene_vcep_dir=null), so generic ACMG applies. No well-established functional assay data (splicing minigene/RT-PCR, enzymatic 2-HG production assay, or other validated functional study) for NM_002168.2:c.374-10G>A was identified in the evidence bundle. OncoKB functional classification field is empty, COSMIC has no record, and the only literature hit (PMID:25741868) is the ACMG/AMP methods paper itself, not a variant-specific functional study; it does not mention this variant and cannot be used per grounding rules. SpliceAI predicts no significant splice impact (max delta 0.003), but this is an in silico prediction, not a functional assay, so it cannot support PS3 directly.

Evidence considered

• case_summary.json cspec.found=false and review_manifest.json gene_vcep_dir=null confirm no IDH2 VCEP specification is available for this case
• compact_evidence.oncokb.classification and evidence_sentence are both empty strings, indicating no OncoKB functional/clinical data was retrieved
• compact_evidence.cosmic.found=false, no somatic functional/observational data available
• literature_packet lists only PMID 25741868 (the ACMG/AMP 2015 standards paper) as high priority; its abstract snippet discusses general variant interpretation methodology, not any IDH2 functional assay, and the variant is not mentioned in it
• compact_evidence.spliceai shows max_delta_score=0.003 (no significant predicted splice impact), which is an in silico prediction and not admissible as PS3 functional assay evidence

Still missing

• No functional/splicing assay (e.g., minigene, RT-PCR on patient RNA, enzymatic activity assay) has been performed or reported for c.374-10G>A
• No IDH2 VCEP-specific PS3 assay validation criteria available to calibrate strength even if a study were found

Reasoning

No affected-case series, case-control enrichment analysis, or statistically interpretable phenotype prevalence evidence for the exact variant NM_002168.2:c.374-10G>A was available in the case-evidence bundle. PS4 cannot be assessed from the variant's ClinVar record or population-frequency data.

Rule

Generic ACMG/AMP fallback: PS4 requires a statistically significant excess of the variant in affected individuals compared with an appropriate control population; no such evidence is available.

Evidence considered

• The consolidated bundle contains no exact-variant affected-case series, case-control dataset, phenotype counts, or enrichment statistic relevant to PS4.
• The only retrieved publication is a general ACMG/AMP standards paper and does not report this exact IDH2 variant or provide PS4 evidence.

Still missing

• Affected-case counts and an appropriate comparison/control cohort for the exact variant are unavailable.
• No case-control enrichment statistic or independently curated clinical case series is available.

Reasoning

No affected-proband observations, second pathogenic or likely pathogenic variant, parental testing, or phase information are documented for NM_002168.2:c.374-10G>A. Therefore, the recessive allelic evidence required for PM3 cannot be assessed.

Rule

Generic ACMG/AMP PM3: apply only when the variant is observed in trans with a pathogenic or likely pathogenic variant in an affected individual with a recessive disorder, with phase and disease relevance adequately established; strength depends on the validated number and quality of observations.

Evidence considered

• The consolidated case-evidence bundle contains no case observations, compound-heterozygous findings, second IDH2 variant, parental testing, or cis/trans phase information for the target variant.
• No applicable IDH2-specific VCEP or gene-specific framework was available; generic ACMG/AMP assessment is therefore the fallback, but its PM3 requirements are not met by the available case data.

Still missing

• Affected proband case-level data are absent.
• No pathogenic or likely pathogenic allele in the same gene is documented in the case.
• No parental testing or validated trans phase is documented.
• Inheritance mode and the relevant IDH2 disease context are not established for this case.

Reasoning

No case-level evidence documents a presumed de novo occurrence of NM_002168.2:c.374-10G>A without parental testing. There is no reported proband phenotype, family history, parental genotype information, or phenotype-consistent clinical report for this exact variant. PM6 cannot be assessed.

Evidence considered

• The consolidated case-evidence bundle contains no parental testing or family-history observations for this variant.
• The literature packet contains no variant-specific case report or segregation study; the sole publication is a general ACMG/AMP guideline.
• Population observations in gnomAD are not proband or parental testing evidence and cannot establish PM6.

Still missing

• Proband phenotype and diagnostic context
• Documented absence of the variant in available parents or an explanation for unavailable parental testing
• Maternity and paternity information
• Variant-specific clinical report describing presumed de novo status

Reasoning

No affected or unaffected relatives, informative meioses, phenotype assessments, or variant genotypes are reported. Consequently, there is no positive familial segregation evidence to support PP1 at any strength.

Evidence considered

• The case bundle contains no pedigree, familial variant observations, segregation counts, or meioses for NM_002168.2:c.374-10G>A.
• The available literature does not report this exact variant in a family or provide segregation data relevant to PP1.
• The ClinVar record is a single-submitter uncertain-significance record with no reported family segregation data in the case bundle.

Still missing

• Pedigree or family structure
• Genotypes of affected relatives
• Genotypes and phenotypes of unaffected relatives
• Number of informative meioses
• Phenotype consistency and assessment of affected relatives

Reasoning

The case materials provide no patient phenotype, highly specific clinical findings, or molecularly distinctive syndrome description for this exact variant. PP4 therefore cannot be assessed.

Rule

Generic ACMG/AMP fallback: PP4 requires a phenotype or highly specific clinical finding consistent with a disease associated with the gene; no qualifying phenotype is available.

Evidence considered

• The case-evidence bundle contains no clinical phenotype description for the tested individual and no phenotype-specific case report for NM_002168.2:c.374-10G>A.
• The exact ClinVar record lists the condition as not specified, which does not establish phenotype specificity for PP4.

Still missing

• Patient-level phenotype and clinical diagnostic information are absent.
• No evidence establishes that the phenotype is highly specific for an IDH2-related disorder.

