CALR c.1066C>G, p.(Gln356Glu)

Reasoning

PM2 is met at supporting strength under the generic fallback rare-variant rule. The variant is present only at an extremely low frequency in population databases: gnomAD v4.1 total AF 9.30e-06 and group maximum FAF 9.20e-05, with 0 homozygotes; gnomAD v2.1 total AF 1.20e-05 and group maximum FAF 2.599e-05, with 0 homozygotes. The maximum observed frequency is 0.0154%, below the 0.1% PM2 cutoff. The variant is therefore rare enough for PM2 supporting, although it is not completely absent from controls.

Rule

generic_fallback_population_threshold: PM2 supporting requires allele frequency <0.1% with no homozygotes

Evidence considered

• gnomAD v4.1 reports 15/1,613,672 total alleles (AF 9.29557e-06), group maximum FAF 9.197e-05, highest population AF 0.0001537515 in South Asians, and 0 homozygotes.
• gnomAD v2.1 reports 3/250,136 total alleles (AF 1.19935e-05), group maximum FAF 2.599e-05, highest population AF 9.81226e-05 in South Asians, and 0 homozygotes.
• gnomAD-Canada v1.0 reports the variant absent; this is concordant with extreme rarity but is not required for PM2 because the variant is already below the rare-variant cutoff in gnomAD v4.1 and v2.1.
• The maximum observed allele frequency, 0.0154%, is below the generic fallback PM2 supporting cutoff of 0.1%; this is a rare-variant assessment, not an absence claim.

Still missing

• The variant is observed at low frequency rather than completely absent from population databases.
• No CALR-specific VCEP PM2 rule or disease-specific prevalence/penetrance model was available.

Sources

Reasoning

No CALR VCEP CSPEC or gene-specific lookup spreadsheet was found (generic_acmg fallback), so generic calibrated computational thresholds apply. This missense variant (NP_004334.1:p.(Gln356Glu)) has a REVEL score of 0.095, below the calibrated benign-supporting threshold (REVEL <=0.290 per SVI calibration), supporting a benign missense in-silico prediction. Independently, SpliceAI predicts no splice impact (max delta score = 0.065, below the 0.2 splice-altering threshold), consistent with no computational evidence of a deleterious splicing effect. Both lines of evidence converge on a benign computational prediction with no conflicting signal; BP4 is applied once at supporting strength per ACMG/AMP convention (not stacked across the two predictor paths).

Rule

Generic ACMG/AMP 2015 (no CALR VCEP CSPEC available); REVEL calibrated benign threshold per SVI calibration (PMID 36413997); SpliceAI threshold per Jaganathan et al. 2019 (PMID 30661751).

Evidence considered

• REVEL score = 0.095, below the calibrated benign-supporting threshold of ~0.290 (Pejaver et al. calibration, PMID 36413997), supporting a benign missense prediction.
• SpliceAI max delta score = 0.065, below the 0.2 splice-altering threshold (Jaganathan et al. 2019, PMID 30661751), indicating no predicted splice disruption and no conflicting evidence against BP4.

Still missing

• BayesDel score is present but not used because no verified published calibration threshold for BayesDel is accessible to this pipeline.

Sources

Reasoning

PS1 requires an established pathogenic variant causing the identical amino-acid change (p.Gln356Glu) via a different nucleotide substitution. ClinVar has no entry for this variant (variant absent from ClinVar), and no VCEP/CSPEC is available for CALR (cspec.found=false, gene_vcep_dir=null) to supply a curated comparator list. The pm5_candidates harvesting artifact, which also collects same-residue comparators relevant to PS1-style analysis, explicitly found zero same-residue candidates and flagged the mode as unable to safely confirm same-residue semantics. No independent source in this bundle identifies any other variant producing p.Gln356Glu as pathogenic.

Rule

generic_ACMG_PS1

Evidence considered

• ClinVar search for CALR c.1066C>G returned no result; evidence_sentence: 'This variant is absent from ClinVar.'
• pm5_candidates.json recommendation.level = 'not_applicable', summary: 'Unable to confirm classic same-residue PM5 semantics safely; skipping automatic candidate harvesting.' candidate_summary.same_residue_candidates = 0, indicating no comparator pathogenic variant at codon 356 (same or alternate amino-acid change) was located in the pipeline's prior search.

