About ATM
No gene-level summary was recorded for this case.
Previous versions
Variant interpretation report
ATM c.3080A>G, p.(His1027Arg)
Classification
Uncertain significance
Variant description:
No variant summary was recorded for this case.
Five-step scale from benign to pathogenic. This variant is uncertain. Select a step to read what it means.
Uncertain.The recorded classification is uncertain significance.
In plain language
Gene-level background from the recorded interpretation.
Words with a dotted underline have a short definition. Hover or tap them.
LYFE Sciences is an AI system, and it can make mistakes.
Criteria may be applied incorrectly, sources may be misread, and a
confident-looking classification can still be wrong.
Double-check every criterion and its underlying evidence
before relying on any call.
01The reasoning
How it was classified
The recorded framework is ClinGen ATM VCEP v1.5. The criteria and strengths shown below come from the saved assessment; this page does not recalculate the classification.
Recorded applied criteria
The applied criteria and their strengths are shown as assessed under the stated framework.
Pathogenic evidence criteria: 1
Benign evidence criteria: 0
Pathogenic criteria applied · 1
PM2Supporting
PM2 · supporting
PM2 (Supporting) is met: the variant is extremely rare. gnomAD v4.1 total allele frequency is 0.00062% (10/1,610,692 alleles, no homozygotes) with a grpmax filtering allele frequency of 0.00036%, both at or below the specification's 0.001% threshold; rarity is corroborated by gnomAD v2.1 (0.00040%) and absence from gnomAD-Canada.
Benign criteria applied · 0
No benign criterion marked as applied in the recorded assessment.
Recorded classification: Uncertain significance
The primary authority was the ClinGen Hereditary Breast, Ovarian and Pancreatic Cancer Expert Panel specification to the ACMG/AMP guidelines for ATM, version 1.5. Exactly one criterion was met: PM2 at supporting strength, because the variant is extremely rare in population databases (gnomAD v4.1 total allele frequency 0.00062% and grpmax FAF 0.00036%, both at or below the 0.001% threshold, with no homozygotes). All other criteria were not met, not applicable under the specification, or could not be assessed for lack of evidence. Because PM2 supporting alone matches no Pathogenic, Likely Pathogenic, Benign, Likely Benign, or conflicting-evidence rule in the specification, the variant remains a Variant of Uncertain Significance.
02The sources
Evidence at a glance
Evidence summaries and captured source records, where available. Missing entries should not be interpreted as negative findings.
gnomAD v4.1
gnomAD v2.1
Population frequency
10 / 1,610,692 alleles
gnomAD v4.1: 10 / 1,610,692 alleles. Recorded AF 6.20851e-06. Recorded homozygote count 0.
gnomAD v2.1
This variant is present in gnomAD v2.1 (AF= 3.98159e-06; MAF= 0.00040%, 1/251156 alleles, homozygotes = 0) and has highest observed frequency in the European (non-Finnish) population (AF= 8.80514e-06; MAF= 0.00088%, 1/113570 alleles, homozygotes = 0).
gnomAD-Canada
Absent from gnomAD-Canada v1.0.
SpliceAI score
0.050 maximum delta
Recorded SpliceAI maximum delta score: 0.050. Refer to the recorded PP3/BP4 rationale for interpretation.
No listed criterion applied
SpliceAI ↗
ClinVar submissions
9 reported laboratory submissions
9 Uncertain significance. ClinVar API review status: criteria provided, multiple submitters, no conflicts.
Context only
ClinVar ↗
Functional evidence
Statement recorded
OncoKB did not identify variant-specific reviewed functional evidence for this variant; gene-level curated context is available for reviewer follow-up. ATM, a kinase involved in the DNA damage response, is mutated in various solid and hematologic malignancies.
No listed criterion applied
OncoKB ↗
COSMIC
Cancer Hotspots
Somatic evidence
No matching record reported
This variant has not previously been reported in somatic cancers (COSMIC).
Cancer Hotspots
This variant does not lie in a statistically significant hotspot.
Context only
For specialists
Technical detail
ACMG/AMP criteria under ClinGen ATM VCEP v1.5, with the data and sources behind each one.
28 criteria with recorded statuses: 1 met · 6 not met · 5 not assessed · 16 not applicable. Open any row for the full reasoning, the rule and the evidence behind it.
Met
Not met
Not assessed
Not applicable
Governing framework: ClinGen HBOP ATM VCEP v1.5 (cspec doc 639508985, framework_mode=vcep). All three criteria in this group were adjudicated under the VCEP specification; generic ACMG/AMP was not used because an applicable specification exists.
PVS1 not_applicable: the variant is a missense substitution (p.His1027Arg), not a null variant (no nonsense, frameshift, canonical splice, initiation-codon, or exon-deletion consequence), and SpliceAI (max delta 0.05) predicts no splice impact that could create an out-of-frame transcript. The ATM PVS1 decision tree therefore cannot be entered.
PM4 not_applicable: the VCEP restricts PM4 to stop-loss variants only; a missense SNV produces no protein length change and does not affect the stop codon.
BP3 not_applicable: the VCEP declares BP3 not applicable (Do not use) for ATM, and the variant is not an in-frame indel in a repeat region in any case.
No literature was required: all three determinations follow deterministically from the variant consequence (missense SNV) and the VCEP rule text. No other group's evidence or any external final classification was used to reach these decisions.
Governing framework: ClinGen HBOP ATM VCEP v1.5 (cspec doc 639508985, framework_mode=vcep, precedence_rank=1). All five criteria in this group were adjudicated under the VCEP specification; generic ACMG/AMP (PMID 25741868) was used only as baseline definitional reference where the VCEP rule text points to it, never as an independent framework.
PS1 not_met: the standard missense PS1 path requires an established P/LP variant producing the same amino acid change (p.His1027Arg). No such comparator exists - the only codon-1027 ClinVar record is the VUA itself (VUS, 9 labs, no expert-panel submissions) - and no alternate nucleotide change (c.3082A>G) or alternate pathogenic missense at residue 1027 appears in any consulted source. SpliceAI (0.05) rules out splicing for the VUA, satisfying the VCEP missense prerequisite, but the comparator is absent.
PM1 not_applicable: the ATM VCEP v1.5 declares PM1 Not Applicable; CancerHotspots shows ATM H1027 is not a hotspot; residue 1027 lies outside the FAT/PI3-K/FATC contiguous-domain region that is the only domain context the spec discusses (PVS1 guide, PMID 28508083).
PM5 not_applicable under the governing rule: the cspec PM5 rule covers frameshifting/truncating and NMD-prone splice PTC variants upstream of p.Arg3047 only, and the VUA is a missense. Under a classic same-residue reading the result would still be not_met (no alternate P/LP missense at residue 1027 documented; pm5_candidates found 0 candidates). The anomalous rule text is flagged for human review (needs_human_review=true on PM5).
