PIK3R1 c.1126G>A, p.(Gly376Arg)

Reasoning

The variant is absent from gnomAD v4.1, meeting the PIK3R1 VCEP PM2_Supporting requirement for total allele frequency <0.00000132 across all populations.

Rule

PIK3R1 VCEP PM2_Supporting: total allele frequency <0.00000132 across all gnomAD v4.1 populations.

Evidence considered

• gnomAD v4.1 reports NM_181523.2:c.1126G>A (p.Gly376Arg) as absent, corresponding to a total observed allele frequency of zero in that dataset.
• The PIK3R1 Antibody Deficiencies VCEP specifies PM2 at Supporting strength when total allele frequency across all gnomAD v4.1 populations is <0.00000132.

Sources

Reasoning

The variant is absent from gnomAD v4.1; therefore it does not meet the PIK3R1 VCEP BA1 threshold of GrpMax filtering allele frequency >=0.00316.

Rule

PIK3R1 VCEP BA1: gnomAD v4.1 GrpMax filtering allele frequency >=0.00316; if unavailable, use the maximum allele frequency among the five major continental populations.

Evidence considered

• gnomAD v4.1 reports NM_181523.2:c.1126G>A (p.Gly376Arg) as absent, so no GrpMax filtering allele frequency approaches the VCEP BA1 cutoff.
• The PIK3R1 Antibody Deficiencies VCEP specifies BA1 for gnomAD v4.1 GrpMax filtering allele frequency >=0.00316, or the maximum frequency in a major continental population when GrpMax is unavailable.

Sources

Reasoning

The variant is absent from gnomAD v4.1; therefore it does not meet the PIK3R1 VCEP BS1 threshold of GrpMax filtering allele frequency >=0.000316.

Rule

PIK3R1 VCEP BS1: gnomAD v4.1 GrpMax filtering allele frequency >=0.000316; if unavailable, use the maximum allele frequency among the five major continental populations.

Evidence considered

• gnomAD v4.1 reports NM_181523.2:c.1126G>A (p.Gly376Arg) as absent, so no GrpMax filtering allele frequency approaches the VCEP BS1 cutoff.
• The PIK3R1 Antibody Deficiencies VCEP specifies BS1 for gnomAD v4.1 GrpMax filtering allele frequency >=0.000316, or the maximum frequency in a major continental population when GrpMax is unavailable.

Sources

Reasoning

NM_181523.2:c.1126G>A (p.Gly376Arg) is a missense variant. The PIK3R1 VCEP BP4 (Supporting) missense clause requires BOTH REVEL <=0.290 AND CADD <=21.5. REVEL = 0.778 for this variant, far above the 0.290 ceiling, so the AND condition fails outright regardless of the (unavailable) CADD value -- BP4 cannot be met via the missense predictor path. The VCEP's BP4 splice-based clause only applies to synonymous or intronic variants not predicted to impact splicing by SpliceAI; this variant is missense, so that clause is not applicable to it. Both applicable BP4 paths are therefore definitively not met.

Rule

PIK3R1 VCEP (cspec v1.0) BP4: missense clause requires REVEL<=0.290 AND CADD<=21.5; splice/synonymous-intronic clause requires SpliceAI delta scores <0.1 and applies only to synonymous or intronic variants.

Evidence considered

• PIK3R1 VCEP (cspec, v1.0) BP4 rule: 'BP4 is met by a missense variant with both REVEL less than or equal to 0.290 and CADD less than or equal to 21.5' and 'BP4 is met by synonymous or intronic variants not predicted to impact splicing by SpliceAI.'
• REVEL score for this variant is 0.778 (source_registry: revel), which exceeds the 0.290 BP4 ceiling, definitively failing the missense BP4 clause independent of CADD.
• Variant is missense (NP_852664.1:p.(Gly376Arg)), not synonymous or intronic, so the VCEP's splice-based BP4 clause does not apply to it.