Still missing

• No ClinVar record exists for NM_004343.3:c.1066C>G or any alternate codon change yielding p.Gln356Glu to serve as a PS1 comparator.
• No CALR-specific VCEP/CSPEC PS1 rule or curated pathogenic variant list was available to check.

Sources

Reasoning

PM1 requires the variant to lie in a well-established mutational hotspot or critical/well-characterized functional domain without benign variation. Residue 356 falls within the C-terminal, low-complexity acidic tail region of calreticulin, not within the structured N-domain or P-domain. cancerhotspots.org returned no statistically significant hotspot at CALR Q356 ('This variant does not lie in a statistically significant hotspot'), and OncoKB has no variant-specific or hotspot-level curated annotation for this position (classification 'Unknown Oncogenic Effect', not_reviewed=true). The well-documented CALR pathogenic mutational cluster for MPN is the somatic frameshift indel hotspot in exon 9 (~residues 367+, causing loss of the C-terminal KDEL motif), which is a distinct region and distinct mutation type from this germline missense substitution at codon 356.

Rule

generic_ACMG_PM1

Evidence considered

• cancerhotspots.org query 'CALR Q356' returned search_status='no_result'; evidence_sentence: 'This variant does not lie in a statistically significant hotspot.' (evidence.json hotspots result; not a registered source_registry key so cited here only, not in evidence_refs)
• OncoKB gene/CALR/Q356E page found but not_reviewed and not variant-specific; classification 'Unknown Oncogenic Effect'.
• The established CALR somatic mutational hotspot (type I/II exon 9 frameshift indels, residues ~367 onward, causing loss of the KDEL ER-retention signal) is spatially and mechanistically distinct from residue 356.

Still missing

• No structural/domain annotation source (e.g., UniProt domain boundaries) was consulted in this bundle to formally confirm residue 356 lies outside any critical structural domain.
• cancerhotspots.org and OncoKB coverage is somatic-cancer-hotspot oriented and may not reflect germline critical-residue significance.

Sources

Reasoning

PP2 requires the gene to have a low rate of benign missense variation and where missense variants are a common mechanism of disease. For CALR, the well-established germline/somatic disease mechanism in myeloproliferative neoplasms is not missense substitution but frameshift insertion/deletion in exon 9 that eliminates the C-terminal KDEL ER-retention signal, producing a novel, positively-charged C-terminal tail that drives constitutive MPL/JAK-STAT activation. Literature reviewed for this gene (PMID 42346284, full text) describes over 50 known somatic CALR mutations as 'insertions/deletions in exon 9 causing frameshift,' with type I (52bp deletion) and type II (5bp insertion) as the two dominant recurrent events; missense substitution is not described as an established CALR pathogenic mechanism. No missense constraint or established missense-pathogenicity precedent for CALR was identified in this bundle.

Rule

generic_ACMG_PP2

Evidence considered

• pvs1_gene_context.json: mechanism_rationale states 'Targeted germline literature review identified disease-focused publications supporting CALR loss of function as a germline disease mechanism' — loss-of-function/frameshift, not missense gain of function, is the supported mechanism.
• Full-text of PMID 42346284 ('Frequency of Non-Type I/II CALR Mutations...'): 'More than 50 somatic CALR mutations are known, which are insertions/deletions in exon 9 causing frameshift, which leads to the loss of the KDEL terminal signal... The two most common mutations are type I, a 52 bp deletion (p.L367fs*46), and type II, a 5 bp TTGTC insertion (p.K385fs*47).' This establishes frameshift indel, not missense, as the CALR pathogenic mechanism.

Still missing

• No gene-level missense constraint metric (e.g., gnomAD missense Z-score or o/e ratio) was retrieved in this bundle to quantitatively confirm a low rate of tolerated benign missense variation.
• No CALR-specific VCEP/CSPEC PP2 rule exists to formally define eligibility.