PP2 and BP1 not_applicable: the ATM VCEP v1.5 declares both Not Applicable; ATM's established LOF disease mechanism (pvs1_gene_context lof_mechanism_supported=true) is consistent with both declarations (PP2's missense-mechanism premise does not hold; BP1's LOF-not-a-mechanism premise is false).
Consistency note: the same-residue negative (no established P/LP at codon/residue 1027) is the single decisive fact for both PS1 (same amino acid change) and PM5 (alternate change at same residue); it was established from ClinVar (only VUS for this variant), pm5_candidates (0 candidates), OncoKB (no variant-specific evidence), COSMIC (absent), CancerHotspots (not a hotspot), and full-text screening of the ClinVar-cited papers (no residue-1027 content). No criterion was inferred from any other source's final classification.
Governing framework is the ClinGen HBOP ATM VCEP v1.5 (CSPEC found, precedence rank 1); generic ACMG/AMP fallback was not needed.
PS2 and PM6 are Not Applicable per VCEP: ATM A-T is autosomal recessive and informative de novo occurrences are not expected/weighted for AR disease; no de novo or parental-testing evidence exists in the case anyway.
PP1 is Applicable (AR condition: segregation in 1/2/>=3 affected relatives = Supporting/Moderate/Strong, with affected relatives required to carry both variants identified in the proband), but the case bundle contains no segregation/family data at all, so PP1 is not_assessed pending family studies.
BS4 is Not Applicable per VCEP: informative lack of co-segregation in A-T families is too rare to weight (may only feed BP2 for biallelic unaffected individuals in an A-T family, which is outside this group).
This group contributes no pathogenic or benign evidence to the classification for NM_000051.4:c.3080A>G (p.His1027Arg).
Framework: ClinGen HBOP ATM VCEP v1.5 governs PS3/BS3 for this case (framework_mode=vcep, cspec version 1.5); generic ACMG/AMP 2015 (PMID:25741868) applies only as background and was not used to assign strength.
VCEP PS3/BS3 rules are assay-gated: strength may only come from the approved assays in clingen_hbop_atm_supplementary_tables_1_and_2_v1.xlsx (kinase activity: Mitui 2009/PMID 18634022, Barone 2009/PMID 19431188, Scott 2002/PMID 11805335, each PS3/BS3_Supporting; radiosensitivity: Mitui 2009 only, PS3_Moderate/BS3_Supporting, combinable only with a kinase-activity result), and only after confirming the variant under assessment appears in one of those papers.
No evidence that c.3080A>G (p.His1027Arg) was tested in any approved assay: the three assay papers are absent from source_registry.json and the case bundle, and the approved-assay table is a methodology/calibration table with no per-variant results.
The only per-variant functional-source data point, Sun et al. 2025 (PMID 40580951) Table S1, reports c.3080A>G as Classification='Functional' but with DeepATM_predicted='Yes' and Confidence='Computational prediction' - a deep-learning extrapolation, not a direct prime-editing measurement; per the VCEP file guidance this must not be presented as functional evidence for PS3/BS3. The paper is also not one of the three VCEP-approved assays and postdates ATM VCEP v1.5.
Conclusion: both PS3 and BS3 are not_assessed due to absence of variant-level validated functional assay evidence; flagged needs_human_review because the sole variant-level functional-source entry is a computational prediction and no approved-assay result exists to confirm or refute ATM functional impact of p.His1027Arg.
PS4 is not met: no case-control study reports c.3080A>G (p.His1027Arg); the ATM case-control study reviewed (PMID:28779002) was restricted to protein-truncating variants, and the case series of Tung et al. (PMID:25186627) and Lerner-Ellis et al. (PMID:32885271) do not report the variant and lack control arms.
PP4 is not applicable under the ClinGen HBOP ATM VCEP v1.5 specification, which explicitly lists PP4 as Not Applicable.
PP5 and BP6 are not applicable: no ClinVar expert-panel classification exists for this exact variant (ClinVar VCV000188327 is Uncertain significance from 9 clinical laboratory submissions with 0 expert panel submissions), and the ATM HBOP VCEP v1.5 marks both criteria Not Applicable.
BP5 is not applicable under the ATM HBOP VCEP v1.5 specification, and no alternate molecular basis of disease is documented for this variant.
Governing framework: ClinGen HBOP ATM VCEP v1.5 (cspec doc 639508985; vcep_atm_pm3_bp2_1_5) applies — ATM is a recessive A-T gene and the VCEP provides a points-based PM3/BP2 table; generic ACMG/AMP was not used for this group.
PM3: not_assessed. The >0.01% population-frequency prerequisite is satisfied (gnomAD v4.1 AF 6.21e-06, grpmax FAF 3.6e-06; v2.1 AF 3.98e-06), but zero A-T proband observations of c.3080A>G exist in the case evidence (no literature mention, no ClinVar zygosity/phase data, 0 gnomAD homozygotes), so 0 points; the required A-T proband source type was never retrieved, so a confident met/not_met call cannot be made.
BP2: not_assessed. No unaffected (non-A-T, >=18y) carrier of c.3080A>G with a P/LP ATM variant in trans, and no cis-phase observation, exists in the case evidence; 0 negative points; the required unaffected-carrier/phase data was never retrieved.
gnomAD v2.1 (1/251,156 alleles) and v4.1 (10/1,610,692 alleles) both report 0 homozygotes — no recessive homozygous signal — consistent with the absence of PM3 proband data but not by itself probative.
None of the 6 available publication full texts (PMID:17508274, 25186627, 32885271, 25741868, 28779002, 34242744) mention c.3080A>G or p.His1027Arg; no publication is cited for PM3 or BP2, and no criterion was inferred from any other source's final classification (ClinVar submissions are all VUS and were not used as evidence of pathogenicity or benignity).
This group contributes no met criteria; the identified gaps (A-T proband observations with phase for PM3; unaffected-carrier-in-trans or cis-phase data for BP2) require dedicated retrieval before these criteria can be scored.
Variant NM_000051.4:c.3080A>G is a missense change (p.His1027Arg, exon 21). Computational-evidence evaluation for PP3/BP4 uses the VCEP-calibrated predictors: REVEL for the missense/protein path and SpliceAI for the splice path; BP7 is scoped to synonymous and deep intronic variants only.
PP3 not met on both VCEP sub-paths: REVEL 0.349 <= 0.7333 (missense path) and SpliceAI max delta 0.05 < 0.2 (splice path). Secondary predictors (BayesDel -0.237, AlphaMissense 0.228, EVE 0.266, CADD 20.3) do not indicate a damaging effect.