Sources

Reasoning

The PIK3R1 Antibody Deficiencies VCEP specification (cspec) defines PS1 for missense variants at Strong/Moderate strength when another variant causing the identical amino acid change (p.Gly376Arg via a different nucleotide substitution) was classified Pathogenic or Likely Pathogenic by VCEP standards without using PS1 itself, provided splicing predictions are equivalent for WT and both mutant alleles. No comparator variant encoding the same p.(Gly376Arg) change via an alternate codon substitution was identified in the case bundle; the automated PM5/PS1 comparator search returned zero same-residue candidates due to a query failure (HTTP 429 Too Many Requests), so this artifact cannot be used to rule PS1 in or out.

Rule

cspec_PS1_PIK3R1

Evidence considered

• PIK3R1 VCEP (cspec, ClinGen Antibody Deficiencies EP) PS1 rule: apply Strong when another missense variant causing the same amino acid change was classified Pathogenic by VCEP standards without using PS1; apply Moderate for a Likely Pathogenic comparator; splicing predictions must remain unchanged for WT and both mutant alleles.
• The pm5_candidates.json artifact (same query infrastructure used for same-residue/same-change comparator harvesting) reports found=false, same_residue_candidates=0, and an HTTP 429 Too Many Requests error, indicating the comparator search for this codon did not complete successfully.

Still missing

• No confirmed search for another nucleotide substitution producing the identical p.(Gly376Arg) change and its VCEP classification status; the automated comparator harvest (pm5_candidates.json) failed with an HTTP 429 error before completing this lookup.
• SpliceAI equivalence between the variant of interest and any candidate comparator could not be evaluated since no comparator was found.

Sources

Reasoning

The PIK3R1 Antibody Deficiencies specification requires documentation that the variant is de novo in an affected proband with no family history; confirmed PS2 additionally requires both maternity and paternity. The case bundle contains no proband phenotype score, parental genotypes, parentage confirmation, or family-history information, so de novo evidence cannot be assessed.

Rule

PIK3R1 VCEP PS2: both maternity and paternity must be confirmed, with no family history of disease; use the VCEP/SVI de novo point framework based on phenotype consistency.

Evidence considered

• PIK3R1 Antibody Deficiencies VCEP PS2 requires both maternity and paternity to be confirmed for confirmed de novo evidence, and directs use of PS2 rather than PM6 for assumed de novo occurrences.
• The VCEP selects de novo evidence strength using the proband's PS4 phenotype score and PIK3CD testing status; neither is supplied for this case.

Still missing

• Proband phenotype and PIK3R1-related disease assessment, including the VCEP PS4 phenotype-point score.
• Targeted parental testing results for NM_181523.2:c.1126G>A.
• Confirmation of both maternity and paternity if the variant is absent from both parents.
• Family-history information and PIK3CD testing status needed to select the VCEP phenotypic-consistency category and de novo point level.

Sources

Reasoning

The PIK3R1 Antibody Deficiencies VCEP specification permits PS3 (Strong/Moderate/Supporting depending on OddsPath or control count) for abnormal results in approved in vitro assays: lipid kinase activity (PMID:28167755), AKT kinase activity (PMID:23810379, PMID:25133428, PMID:25488983, PMID:28167755), protein binding (PMID:25488983), conformational dynamics (PMID:28167755), and the pS6/pAKT T-cell enrichment ratio assay (PMID:40543502, OddsPath 18.7 for Strong). However, the case-evidence bundle contains no data indicating that p.(Gly376Arg)/c.1126G>A was actually tested in any of these approved assays. The vcep_materials functional-assay spreadsheet (04_08_26_PIK3R1_Functional_Assays_PS3_BS3.xlsx) lists the general assay methodology and source PMIDs but no row confirming this specific variant was among the tested alleles. No animal model data for this variant were identified either.

Evidence considered

• cspec PS3 rule (PIK3R1 VCEP v1.0): approved assay classes are lipid kinase activity, AKT kinase activity, protein binding, conformational dynamics, and pS6/pAKT T-cell enrichment ratio; no assay result for this variant located in the case bundle.
• vcep_materials 04_08_26_PIK3R1_Functional_Assays_PS3_BS3.xlsx markdown extract lists assay methodology and source PMIDs (23810379, 25133428, 25488983, 40543502) but the captured excerpt does not list p.G376R/c.1126G>A as a tested variant.