Sources

Reasoning

No ClinGen VCEP-specific CSPEC or gene-specific PP3/BP4 lookup spreadsheet was found for CALR (cspec.found=false; vcep_materials empty; framework_mode=generic_acmg), so generic calibrated computational thresholds apply. NM_004343.3:c.1066C>G (p.Gln356Glu) is a missense variant. SpliceAI predicts no splice impact (max delta score = 0.065, well below the 0.2 splice-altering threshold established by Jaganathan et al. 2019), so the splice-impact path does not support PP3. On the missense path, REVEL score = 0.095, far below the calibrated pathogenic-supporting threshold (REVEL >=0.644 per SVI calibration), so it does not support PP3 either. BayesDel score (-0.728488) is available but has no verified published calibration threshold accessible to this pipeline and is therefore not usable as evidence for PP3, consistent with the standing rule that an uncited/uncalibrated predictor score must be treated as not_available. No computational evidence supports a pathogenic/damaging effect via either the splice or missense in-silico path.

Rule

Generic ACMG/AMP 2015 (no CALR VCEP CSPEC available); SpliceAI splice-altering threshold per Jaganathan et al. 2019 (PMID 30661751); REVEL calibrated pathogenic threshold per SVI calibration (PMID 36413997).

Evidence considered

• SpliceAI max delta score = 0.065 (DS_AG=0.014, DS_AL=0.065, DS_DG=0.0, DS_DL=0.0), below the 0.2 splice-altering threshold from Jaganathan et al. 2019 (PMID 30661751), indicating no predicted splice disruption.
• REVEL score = 0.095, well below the calibrated pathogenic-supporting threshold of ~0.644 (Pejaver et al., calibration referenced as PMID 36413997), indicating the missense in-silico predictor does not support a damaging effect.

Still missing

• BayesDel score is present but excluded because no verified publication/threshold for BayesDel is available to this pipeline.
• No gene-specific VCEP PP3/BP4 lookup table was available for CALR to override generic thresholds.

Sources

Reasoning

BA1 is not met. NM_004343.3:c.1066C>G (CALR p.Gln356Glu) has a maximum observed population frequency far below a benign stand-alone frequency threshold: gnomAD v4.1 total AF is 9.30e-06, the highest population AF is 1.54e-04 in South Asians, and the group maximum FAF is 9.20e-05; gnomAD v2.1 group maximum FAF is 2.599e-05. These values are well below the generic fallback BA1 threshold of 1% and do not represent a threshold artifact caused by a single inadequately sized ancestry group.

Rule

generic_fallback_population_threshold: BA1 requires allele frequency >1%

Evidence considered

• gnomAD v4.1 reports 15/1,613,672 total alleles (AF 9.29557e-06), 14/91,056 South Asian alleles (AF 0.0001537515), and group maximum FAF 9.197e-05; homozygotes = 0.
• gnomAD v2.1 reports 3/250,136 total alleles (AF 1.19935e-05), 3/30,574 South Asian alleles (AF 9.81226e-05), and group maximum FAF 2.599e-05; homozygotes = 0.
• The observed frequencies are orders of magnitude below the generic fallback BA1 threshold of 1%; BA1 therefore cannot be applied as a stand-alone benign criterion.

Sources

Reasoning

BS1 is not met. The variant is rare in both gnomAD datasets: gnomAD v4.1 group maximum FAF is 9.20e-05 and gnomAD v2.1 group maximum FAF is 2.599e-05, both below the generic fallback BS1 threshold of 0.3%. The data do not show a frequency greater than expected for a CALR disease allele. The South Asian enrichment is based on 14 alleles in gnomAD v4.1 and 3 alleles in gnomAD v2.1, but remains far below the BS1 threshold and is not sufficient to establish a benign frequency signal.

Rule

generic_fallback_population_threshold: BS1 requires allele frequency >0.3%

Evidence considered

• gnomAD v4.1 group maximum FAF is 9.197e-05, with the highest population AF 0.0001537515 in South Asians (14/91,056 alleles).
• gnomAD v2.1 group maximum FAF is 2.599e-05, with the highest population AF 9.81226e-05 in South Asians (3/30,574 alleles).
• Both group-maximum frequencies are below the generic fallback BS1 threshold of 0.3%; no disease prevalence or penetrance model supports a lower benign threshold for CALR in this case.