BP4 not met: REVEL 0.349 > 0.249 on the missense sub-path (only sub-path allowed for a missense variant under the lab's BP4 curator override). Explicitly noted: the literal ATM VCEP v1.5 text would permit BP4 via the splicing sub-path (SpliceAI 0.05 <= 0.1), but the override restricts missense variants to the REVEL sub-path and treats the 0.249-0.7333 gray zone as not_met - this override changes the outcome from the literal VCEP text.
BP7 not applicable: variant is missense, not synonymous or deep intronic.
Sun et al. 2025 Table S1 lists c.3080A>G as Classification 'Functional' but DeepATM_predicted=Yes / Confidence='Computational prediction' (trained-model extrapolation, not a direct functional measurement); per VCEP index guidance this is at most weak computational corroboration for PP3/BP4 and was not used as functional evidence or to override the VCEP-calibrated REVEL/SpliceAI thresholds.
No literature source mentions the variant: full-text searches of the PP3-indexed candidate papers (PMID:25186627, PMID:32885271) returned no occurrence of c.3080A>G, 3080A, His1027, or H1027; no publication was cited for PP3/BP4/BP7.
Governing framework: ClinGen HBOP ATM VCEP v1.5 (CSPEC doc 639508985), precedence_rank 1, framework_mode=vcep. Generic ACMG/AMP is not used for population criteria because this gene-specific specification exists and covers BA1, BS1, BS2 and PM2.
BA1 not met: gnomAD v4.1 grpmax FAF 3.6e-06 (0.00036%) is far below the VCEP BA1 threshold of >0.5%.
BS1 not met: gnomAD v4.1 grpmax FAF 3.6e-06 (0.00036%) is far below the VCEP BS1 threshold of >0.05%; highest subpopulation AF is 0.0016%.
BS2 not applicable: ATM VCEP v1.5 explicitly lists BS2 as Not Applicable.
PM2_supporting met: gnomAD v4.1 total AF 0.00062% and grpmax FAF 0.00036% are both <= 0.001%; n=1 in the 'Remaining individuals' subpopulation satisfies the VCEP's n=1 rarity clause; 0 homozygotes across gnomAD v2.1/v4.1 and gnomAD-Canada.
No population-frequency or allele-count information for this variant was found in any retrieved publication (full-text grep for c.3080A>G / p.His1027Arg returned no hits); population assessment rests on gnomAD v2.1, gnomAD v4.1, gnomAD-Canada v1.0, and the VCEP framework itself.
No criteria match.
Met · 1
PM2 SupportingPM2 (Supporting) is met: the variant is extremely rare. gnomAD v4.1 total allele frequency is 0.00062% (10/1,610,692 alleles, no homozygotes) with a grpmax filtering allele frequency of 0.00036%, both at or below the specification's 0.001% threshold; rarity is corroborated by gnomAD v2.1 (0.00040%) and absence from gnomAD-Canada. Details ▾Hide ▴
Reasoning
ATM VCEP v1.5 defines PM2_supporting as Frequency <= 0.001% in the gnomAD v4 dataset, with an explicit note that if n=1 in a single subpopulation, that is sufficiently rare and PM2_supporting applies. gnomAD v4.1 total AF is 6.21e-06 (0.00062%; 10/1,610,692 alleles, 0 homozygotes) and grpmax FAF is 3.6e-06 (0.00036%) - both at or below the 0.001% threshold. The variant also satisfies the VCEP's n=1 single-subpopulation clause ('Remaining individuals', 1/62,392 alleles; 0 homozygotes). Rarity is corroborated across datasets: gnomAD v2.1 AF 3.98e-06 (0.00040%; 1/251,156 alleles, 0 homozygotes) and absence from gnomAD-Canada v1.0. Note: the highest subpopulation AF ('Remaining individuals', 0.0016%) nominally exceeds 0.001% when expressed as a percentage of that small subpopulation, but n=1 there and total/grpmax AFs are below threshold, matching the VCEP's own rarity guidance; no homozygotes in any dataset. PM2_supporting is met.
Rule
Frequency <=0.001% in gnomAD v4 dataset; if n=1 in a single sub population, that is sufficiently rare and PM2_supporting would apply
Evidence considered
- gnomAD v4.1 total AF 6.21e-06 (0.00062%, 10/1,610,692, 0 homozygotes)
- gnomAD v4.1 grpmax FAF 3.6e-06 (0.00036%)
- gnomAD v4.1 'Remaining individuals' subpopulation n=1/62,392 (satisfies VCEP n=1 clause)
- gnomAD v2.1 AF 3.98e-06 (0.00040%, 1/251,156, 0 homozygotes)
- gnomAD-Canada v1.0: absent
Sources
Not met · 6 · evaluated, threshold not reached
PS1 PS1 requires the same amino acid change to have been established as pathogenic (or likely pathogenic) in another variant. No such comparator exists: the only ClinVar record for a change at codon 1027 is this exact variant itself, classified Uncertain significance by 9 clinical laboratories with no expert-panel submissions, so PS1 is not met. Details ▾Hide ▴
Reasoning
ATM VCEP v1.5 PS1 (standard missense path) requires the same amino acid change as an established pathogenic (P) or likely pathogenic (LP) variant, with splicing ruled out for both alterations. The VUA is NM_000051.4:c.3080A>G = NP_000042.3:p.(His1027Arg). The only ClinVar record for a codon-1027 change is this exact variant itself (VCV000188327), classified Uncertain significance by 9 clinical laboratories with no expert-panel submissions (ClinVar expert_panels = {}), so there is no established P/LP variant producing p.His1027Arg (e.g., via the alternative nucleotide change c.3082A>G) in any consulted source. The VCEP splicing prerequisite is satisfied for the VUA only (SpliceAI max delta 0.05, no splice impact), but the essential same-amino-acid pathogenic comparator is absent, so PS1 cannot be applied at any strength. The generic ACMG/AMP definition (PMID 25741868: a different nucleotide change resulting in the same amino acid can be assumed pathogenic once one missense is established) confirms the comparator requirement; that comparator does not exist here. A dedicated ClinVar/LOVD search for c.3082A>G was not executed in this case (pm5_candidates.clinvar_queries is empty), so the negative rests on the absence of any such variant in the collected evidence bundle.