Still missing

• No functional assay result (lipid kinase, AKT kinase, protein binding, conformational dynamics, or pS6/pAKT T-cell ratio) specific to p.(Gly376Arg)/c.1126G>A was found in the evidence bundle or vcep_materials extract.
• No animal model data for this variant identified.
• Reviewer should check the full 04_08_26_PIK3R1_Functional_Assays_PS3_BS3.xlsx variant list directly (only summary/header rows were captured in the bundle) to confirm whether p.G376R appears among tested variants.

Reasoning

The available evidence contains no germline PIK3R1-related immunodeficiency probands with c.1126G>A (p.Gly376Arg) whose phenotype score and PIK3CD testing status can be evaluated; PS4 cannot be assigned.

Rule

PIK3R1 VCEP PS4: count only non-PP4 probands meeting the Table 3 phenotype and PIK3CD-testing requirements; 1/2-3/>=4 qualifying probands correspond to Supporting/Moderate/Strong.

Evidence considered

• The PIK3R1 Antibody Deficiencies VCEP specification requires qualifying probands to score at least 6 phenotype points and to have PIK3CD adequately excluded; one, two to three, and four or more qualifying probands support PS4_Supporting, PS4_Moderate, and PS4_Strong, respectively.
• The exact ClinVar record is a somatic submission for vascular malformations/overgrowth and contains no germline immunodeficiency proband-level phenotype or PIK3CD exclusion data.

Still missing

• Variant-specific germline proband reports with sufficient clinical findings to calculate the VCEP Table 3 phenotype score.
• For each candidate proband, primary immunodeficiency/antibody panel or exome/genome results demonstrating no pathogenic or likely pathogenic PIK3CD variant; if no such testing was performed, documentation of a phenotype score of at least 10.
• At least one qualifying proband for PS4_Supporting, two to three for PS4_Moderate, or four or more for PS4_Strong; probands used for PP4 must be excluded from PS4.

Sources

Reasoning

The PIK3R1 VCEP defines PM5 (Moderate) for a missense variant when a different amino acid change at the same codon (376) was classified Pathogenic by VCEP standards without using PM5, with the variant of interest required to have a higher Grantham score than the comparator, no benign variants known at the codon, and SpliceAI delta scores <0.2 for both variants; PM5_Supporting applies analogously for a Likely Pathogenic comparator. The dedicated same-codon comparator search (pm5_candidates.json) found zero same-residue candidates and explicitly failed with an HTTP 429 error, so no comparator variant at codon 376 with a different amino acid substitution and its VCEP classification could be identified or evaluated against the Grantham/benign-variant/splicing gating conditions required by this rule.

Rule

cspec_PM5_PIK3R1

Evidence considered

• PIK3R1 VCEP PM5 rule: Moderate strength requires a different amino acid change at the same codon classified Pathogenic by VCEP standards without PM5, variant of interest must have a higher Grantham score than the comparator, no benign variants known at the codon, and SpliceAI delta scores <0.2 for both the variant of interest and comparator; Supporting strength applies analogously for a Likely Pathogenic comparator.
• pm5_candidates.json reports found=false, same_residue_candidates=0, expert_panel_plp_candidates=0, with an HTTP 429 Too Many Requests error recorded, indicating the same-codon (residue 376) comparator harvest for alternate amino acid changes did not complete.

Still missing

• The automated same-codon comparator search (pm5_candidates.json) did not complete due to an HTTP 429 (rate limit) error and returned zero candidates; a manual ClinVar/literature search for other PIK3R1 codon 376 missense variants with VCEP classifications and their Grantham scores is required before PM5 can be adjudicated.
• Whether any benign variant has been reported at codon 376 (which would preclude PM5 use per the VCEP rule) has not been confirmed.

Sources

Reasoning

No informative relatives, genotypes, phenotypes, or meioses are documented in the case bundle. Therefore, co-segregation cannot be counted.

Rule

PIK3R1 VCEP PP1: affected relatives must each reach at least 6 PS4 phenotype points; Supporting requires 1 meiosis, Moderate 2 meioses, and Strong at least 4 meioses; do not apply if BA1 or BS1 is met.

Evidence considered

• PIK3R1 VCEP PP1 requires each counted affected family member to have at least 6 PS4 phenotype points and prohibits PP1 if BA1 or BS1 is met.
• VCEP PP1 thresholds are Supporting for one informative meiosis, Moderate for two, and Strong for at least four meioses. No family segregation data are supplied.