Sources

Reasoning

BS2 is not met. gnomAD v4.1 and v2.1 each report zero homozygotes for this variant, so there is no observation of the variant in healthy adults in the zygosity expected to support BS2. The absence of homozygotes is not positive BS2 evidence; it is also not strong evidence against pathogenicity because the allele is rare and the available population data provide no phenotype, ascertainment, penetrance, or disease-prevalence information.

Rule

BS2 requires observed healthy adult individuals with the variant in the relevant zygosity; absence of homozygotes does not meet BS2

Evidence considered

• gnomAD v4.1 reports 0 homozygotes among 1,613,672 alleles, and the South Asian subgroup reports 0 homozygotes among 91,056 alleles.
• gnomAD v2.1 reports 0 homozygotes among 250,136 alleles, and the South Asian subgroup reports 0 homozygotes among 30,574 alleles.
• BS2 requires observation in healthy adults in the relevant zygosity; zero homozygotes cannot satisfy that positive-evidence requirement.

Still missing

• No phenotype-level or penetrance data for carriers in population cohorts.
• No disease-prevalence/inheritance model permitting a stronger interpretation of zero observed homozygotes.

Sources

Reasoning

No proband phenotype, de novo status, or parental testing results are provided for NM_004343.3:c.1066C>G (NP_004334.1:p.(Gln356Glu)). Confirmed de novo occurrence cannot be established.

Rule

Generic ACMG/AMP PS2: confirmed de novo variant in a patient with the disease and no family history, with maternity/paternity confirmed as applicable; strength requires the applicable de novo evidence framework.

Evidence considered

• The consolidated case-evidence bundle contains no proband-level clinical description or parental testing results.
• The case is in generic_acmg fallback mode because no applicable CALR ClinGen specification or VCEP framework was found.

Still missing

• Proband clinical phenotype and diagnosis
• Variant testing results for both biological parents
• Confirmation that the variant is absent in both parents
• Assessment for parental relationships, maternity/paternity, and mosaicism where relevant
• Phenotype consistency with a CALR-associated germline disorder

Reasoning

No CALR VCEP specification is available for this case (gene_vcep_dir null, cspec.found false), so generic ACMG applies. No functional/experimental assay data for NM_004343.3:c.1066C>G (p.Gln356Glu) was found in evidence.json, literature_pass.json, vcep_materials.json, or fulltext_map.json; literature_pass.json explicitly reports total_pmids=0 and no PMIDs across evidence sources. There is no validated functional assay (e.g., cell-based CALR mutant calreticulin/MPL activation assay) evidence to evaluate for pathogenic effect.

Still missing

• No published functional assay data identified for this specific variant (no PMIDs retrieved for CALR at all)
• No CALR-specific VCEP functional assay specification available to define approved assay types or strength calibration

Reasoning

No affected-case series, case-control enrichment analysis, or phenotype prevalence data for the exact variant NM_004343.3:c.1066C>G (NP_004334.1:p.Gln356Glu) were provided. The available bundle contains no literature PMIDs or case-level phenotype evidence, so PS4 cannot be assessed.

Rule

Generic ACMG/AMP PS4: increased prevalence of the variant in affected individuals compared with controls; no qualifying exact-variant case-control evidence is available.

Evidence considered

• The consolidated bundle reports zero literature PMIDs and provides no affected-case or case-control phenotype dataset for the exact variant.

Still missing

• Exact-variant affected-case counts and an appropriate control comparison are unavailable.

Reasoning

No applicable CALR-specific VCEP or gene-specific framework was identified, so the generic ACMG/AMP PM3 concept was considered. The case bundle contains no affected proband observations, no second pathogenic or likely pathogenic variant, no biallelic genotype, and no phase or parental testing information. PM3 therefore cannot be assessed.

Rule

Generic ACMG/AMP PM3: use affected individuals with a recessive disorder who carry the variant in trans with a pathogenic variant; phase and disease/inheritance context must support the observation.