Rule
ClinGen HBOP ATM VCEP v1.5 PS1 (Pathogenic Strong): use for missense changes as long as splicing is ruled-out for both alterations - standard missense path = same amino acid change as an established P/LP variant; splicing-region PS1 table for splice variants with similar predicted/observed splice defect
Evidence considered
- ClinVar VCV000188327: c.3080A>G (p.His1027Arg) = Uncertain significance, 9 laboratories, no conflicts, expert_panels={} (no expert-panel classification of this or any codon-1027 variant)
- pm5_candidates.json: 0 same-residue comparator candidates seen; classic same-residue PM5 search not eligible (PM5 treated as truncation-cutoff rule), clinvar_queries empty
- ATM VCEP v1.5 cspec PS1 rule: 'Use for missense changes as long as splicing is ruled-out for both alterations' - standard path requires same amino acid change as an established P/LP variant
- ATM PS1 v1.5 guide (index guidance): standard missense PS1 = 'same amino acid change as an established P/LP variant'; splicing-region table not applicable (variant is exonic missense with no splice prediction)
- SpliceAI: max delta score 0.05 - splicing ruled out for the VUA (prerequisite met), but no pathogenic comparator exists to compare against
- PMID 25741868 (ACMG/AMP 2015): PS1 defined as same amino acid change via a different nucleotide change, assumed pathogenic only when one missense at that change is already established pathogenic
Still missing
- No established P/LP variant with the same amino acid change (p.His1027Arg via c.3082A>G) was identified in the evidence bundle; a dedicated ClinVar same-codon search was not run (pm5_candidates.clinvar_queries empty) - optional reviewer confirmation before final sign-out
Sources
clinvar ↗pm5_candidatescspec ↗vcep_atm_ps1_1_5spliceai ↗PMID:25741868 ↗PMID:25186627 ↗PMID:32885271 ↗
PS4 No case-control or cohort enrichment data exist for this variant. The available ATM case-control study covered only protein-truncating variants, and the reviewed case series do not report this variant, so PS4 is not met. Details ▾Hide ▴
Reasoning
No case-control or cohort enrichment data exist for NM_000051.4:c.3080A>G (p.His1027Arg). The available ATM case-control study (PMID:28779002; 13,087 breast cancer cases vs 5,488 controls) was restricted to rare protein-truncating variants, so this missense variant is outside its scope, and the variant is not reported in it. The case series of Tung et al. (PMID:25186627; 2,158 breast cancer patients, no control arm) and Lerner-Ellis et al. (PMID:32885271; 3,251 HBOC patients, no control arm) also do not report the variant. With no variant-specific case-control data satisfying the VCEP PS4 threshold (p<=0.05 AND OR/HR/RR >=2 OR lower 95% CI >=1.5), PS4 is not met.
Rule
ATM HBOP VCEP v1.5 PS4: Case-control studies; p-value <=0.05 AND (Odds ratio, hazard ratio, or relative risk >=2 OR lower 95% CI >=1.5).
Evidence considered
- PMID:28779002 (Decker 2017): population-based case-control (13,087 BC cases vs 5,488 controls) of rare protein-truncating variants in ATM/CHEK2/PALB2/XRCC2; c.3080A>G is a missense variant and is not reported.
- PMID:25186627 (Tung 2015): cross-sectional 25-gene panel study in 2,158 breast cancer patients; no control arm; c.3080A>G not reported.
- PMID:32885271 (Lerner-Ellis 2021): multigene panel testing in 3,251 HBOC cases; diagnostic yield study with no control arm; c.3080A>G not reported.
Still missing
- No case-control study reporting c.3080A>G (p.His1027Arg) with odds ratio or confidence interval is available; variant-specific enrichment cannot be established.
Sources
PP3 PP3 is not met on either computational path calibrated by the ATM specification: REVEL is 0.349, well below the >0.7333 pathogenic threshold, and SpliceAI max delta is 0.05, below the >=0.2 threshold. Secondary predictors (BayesDel, AlphaMissense, EVE, CADD) also do not indicate a damaging effect. Details ▾Hide ▴
Reasoning
Missense variant NM_000051.4:c.3080A>G (p.His1027Arg). ATM HBOP VCEP v1.5 PP3 (Pathogenic Supporting) has two sub-paths. Missense sub-path: REVEL > 0.7333 required; REVEL = 0.349, well below threshold -> not met. Splicing sub-path (which the VCEP applies to missense variants as well): SpliceAI >= 0.2 required; SpliceAI max delta = 0.05 (DS_AG 0.05, DS_DG 0.02) -> not met. Both VCEP-calibrated computational paths fail, so PP3 is not met on either mechanism. Secondary in-silico predictors (not VCEP-calibrated): BayesDel noAF -0.237229, AlphaMissense 0.2278, EVE 0.266, CADD phred 20.3 — none indicate damaging effect. Sun et al. 2025 Table S1 row for c.3080A>G lists Classification 'Functional' but with DeepATM_predicted=Yes and Confidence='Computational prediction', i.e. a trained-model extrapolation, not a functional measurement; per VCEP index guidance it can only serve as weak computational corroboration and does not support PP3. Literature papers indexed as PP3 candidates (PMID:25186627, PMID:32885271) were full-text searched and do not mention this variant. Decision is deterministic from VCEP thresholds.
Rule
PP3 (Pathogenic Supporting): Missense: REVEL > .7333; Splicing: Predicted impact via splicing (SpliceAI >= 0.2) for silent, missense/in-frame and for intronic variants outside of donor and acceptor 1,2 sites.
Evidence considered
- REVEL score 0.349 (missense path; VCEP PP3 threshold >0.7333 not met)
- SpliceAI max delta 0.05, DS_AG 0.05, DS_DG 0.02 (splice path; VCEP PP3 threshold >=0.2 not met)
- ATM HBOP VCEP v1.5 PP3 rule from cspec (rule text quoted in 'rule')
- Sun et al. 2025 (PMID 40580951) Table S1 row c.3080A>G: REVEL 0.349, AlphaMissense 0.2278, EVE_scores_ASM 0.266, CADD.phred 20.3, boostDM 0.081, Classification 'Functional', DeepATM_predicted=Yes (computational prediction) - secondary corroboration only
- BayesDel noAF score -0.237229 (not VCEP-calibrated; benign-leaning, no PP3 support)
Still missing
- No RNA/minigene splice assay data for this variant (not required for the PP3 computational sub-paths, but would be the only route to a splice-impact finding given SpliceAI < 0.2).
Sources
BA1 BA1 is not met: it requires a grpmax filtering allele frequency above 0.5%, and the observed grpmax FAF is 0.00036% - more than a thousand-fold below the threshold, with no homozygotes and no ancestry-specific enrichment. Details ▾Hide ▴
Reasoning
ATM VCEP v1.5 (ClinGen HBOP, CSPEC doc 639508985) defines BA1 as Grpmax Filtering AF > 0.5% in the gnomAD v4 dataset. The variant is present in gnomAD v4.1 at total AF 6.21e-06 (0.00062%; 10/1,610,692 alleles, 0 homozygotes) with a grpmax FAF of 3.6e-06 (0.00036%). The grpmax FAF is more than three orders of magnitude below the 0.5% BA1 threshold (0.00036% vs 0.5%), so BA1 is not met. Highest observed subpopulation AF is 1.60e-05 (0.0016%; 'Remaining individuals', 1/62,392 alleles), still far below the BA1 threshold. gnomAD v2.1 corroborates rarity (AF 3.98e-06, 1/251,156 alleles) and gnomAD-Canada v1.0 reports absence; no threshold artifact (no ancestry-specific enrichment, no homozygotes) is present.