Still missing

• Pedigree with relationship structure and meiosis count.
• Targeted PIK3R1 testing in informative affected relatives.
• Phenotype documentation showing each counted affected relative reaches at least 6 PIK3R1 VCEP PS4 phenotype points.
• Confirmation that population evidence does not meet BA1 or BS1 before applying PP1.

Sources

Reasoning

NM_181523.2:c.1126G>A (p.Gly376Arg) is a missense variant. The PIK3R1 Antibody Deficiencies VCEP specification defines PP3 (Supporting) via two independent clauses: (1) a missense clause requiring BOTH REVEL >=0.644 AND CADD >=26.0, or (2) a splice-impact clause (applicable to missense/synonymous/intronic variants outside +/-1,2 splice-consensus positions) requiring any SpliceAI delta score (donor gain/loss, acceptor gain/loss) >=0.2. The splice clause is not met: SpliceAI predicts no significant splice impact for this variant (max delta score = 0.00), so PP3 cannot be applied via the splice-impact path. For the missense clause, REVEL = 0.778, which exceeds the 0.644 threshold, but no CADD score for this variant is present anywhere in the case evidence bundle (prefetch.json, evidence.json). Per pipeline policy, only calibrated predictors with a value actually retrieved and a named governing threshold can be used; the VCEP explicitly requires CADD as a co-requisite with REVEL (an AND condition), and that value is unavailable, so the missense clause cannot be confirmed as met even though REVEL alone clears its half of the threshold. BayesDel (score 0.536814) is not part of the PIK3R1 VCEP's PP3 lookup (which specifies REVEL+CADD only) and, independent of that, this pipeline has no verified published calibration threshold for BayesDel, so it cannot substitute for the missing CADD value.

Rule

PIK3R1 VCEP (cspec v1.0) PP3: missense clause requires REVEL>=0.644 AND CADD>=26.0; splice clause requires any SpliceAI delta score >=0.2 for missense/synonymous/intronic variants outside +/-1,2 splice sites.

Evidence considered

• PIK3R1 VCEP (cspec, v1.0) PP3 rule: 'PP3 is met by a missense variant with both REVEL greater than or equal to 0.644 and CADD greater than or equal to 26.0.' and 'PP3 is met by missense, synonymous, or intronic variants outside the +/-1,2 dinucleotide positions that are predicted as damaging using SpliceAI (delta score for donor gain, donor loss, acceptor gain, or acceptor loss score greater than or equal to 0.2, PMID: 37352859).'
• REVEL score for this variant is 0.778 (source_registry: revel), which exceeds the VCEP's 0.644 REVEL threshold on its own, but CADD is required in conjunction and was not retrievable in this case's evidence bundle.
• SpliceAI evidence sentence: 'SpliceAI predicts no significant splice impact for this variant (max delta score = 0.00).' (source_registry: spliceai) -- rules out the splice-impact path for PP3.
• BayesDel score of 0.536814 was retrieved (source_registry: bayesdel) but is not used for PP3/BP4 because (a) the PIK3R1 VCEP PP3/BP4 lookup specifies REVEL+CADD, not BayesDel, and (b) this pipeline has no verified published calibration threshold for BayesDel.

Still missing

• CADD score for NM_181523.2:c.1126G>A / NC_000005.9:g.67589138G>A is not present in prefetch.json, evidence.json, or any other bundled source; it is required as a co-requisite with REVEL under the PIK3R1 VCEP PP3 missense rule and must be obtained to fully adjudicate PP3.

Sources

Reasoning

No individual with the germline PIK3R1-related phenotype is documented for this variant, so the required phenotype score and exclusion of an alternative PIK3CD molecular diagnosis cannot be assessed.

Rule

PIK3R1 VCEP PP4: phenotype score >=10 plus no alternative PIK3CD diagnosis; add patient-cell PI3K-delta hyperactivity for Moderate.