Evidence considered

• The consolidated case bundle identifies CALR NM_004343.3:c.1066C>G (p.Gln356Glu) but reports no case-level clinical observations or genotype-phase data.
• The bundle reports no applicable CSPEC/VCEP materials for CALR and indicates generic ACMG fallback.
• No relevant publication was available in the case literature pass or full-text map for evaluation of PM3.

Still missing

• Affected-proband or affected-case observations involving NM_004343.3:c.1066C>G
• A confirmed pathogenic or likely pathogenic allele in the same individual, if relevant to the disease's inheritance model
• Phase information establishing trans or cis configuration
• Inheritance mode and parental segregation data

Reasoning

No case or publication evidence indicates that NM_004343.3:c.1066C>G occurred de novo without parental confirmation. The available bundle contains no proband, family-history, or parental-testing data.

Rule

Generic ACMG/AMP PM6: assumed de novo, without confirmation of maternity/paternity, in a patient with the disease and no family history; apply only when the de novo observation itself is documented.

Evidence considered

• The consolidated case-evidence bundle reports no literature PMIDs and provides no de novo case report or family data for this variant.
• The consolidated case-evidence bundle contains no parental testing or family-history information.

Still missing

• Evidence of a de novo occurrence
• Parental testing status and whether parentage was confirmed
• Proband phenotype and disease diagnosis
• Family history relevant to de novo assessment

Reasoning

No affected relatives, informative meioses, pedigree, or segregation testing results are available. Segregation of NM_004343.3:c.1066C>G with a consistent phenotype cannot be evaluated.

Rule

Generic ACMG/AMP PP1: co-segregation with disease in multiple affected family members; assess the number and informativeness of meioses and phenotype specificity before assigning strength.

Evidence considered

• The consolidated case-evidence bundle contains no pedigree, family segregation, meiosis, or relative-testing information.
• The available sources concern variant normalization and population or database lookups and do not provide segregation evidence.

Still missing

• Pedigree and number of affected and unaffected relatives tested
• Genotype results for informative relatives
• Phenotype definitions and consistency across relatives
• Number of informative meioses
• Potential phenocopies, reduced penetrance, or alternative diagnoses

Reasoning

No patient phenotype, clinical diagnosis, or disease-specific phenotype spectrum for an individual carrying the exact variant was provided. Therefore, phenotype specificity cannot be assessed for PP4.

Rule

Generic ACMG/AMP PP4: the patient's phenotype or family history is highly specific for a disease with a single-gene etiology; no qualifying patient phenotype is available.

Evidence considered

• The consolidated bundle contains no case-level clinical phenotype information and no literature PMIDs for this exact variant.

Still missing

• Clinical phenotype and diagnostic specificity for an exact-variant carrier are unavailable.

Reasoning

PP5 is restricted here to an exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic assertion. The bundle records no exact ClinVar match and states that the variant is absent from ClinVar; consequently, no qualifying expert-panel assertion is available.

Rule

PP5 may be applied only when an exact-variant ClinVar expert-panel classification is Pathogenic or Likely pathogenic; ordinary submissions, aggregate labels, and non-expert assertions are excluded.

Evidence considered

• ClinVar search for CALR NM_004343.3:c.1066C>G returned no result; no expert-panel assertion was identified.

Still missing

• A qualifying exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification is unavailable.

Sources

Reasoning

No CALR VCEP specification is available for this case (gene_vcep_dir null, cspec.found false), so generic ACMG applies. No functional/experimental assay data showing lack of damaging effect for NM_004343.3:c.1066C>G (p.Gln356Glu) was found in evidence.json, literature_pass.json, vcep_materials.json, or fulltext_map.json; literature_pass.json explicitly reports total_pmids=0 and no PMIDs across evidence sources.

Still missing

• No published functional assay data identified for this specific variant (no PMIDs retrieved for CALR at all)
• No CALR-specific VCEP functional assay specification available to define approved assay types or strength calibration

Reasoning

No unaffected relatives with reliable phenotype assessment and confirmed absence of NM_004343.3:c.1066C>G are reported. Non-segregation cannot be evaluated.

Rule

Generic ACMG/AMP BS4: absence of segregation in affected members of a family, or presence of the variant in unaffected relatives when the phenotype and penetrance make that observation informative; require adequate phenotype and genotype data.