Rule
Grpmax Filtering AF >0.5% in gnomAD v4 dataset
Evidence considered
- gnomAD v4.1 grpmax FAF 3.6e-06 (0.00036%), total AF 6.21e-06 (0.00062%), 10/1,610,692 alleles, 0 homozygotes
- gnomAD v4.1 highest subpopulation AF 1.60e-05 ('Remaining individuals', 1/62,392)
- gnomAD v2.1 AF 3.98e-06 (0.00040%, 1/251,156 alleles)
- gnomAD-Canada v1.0: absent
Sources
BS1 BS1 is not met: it requires a grpmax filtering allele frequency above 0.05%, and the observed grpmax FAF of 0.00036% is roughly 140-fold lower. No homozygotes are observed in any dataset and the variant is absent from gnomAD-Canada, consistent with a rare variant rather than a common benign polymorphism. Details ▾Hide ▴
Reasoning
ATM VCEP v1.5 defines BS1 as Grpmax Filtering AF > 0.05% in the gnomAD v4 dataset. Observed grpmax FAF in gnomAD v4.1 is 3.6e-06 (0.00036%), and the highest subpopulation AF is 1.60e-05 (0.0016%; 'Remaining individuals', n=1/62,392). Both are well below the 0.05% (5e-04) BS1 threshold (0.00036% vs 0.05%, ~140-fold lower on the grpmax metric), so BS1 is not met. No homozygotes are observed in any dataset, and the variant is absent from gnomAD-Canada v1.0, consistent with a rare variant rather than a common benign polymorphism.
Rule
Grpmax Filtering AF >0.05% in gnomAD v4 dataset
Evidence considered
- gnomAD v4.1 grpmax FAF 3.6e-06 (0.00036%) vs BS1 threshold 0.05%
- gnomAD v4.1 highest subpopulation AF 1.60e-05 (0.0016%, 1/62,392)
- gnomAD v4.1 total AF 6.21e-06 (10/1,610,692), 0 homozygotes
- gnomAD v2.1 AF 3.98e-06, 0 homozygotes; gnomAD-Canada absent
Sources
BP4 BP4 is not met: under the rule as applied, missense variants are evaluated only on the REVEL threshold (<= 0.249), and REVEL is 0.349 - above the threshold and inside the specification's gray zone. Although the SpliceAI score is low, the splicing sub-path does not apply to missense variants because it only rules out one damaging mechanism, so BP4 remains not met. Details ▾Hide ▴
Reasoning
Missense variant. Under the lab's BP4 curator override (stated in vcep_materials gene_notes), exactly one mechanism is evaluated per variant: for missense variants ONLY the missense sub-path applies (REVEL <= 0.249). REVEL = 0.349 > 0.249, and 0.349 falls in the VCEP gray zone (0.249-0.7333); per the override, BP4 is not_met for this variant regardless of SpliceAI. OVERRIDE CHANGES OUTCOME vs. the literal ATM VCEP v1.5 text: the VCEP's splicing sub-path (SpliceAI <= 0.1) would be met here (max delta 0.05 <= 0.1), but the override explicitly forbids applying the splicing sub-path to a missense variant because a clean SpliceAI score only rules out one damaging mechanism (splicing) and does not clear the variant's own protein-level effect, which REVEL is calibrated to assess. The Sun et al. 2025 row's model-derived 'Functional' classification is DeepATM_predicted=Yes (computational prediction) and does not satisfy the REVEL <= 0.249 missense threshold; it is weak computational corroboration only. BP4 not_met.
Rule
BP4 (Benign Supporting): Missense: REVEL score <= .249; Splicing: No predicted impact via splicing (SpliceAI <= 0.1). Lab curator override for missense variants: missense sub-path only (REVEL <= 0.249); splicing sub-path not applied; if REVEL is in the 0.249-0.7333 gray zone or unavailable, BP4 is not_met/not_assessed regardless of SpliceAI.
Evidence considered
- REVEL score 0.349 (missense sub-path; VCEP BP4 threshold <=0.249 not met; gray zone 0.249-0.7333)
- SpliceAI max delta 0.05 (<=0.1 would satisfy the VCEP splicing sub-path but is suppressed for missense variants by the lab BP4 curator override)
- ATM HBOP VCEP v1.5 BP4 rule from cspec plus lab curator BP4 missense/splicing override from vcep_materials gene_notes
- Sun et al. 2025 (PMID 40580951) Table S1 row c.3080A>G: Classification 'Functional', DeepATM_predicted=Yes (computational prediction) - secondary corroboration only, does not meet REVEL threshold
- BayesDel noAF score -0.237229 (not VCEP-calibrated; benign-leaning direction, secondary only)
Still missing
- REVEL 0.349 sits in the VCEP gray zone (0.249-0.7333): the missense sub-path cannot meet BP4 and no functional (kinase/radiosensitivity) or RNA assay result is available to resolve the protein-level effect of p.His1027Arg.
Sources
Not assessed · 5 · no assessment recorded
PS3 PS3 can only be applied through functional assays approved by the ATM expert panel (three kinase-activity assays and one radiosensitivity assay). No variant-level result from an approved assay exists for this variant - the only functional-source entry (Sun et al. 2025) is a computational model prediction, not a direct measurement - so insufficient evidence was available and PS3 was not assessed. Details ▾Hide ▴
Reasoning
Governing framework is the ClinGen HBOP ATM VCEP v1.5 specification (framework_mode=vcep; cspec found, version 1.5). Under this VCEP, PS3 may only be applied via approved functional assays whose methodology and strength calibration are defined in clingen_hbop_atm_supplementary_tables_1_and_2_v1.xlsx: three kinase-activity assays (Mitui 2009/PMID 18634022, Barone 2009/PMID 19431188, Scott 2002/PMID 11805335; stable transfection of ATM cDNA into ATM-null LCLs; IR-induced phosphorylation of ATM substrates by Western blot; approved, proposed strength PS3_Supporting) and the Mitui 2009 radiosensitivity assay (survival fraction; PS3_Moderate only when combined with a kinase-activity PS3_Supporting result; radiosensitivity alone gets no weight). The VCEP rule set additionally states PS3 Strong: 'Do not use as strong.' None of these three papers is present in source_registry.json and no fulltext is available locally, and the approved-assay table is a methodology/calibration table only - it does not report which variants were tested. No variant-level functional data for c.3080A>G (p.His1027Arg) from an approved assay therefore exists in this case bundle. The only per-variant functional-source row available is Sun et al. 2025 (PMID 40580951) Table S1 for c.3080A>G: Combined_score=-0.276, Classification='Functional', but DeepATM_predicted='Yes' with Confidence='Computational prediction', meaning the variant was NOT directly assayed in the prime-editing saturation screen and its classification is the DeepATM model's extrapolation; per the VCEP file guidance this must not be presented as functional evidence for PS3/BS3 (it is at most weak computational corroboration for PP3/BP4, which is outside this group). The Sun et al. paper also postdates ATM VCEP v1.5 and is not one of the three VCEP-approved assays. Evidence is therefore insufficient: PS3 is not_assessed pending a confirmed variant-level result in an approved assay (or human review of whether the Sun et al. screen qualifies as a novel validated assay under the specification).