Reasoning

BS1 cannot be adjudicated from the available evidence because no disease-specific expected allele-frequency threshold, disease prevalence, inheritance model, or penetrance estimate is provided for the relevant IDH2 disorder. The observed population frequencies are very low, but a direct comparison with the expected frequency for disease is required for BS1.

Rule

Generic ACMG/AMP 2015 BS1: allele frequency greater than expected for the disorder supports strong benign evidence; the expected frequency must be established for the disease context.

Evidence considered

• gnomAD v4.1 reports an overall AF of 3.09821e-06 and grpmax FAF of 1.064e-05; the highest listed subpopulation AF is 4.00662e-05 in African/African American individuals.
• Richards et al. 2015 (PMID:25741868) defines BS1 as a control-population allele frequency greater than expected for the disorder, so a disorder-specific expected frequency is necessary rather than a generic frequency comparison.

Still missing

• A disease-specific maximum credible allele frequency or equivalent BS1 threshold is not available in the case bundle.
• Disease prevalence, inheritance model, and penetrance needed to derive the expected allele frequency are not established in the case evidence.

Sources

Reasoning

BS2 is not assessable from the available population data. gnomAD reports no homozygotes, but the case bundle does not document a well-phenotyped healthy adult carrying the variant, does not establish that the relevant disease is fully penetrant and early onset, and does not provide the clinical status or zygosity of individual carriers. Absence of homozygotes alone does not satisfy BS2.

Rule

Generic ACMG/AMP 2015 BS2: a variant observed in a well-documented healthy adult is strong benign evidence when the disease is fully penetrant at an early age, with zygosity appropriate to the inheritance model.

Evidence considered

• gnomAD v2.1 and v4.1 each report zero homozygotes; neither dataset supplies the clinical phenotype, age, disease ascertainment, or individual-level documentation required for BS2.
• Richards et al. 2015 (PMID:25741868) requires a well-documented healthy adult observation and, for a recessive condition, homozygosity; it also specifies a fully penetrant disease with early onset.

Still missing

• No documented healthy adult carrier with verified phenotype and zygosity is available.
• The relevant disease's early-onset full penetrance is not established in the case bundle.

Sources

Reasoning

As with PS3, no IDH2 VCEP specification is available (generic ACMG framework applies), and no functional assay data (splicing, enzymatic, or other validated functional study) showing normal/wild-type function for NM_002168.2:c.374-10G>A was identified in the evidence bundle. No functional study of any kind, supportive or refuting, exists in the case evidence.

Evidence considered

• case_summary.json cspec.found=false and review_manifest.json gene_vcep_dir=null confirm no IDH2 VCEP specification is available for this case
• No functional assay results present in evidence.json, prefetch.json, or literature_pass.json high_priority_papers beyond the ACMG/AMP methods paper (PMID 25741868), which does not mention this variant

Still missing

• No functional/splicing assay data of any kind is available for this variant to assess whether it shows normal function

Reasoning

No affected family members or tested relatives are described, so there is no reliable non-segregation observation showing that the variant is absent in affected relatives. BS4 cannot be assessed.

Evidence considered

• The consolidated case-evidence bundle contains no family members, affected-relative genotypes, non-segregation findings, or pedigree data.
• Absence of family data is not evidence that the variant fails to segregate; therefore BS4 is not met and is not assessable.
• No variant-specific publication in the available literature provides non-segregation evidence for this variant.

Still missing

• Affected relatives with a phenotype relevant to the suspected disorder
• Reliable genotypes for affected relatives
• Phenotype verification and age-appropriate assessment of relatives
• Family structure and number of informative relatives/meioses

Reasoning

No phase-resolved observation places NM_002168.2:c.374-10G>A in cis with a pathogenic variant or in trans with a pathogenic variant in a disorder for which the relevant inheritance model supports BP2. The available evidence is limited to population data, ClinVar uncertainty, and splice prediction, none of which evaluates allelic configuration.

Rule

Generic ACMG/AMP BP2: apply only with documented allelic configuration showing the variant in trans with a pathogenic variant in a fully penetrant dominant disorder or in cis with a pathogenic variant, as appropriate to the disease model; phase and the pathogenic status of the comparator must be established.

Evidence considered

• The consolidated case-evidence bundle contains no phase-resolved family or proband data relevant to BP2.
• Population frequency, ClinVar classification, and SpliceAI output do not provide evidence for cis or trans allelic configuration and were not used to apply BP2.
• No applicable IDH2-specific VCEP or gene-specific framework was available; generic ACMG/AMP assessment is therefore the fallback, but BP2 cannot be assessed without phase-resolved allelic observations.

Still missing

• No cis/trans phase information is documented.
• No second pathogenic variant in IDH2 is documented.
• No affected or unaffected individual-level genotype and phenotype observations are documented.
• The inheritance model relevant to this variant is not established.

Reasoning

No evidence in the case bundle shows that the patient's phenotype is better explained by an alternate molecular cause, and no relevant phenotype or alternate diagnosis is provided. BP5 cannot be assessed.

Rule

Generic ACMG/AMP fallback: BP5 requires a pathogenic variant in an alternative gene or another molecular explanation that better accounts for the patient's phenotype; no such evidence is available.

Evidence considered

• The consolidated evidence bundle contains no patient-level phenotype, alternate molecular diagnosis, or evidence that another variant better explains the putative disorder.
• The exact ClinVar submission has condition labels of not specified and cannot establish an alternate molecular cause.

Still missing

• Patient phenotype and diagnostic workup are unavailable.
• No documented alternate pathogenic variant or molecular explanation is available for comparison.