Evidence considered

• The consolidated case-evidence bundle contains no unaffected-relative testing or non-segregation observations.
• No publication was available in the case bundle that could establish non-segregation for this variant.

Still missing

• Phenotype-assessed unaffected relatives
• Genotype results for those relatives
• Informative family structure and expected transmission
• Consideration of age-dependent penetrance, variable expressivity, and phenocopies

Reasoning

BP1 (missense variant in a gene where primarily truncating variants are known to cause disease) requires cautious application because the CALR disease mechanism is not classic loss-of-function truncation: the pathogenic exon 9 frameshift indels create a shifted reading frame that generates a novel, non-truncated C-terminal peptide with altered charge (loss of the KDEL retention signal and gain of a novel tail), which is mechanistically a gain-of-function/neomorphic event rather than simple truncating loss-of-function. Applying BP1's premise ('primarily truncating variants known to cause disease') to CALR without a VCEP-endorsed rule risks mischaracterizing this non-classic mechanism. No formal missense-vs-truncating constraint statistics were available in this bundle to support or refute BP1 quantitatively.

Rule

generic_ACMG_BP1

Evidence considered

• Full-text of PMID 42346284 and pvs1_gene_context.json both describe the CALR pathogenic mechanism as frameshift insertion/deletion producing a novel C-terminal sequence (not simple nonsense/premature-truncation loss of KDEL alone), which complicates direct application of the generic BP1 'primarily truncating variants' premise without VCEP guidance.

Still missing

• No CALR VCEP/CSPEC BP1 rule is available to define whether/how BP1 should be applied given CALR's atypical (frameshift-generated neomorphic C-terminus, not simple truncation) disease mechanism.
• No gene-level missense vs. truncating variant constraint data (e.g., gnomAD constraint metrics, ClinVar variant-type distribution) was retrieved to quantitatively assess this criterion.

Reasoning

No applicable CALR-specific VCEP or gene-specific framework was identified, so the generic ACMG/AMP BP2 concept was considered. The case bundle contains no individual-level observation of this variant with another pathogenic variant, and no phase, cis/trans, affected-status, or inheritance information. BP2 therefore cannot be assessed.

Rule

Generic ACMG/AMP BP2: consider an observation of the variant in trans with a pathogenic variant for a dominant disorder or in cis with a pathogenic variant for a recessive disorder, with disease and phase context established.

Evidence considered

• The consolidated case bundle contains no affected-proband, family, co-occurrence, phase, or parental segregation observations for NM_004343.3:c.1066C>G.
• The bundle reports no applicable CSPEC/VCEP materials for CALR and indicates generic ACMG fallback.
• No relevant publication was available in the case literature pass or full-text map for evaluation of BP2.

Still missing

• A documented individual or family carrying NM_004343.3:c.1066C>G together with a pathogenic variant
• Reliable phase determination showing cis or trans configuration
• Affected status and phenotype of the individual(s) carrying both variants
• Disease inheritance model needed to interpret the allelic observation

Reasoning

No clinical phenotype or confirmed pathogenic variant in an alternative gene explaining the presented phenotype was provided. BP5 therefore cannot be assessed.

Rule

Generic ACMG/AMP BP5: a pathogenic variant in an alternative gene explains the phenotype; no qualifying alternative diagnosis is available.

Evidence considered

• The consolidated bundle contains no case phenotype and no alternative-gene pathogenic diagnosis.

Still missing

• An alternative molecular diagnosis and phenotype evidence explaining the case are unavailable.

Reasoning

BP6 is restricted here to an exact-variant ClinVar expert-panel Benign or Likely benign assertion. The bundle records no exact ClinVar match and states that the variant is absent from ClinVar; consequently, no qualifying expert-panel assertion is available.

Rule

BP6 may be applied only when an exact-variant ClinVar expert-panel classification is Benign or Likely benign; ordinary submissions, aggregate labels, and non-expert assertions are excluded.

Evidence considered

• ClinVar search for CALR NM_004343.3:c.1066C>G returned no result; no expert-panel assertion was identified.

Still missing

• A qualifying exact-variant ClinVar expert-panel Benign or Likely benign classification is unavailable.

Sources