Rule
ATM VCEP v1.5 PS3: Strong: 'Do not use as strong.' Moderate: 'Use when a variant fails to rescue both an ATM specific feature (e.g. phosphorylation of ATM-specific targets) AND radiosensitivity.' Supporting: 'Use when a variant fails to rescue an ATM specific feature, only (e.g. phosphorylation of ATM-specific targets). Do not use for radiosensitivity-only as that is not a feature specific to ATM deficiency.' Application restricted to approved assays in clingen_hbop_atm_supplementary_tables_1_and_2_v1.xlsx; variant must be confirmed tested in Mitui 2009 (PMID 18634022), Barone 2009 (PMID 19431188), or Scott 2002 (PMID 11805335).
Evidence considered
- VCEP-approved assay methodology/strength table (kinase activity: Mitui 2009, Barone 2009, Scott 2002 - PS3_Supporting; radiosensitivity: Mitui 2009 - PS3_Moderate, combinable only with a kinase-activity PS3_Supporting result): methodology and calibration only, no per-variant results for c.3080A>G.
- Sun et al. 2025 Table S1 row c.3080A>G: Classification='Functional', Confidence='Computational prediction', DeepATM_predicted='Yes' - excluded from PS3 as not a direct functional measurement (per VCEP file guidance).
- Local publication fulltexts (PMID:25741868, PMID:28779002 and remaining case papers) checked for c.3080A>G / p.His1027Arg / H1027R: zero mentions; no functional assay data for this variant.
Still missing
- No variant-level functional assay result for c.3080A>G from any VCEP-approved assay (Mitui 2009/PMID 18634022, Barone 2009/PMID 19431188, Scott 2002/PMID 11805335); the three papers are not in source_registry.json and no fulltexts are available to confirm whether p.His1027Arg was tested.
- Sun et al. 2025 (PMID 40580951) Table S1 lists c.3080A>G as 'Functional' but with DeepATM_predicted=Yes / Confidence='Computational prediction' - a model extrapolation, not a direct functional measurement; it is explicitly excluded from PS3/BS3 use by the VCEP file guidance.
- If a human reviewer accepts the Sun et al. 2025 prime-editing screen as a novel validated assay, only then could its direct-measurement (DeepATM_predicted=No) classification be considered, and even that would require VCEP calibration; a 'Functional' classification would not support PS3 in any case.
Sources
vcep_clingen_hbop_atm_supplementary_tables_1_and_2_v1vcep_suppl_tables1_pmid_40580951PMID:25741868 ↗PMID:28779002 ↗
PM3 PM3 requires observations of the variant in ataxia-telangiectasia probands carrying a second ATM variant. No such observation exists - none of the reviewed publications mention this variant, all 9 ClinVar submissions are Uncertain significance with no phase or zygosity detail, and no homozygotes are reported - so insufficient evidence was available and PM3 was not assessed. Details ▾Hide ▴
Reasoning
Governing framework is the ClinGen HBOP ATM VCEP v1.5 (cspec doc 639508985; source key vcep_atm_pm3_bp2_1_5), whose points-based PM3 table awards 4.0/2.0/1.0/2.0 points per unrelated A-T proband (phenotype 'confident': confirmed in trans / phase unknown / second variant unidentified or VUS / homozygous [max 2]; 'consistent' = half) with a prerequisite that the variant not exceed 0.01% population frequency. The frequency prerequisite IS satisfied (gnomAD v4.1 AF=6.21e-06 = 0.00062%, grpmax FAF=3.6e-06; gnomAD v2.1 AF=3.98e-06), but there is no observation of c.3080A>G in ANY A-T proband anywhere in the case evidence, so zero PM3 points can be awarded: (1) none of the 6 available publication full texts mention c.3080A>G or p.His1027Arg (screened: 17508274, 25186627, 32885271 with content; 25741868, 28779002, 34242744 with empty extractions); (2) all 9 ClinVar submissions (VCV000188327) are 'Uncertain significance' with no assertion-criteria detail and no zygosity/phase information (the one submission under condition 'Ataxia-telangiectasia', SCV002079721/Natera, provides no genotype detail and cannot be counted as an A-T proband observation); (3) gnomAD v2.1 and v4.1 both report 0 homozygotes. Because the required evidence type (A-T proband observations with phase) was never retrieved — the case literature set is derived from ClinVar citations and contains no A-T cohort/mutation-compilation source — the criterion is not_assessed rather than not_met; a dedicated A-T proband source (e.g., LOVD, ATM mutation compilations, A-T cohort papers) is required to score it. No other source's final classification (all ClinVar VUS) was used to infer this criterion.
Rule
ClinGen HBOP ATM VCEP v1.5 PM3 (vcep_atm_pm3_bp2_1_5 / cspec doc 639508985): points per unrelated A-T proband — phenotype 'confident' 4.0 (confirmed in trans) / 2.0 (phase unknown) / 1.0 (second variant unidentified or VUS) / 2.0 (homozygous, max 2 individuals); phenotype 'consistent' = half; PM3_Supporting=1.0, PM3=2.0 (Moderate), PM3_Strong=4.0, PM3_VeryStrong>=8.0; prerequisite variant population frequency <=0.01%. Not assessed: 0 A-T proband observations of c.3080A>G exist in the case evidence, and no A-T proband source was retrieved to score points.