Evidence considered

• The PIK3R1 VCEP specification requires a phenotype score of at least 10 plus exclusion of an alternative PIK3CD diagnosis for PP4; PP4_Moderate additionally requires abnormal disease-relevant PI3K-delta pathway activity in patient cells.
• The exact ClinVar record supplies no patient phenotype, germline testing, or patient-cell pathway data relevant to PIK3R1-related immunodeficiency.

Still missing

• One variant-positive proband with clinical data sufficient to calculate a Table 3 phenotype score of at least 10.
• Genotyping showing no alternative pathogenic or likely pathogenic variant at PIK3CD.
• For PP4_Moderate, patient-cell evidence of abnormally elevated disease-relevant PI3K-delta pathway activity.

Sources

Reasoning

There is no exact-variant ClinVar expert-panel Pathogenic or Likely pathogenic classification. The available exact record contains only non-expert somatic submissions and therefore cannot support PP5.

Rule

Apply PP5 only to an exact-variant ClinVar expert-panel Pathogenic/Likely pathogenic assertion.

Evidence considered

• ClinVar variation 376064 exactly matches NM_181523.3:c.1126G>A (p.Gly376Arg), but the submission audit identifies zero expert-panel submissions. The listed submissions are somatic and non-expert, with classifications Pathogenic (no assertion criteria) and Likely oncogenic; neither is eligible for PP5.

Still missing

• An exact-variant ClinVar submission from an expert panel classifying the variant as Pathogenic or Likely pathogenic.

Sources

Reasoning

The PIK3R1 VCEP specification permits BS3 (Strong/Moderate/Supporting) for a non-damaging result in the same set of approved in vitro assays (lipid kinase, AKT kinase, protein binding, conformational dynamics, or the pS6/pAKT T-cell ratio assay PMID:40543502). No functional assay data — damaging or non-damaging — specific to p.(Gly376Arg)/c.1126G>A was identified in the case-evidence bundle, so a benign functional result cannot be confirmed.

Evidence considered

• cspec BS3 rule (PIK3R1 VCEP v1.0): same approved assay classes as PS3; no non-damaging result for this variant located in the case bundle.

Still missing

• No functional assay result specific to p.(Gly376Arg)/c.1126G>A was found supporting a non-damaging/normal readout.
• No animal model data for this variant identified.

Reasoning

No affected relatives reported as lacking NM_181523.2:c.1126G>A are documented in the case bundle. Non-segregation evidence therefore cannot be assessed.

Rule

PIK3R1 VCEP BS4: each affected non-carrier must reach at least 6 PS4 phenotype points; Supporting requires 1 affected non-carrier and Strong requires at least 2.

Evidence considered

• PIK3R1 VCEP BS4 requires affected non-carriers to reach at least 6 PS4 phenotype points; one such relative supports BS4_Supporting and at least two support BS4 at Strong strength.
• No affected-relative genotype or phenotype data are supplied.

Still missing

• Pedigree and targeted testing of affected relatives who do not carry the variant.
• Phenotype documentation showing each affected non-carrier reaches at least 6 PIK3R1 VCEP PS4 phenotype points.

Sources

Reasoning

No case evidence documents this variant in individuals whose phenotype has an independently established alternative molecular cause; BP5 cannot be assigned.

Rule

PIK3R1 VCEP BP5: at least two cases with an alternative molecular basis for disease are required.

Evidence considered

• The PIK3R1 VCEP specification requires at least two cases with an alternative molecular basis for disease before applying BP5.
• The available ClinVar evidence is limited to somatic cancer/overgrowth submissions and does not provide two cases with an independently established alternative molecular diagnosis explaining a germline PIK3R1-related phenotype.

Still missing

• At least two variant-positive cases with a documented alternative molecular basis that explains the disease phenotype.

Sources

Reasoning

There is no exact-variant ClinVar expert-panel Benign or Likely benign classification. Non-expert and aggregate ClinVar assertions cannot support BP6.

Rule

Apply BP6 only to an exact-variant ClinVar expert-panel Benign/Likely benign assertion.

Evidence considered

• ClinVar variation 376064 exactly matches NM_181523.3:c.1126G>A (p.Gly376Arg), but the submission audit identifies zero expert-panel submissions and no Benign/Likely benign assertion.

Still missing

• An exact-variant ClinVar submission from an expert panel classifying the variant as Benign or Likely benign.

Sources