Evidence considered
- vcep_atm_pm3_bp2_1_5 (ATM PM3/BP2 decision guide v1.5): points-based PM3 table (confident phenotype 4.0 confirmed-in-trans / 2.0 phase-unknown / 1.0 second-variant-unidentified-or-VUS / 2.0 homozygous, max 2; consistent phenotype = half) and >0.01% population-frequency prerequisite
- cspec (doc 639508985): PM3 strength mapping PM3_Supporting=1.0 / PM3=2.0 (moderate) / PM3_Strong=4.0 / PM3_VeryStrong>=8.0 points
- clinvar (VCV000188327): 9 submissions, all 'Uncertain significance', none with assertion-criteria detail or zygosity/phase data; no A-T proband observation extractable
- gnomad_v2 (11-108143261-A-G): AF=3.98e-06 (1/251,156 alleles), 0 homozygotes — frequency prerequisite satisfied; no homozygous proband signal
- gnomad_v4 (chr11-108272534-A-G): AF=6.21e-06 (10/1,610,692 alleles), 0 homozygotes, grpmax FAF=3.6e-06 — frequency prerequisite satisfied; no homozygous proband signal
- Publications screened (see publications_reviewed): none of the 6 full texts mention c.3080A>G or p.His1027Arg, so no literature-based proband observation exists
Still missing
- No A-T proband observation source was retrieved: the case literature set (built from ClinVar citations) contains no A-T cohort or ATM mutation-compilation reports; all 6 available full texts were screened and none mention c.3080A>G/p.His1027Arg, but a systematic A-T proband search (LOVD, ATM mutation databases, A-T cohort papers) is required to determine whether the variant has been observed homozygous or in trans with a P/LP ATM variant in affected patients
- ClinVar submissions (9, all VUS) expose no zygosity or phase; SCV002079721 (Natera) lists condition 'Ataxia-telangiectasia' but with no genotype/phase detail, so no A-T proband count can be extracted from ClinVar
- gnomAD v2.1 (1/251,156 alleles) and v4.1 (10/1,610,692 alleles) report 0 homozygotes, consistent with no recessive proband signal, but population data cannot substitute for A-T proband observations
Sources
PP1 PP1 requires segregation of the variant in affected relatives, with both variants identified in the proband. No segregation data exist for this variant (no affected relatives tested, no meiosis or parental genotype information), so insufficient evidence was available and PP1 was not assessed. Details ▾Hide ▴
Reasoning
PP1 is Applicable under the governing ATM HBOP VCEP v1.5 for the AR condition, with points-based strength tiers: segregation in >=3 affected relatives = Strong, 2 affected relatives = Moderate, 1 affected relative = Supporting; prerequisite per VCEP instruction: 'AR Condition: Affected relatives must have both variants identified in proband.' However, the consolidated case-evidence bundle contains no segregation data whatsoever for this variant: no proband family history, no affected relatives tested, no meiosis/segregation counts, and no parental or sibling genotype data appear in review_manifest, case_summary, prefetch, evidence, literature_pass, vcep_materials, or fulltext_map. None of the six full-text publications available in /publications concerns familial segregation of c.3080A>G (they are the ACMG guidelines, breast-cancer risk/panel cohorts, St. Gallen consensus, and genetic-counseling recommendations). PP1 therefore cannot be assessed; no criterion should be inferred from any source's final classification.
Rule
ATM HBOP VCEP v1.5 PP1 (AR condition): Segregation in >=3 affected relatives = Strong; 2 affected relatives = Moderate; 1 affected relative = Supporting. 'AR Condition: Affected relatives must have both variants identified in proband.'
Still missing
- No segregation data: no affected relatives (A-T) tested for the variant and for both variants identified in the proband; family segregation studies are required to apply PP1 under the VCEP AR rule
Sources
BS3 BS3 requires demonstration that the variant does not damage ATM function in a panel-approved assay. No approved-assay result exists for this variant - the only functional-source entry (Sun et al. 2025) is a computational model prediction, not a direct measurement - so insufficient evidence was available and BS3 was not assessed. Details ▾Hide ▴
Reasoning
Under the governing ClinGen HBOP ATM VCEP v1.5 specification, BS3 may only be applied via the same approved assays as PS3 (clingen_hbop_atm_supplementary_tables_1_and_2_v1.xlsx): BS3_Supporting for rescue of EITHER an ATM-specific feature (kinase activity: phosphorylation of ATM substrates) OR radiosensitivity, and BS3_Moderate for rescue of BOTH. The three approved assay papers (Mitui 2009/PMID 18634022, Barone 2009/PMID 19431188, Scott 2002/PMID 11805335) are not in source_registry.json, no fulltexts are available locally, and the calibration table does not report per-variant results, so there is no evidence that c.3080A>G was tested in or rescued function in any approved assay. The only variant-level row in a functional source is Sun et al. 2025 (PMID 40580951) Table S1, which lists c.3080A>G as Classification='Functional' - however DeepATM_predicted='Yes' and Confidence='Computational prediction', i.e. the variant was not directly measured in the prime-editing screen, and per the VCEP file guidance this computational extrapolation must not be routed to BS3 (it is at most weak computational corroboration for BP4, outside this group). The Sun et al. paper is also not one of the three VCEP-approved assays and postdates ATM VCEP v1.5. Evidence is insufficient to demonstrate a well-established lack of damaging effect on protein function: BS3 is not_assessed. Note that even if the Sun et al. 'Functional' call were accepted as a direct measurement, the index guidance instructs flagging needs_human_review rather than silently assigning a BS3 strength the VCEP has not calibrated for that assay.
Rule
ATM VCEP v1.5 BS3: Moderate: 'Use when a variant rescues both an ATM specifc feature (e.g. phosphorylation of ATM-specific targets) AND radiosensitivity.' Supporting: 'Use when a variant rescues EITHER an ATM specifc feature OR rescues radiosensitivity.' Application restricted to approved assays in clingen_hbop_atm_supplementary_tables_1_and_2_v1.xlsx; variant must be confirmed tested in Mitui 2009 (PMID 18634022), Barone 2009 (PMID 19431188), or Scott 2002 (PMID 11805335).
Evidence considered
- VCEP-approved assay methodology/strength table (BS3_Supporting for rescue of an ATM-specific feature or radiosensitivity; BS3_Moderate for rescue of both): methodology and calibration only, no per-variant results for c.3080A>G.
- Sun et al. 2025 Table S1 row c.3080A>G: Classification='Functional', Confidence='Computational prediction', DeepATM_predicted='Yes' - excluded from BS3 as not a direct functional measurement (per VCEP file guidance); paper not among VCEP-approved assays.
- Local publication fulltexts (PMID:25741868, PMID:28779002 and remaining case papers) checked for c.3080A>G / p.His1027Arg / H1027R: zero mentions; no functional rescue data for this variant.
Still missing
- No variant-level result for c.3080A>G in any VCEP-approved assay demonstrating rescue of an ATM-specific feature and/or radiosensitivity (Mitui 2009/PMID 18634022, Barone 2009/PMID 19431188, Scott 2002/PMID 11805335 not available in this case bundle).
- Sun et al. 2025 Table S1 'Functional' classification for c.3080A>G is DeepATM_predicted=Yes / Confidence='Computational prediction' (model extrapolation, not a direct functional measurement) and is explicitly excluded from BS3 use by the VCEP file guidance; its assay is not VCEP-approved and postdates v1.5.
- A human reviewer should decide whether to obtain/perform an approved functional assay (e.g., ATM kinase activity or Mitui 2009 radiosensitivity rescue) before BS3 can be applied.
Sources
vcep_clingen_hbop_atm_supplementary_tables_1_and_2_v1vcep_suppl_tables1_pmid_40580951PMID:25741868 ↗PMID:28779002 ↗
BP2 BP2 requires observations of unaffected carriers who carry this variant in trans with a pathogenic ATM variant (or evidence of a cis configuration). No such observations exist - no submission or publication reports this variant alongside a pathogenic allele, and population-database carrier alleles carry no phase information - so insufficient evidence was available and BP2 was not assessed. Details ▾Hide ▴
Reasoning
Governing framework is the ClinGen HBOP ATM VCEP v1.5 (vcep_atm_pm3_bp2_1_5 / cspec doc 639508985), whose BP2 table awards negative points per unaffected (non-A-T) individual carrying the variant with a P/LP ATM variant in trans: -4.0 (confirmed trans) / -2.0 (phase unknown) / -2.0 laboratory or -1.0 database setting (homozygous, max -2.0), with BP2=-1.0 (Supporting), BP2_Moderate=-2.0, BP2_Strong<=-4.0; instructions require the unaffected individual to be >=18 years old with no evidence of A-T. No such observation exists in the case evidence: no ClinVar submission or publication reports c.3080A>G together with a P/LP ATM variant in trans in any unaffected carrier (all 6 available full texts were screened and none mention the variant; all 9 ClinVar submissions are VUS without zygosity/phase detail). gnomAD carrier alleles (10 in v4.1) carry no phase information relative to a second ATM variant. The generic-ACMG cis clause ('observed in cis with a pathogenic variant in any inheritance pattern') also finds no support — no cis-configuration evidence for this variant exists in any source. Because the required evidence type (unaffected-carrier observations with a P/LP allele in trans, or cis phase data) was never retrieved, BP2 is not_assessed rather than not_met; phase data (parental testing / single-molecule haplotyping) or unaffected-carrier cohort data with a second ATM variant is required to score it. No other source's final classification (all ClinVar VUS) was used to infer this criterion.
Rule
ClinGen HBOP ATM VCEP v1.5 BP2 (vcep_atm_pm3_bp2_1_5 / cspec doc 639508985): -4.0 points per unaffected (non-A-T, >=18y, no A-T evidence) individual carrying the variant with a P/LP variant confirmed in trans; -2.0 phase unknown; -2.0 laboratory / -1.0 database homozygous (max -2.0); BP2=-1.0 (Supporting), BP2_Moderate=-2.0, BP2_Strong<=-4.0. Not assessed: no unaffected-carrier-in-trans or cis-phase observations of c.3080A>G exist in the case evidence.
Evidence considered
- vcep_atm_pm3_bp2_1_5 (ATM PM3/BP2 decision guide v1.5): BP2 points per unaffected (non-A-T) individual with P/LP variant in trans — -4.0 confirmed trans / -2.0 phase unknown / -2.0 lab or -1.0 database homozygous (max -2.0); BP2=-1.0 (Supporting), BP2_Moderate=-2.0, BP2_Strong<=-4.0; unaffected individual must be >=18y with no evidence of A-T
- cspec (doc 639508985): BP2 strength mapping and instructionsToUse (points-based assignment; >=18y, no A-T requirement); originalACMGSummary retains cis clause ('observed in cis with a pathogenic variant in any inheritance pattern')
- clinvar (VCV000188327): 9 submissions, all 'Uncertain significance' with no zygosity/phase detail; no unaffected-carrier-with-P/LP-in-trans observation extractable
- gnomad_v2 / gnomad_v4: population carrier alleles without phase (no second-variant or trans/cis information)
- Publications screened (see publications_reviewed): none of the 6 full texts mention c.3080A>G or p.His1027Arg, so no literature-based unaffected-carrier or cis-phase observation exists
Still missing
- No unaffected-carrier data with a second P/LP ATM variant in trans: no publication or ClinVar submission reports c.3080A>G in a >=18-year-old unaffected individual together with a P/LP ATM allele; gnomAD carrier alleles (1 in v2.1, 10 in v4.1) lack phase information
- No cis/trans phase data of any kind (parental testing, single-molecule haplotyping, familial segregation) is available for any observation of this variant
- No systematic unaffected-carrier cohort source (e.g., ATM heterozygote studies reporting second-variant phase) was included in the case bundle
Sources
Not applicable · 16· recorded as not applicable Show reasons ▾Hide reasons ▴
PVS1 N/APVS1 applies only to null variants (nonsense, frameshift, canonical splice-site, initiation-codon, and single/multi-exon deletion variants). This variant is a missense change (p.His1027Arg) with no predicted splicing impact (SpliceAI max delta score 0.05), so it cannot trigger PVS1; it would only be reconsidered if RNA evidence demonstrated an out-of-frame splicing consequence, for which none exists. Details ▾Hide ▴
Reasoning
PVS1 is restricted to null variants: nonsense, frameshift, canonical +-1,2 splice-site, initiation-codon, and single/multi-exon deletion variants (ClinGen HBOP ATM VCEP v1.5 PVS1 decision tree; generic ACMG/AMP 2015 definition identical in scope). NM_000051.4:c.3080A>G is a single-nucleotide missense substitution (NP_000042.3:p.(His1027Arg)), consequence_class=missense, not at a canonical splice consensus position (canonical_splice_consensus=false), and SpliceAI predicts no significant splice impact (max delta score = 0.05). The pipeline's pvs1_variant_assessment confirms the variant does not fall into any default PVS1 null-variant bucket and apply_generic_pvs1_framework=false. The gene-level PVS1 gate is eligible (ATM LOF is an established mechanism, pvs1_gene_context pvs1_gene_gate=eligible), but the variant-level consequence does not trigger the criterion, so PVS1 cannot be applied at any strength. Would only be reconsidered if RNA evidence demonstrated an out-of-frame splice consequence, for which no evidence exists here.
Rule
ClinGen HBOP ATM VCEP v1.5: PVS1 null variant (nonsense, frameshift, canonical +/-1 or 2 splice sites, initiation codon, single or multi-exon deletion) in a gene where LOF is a known mechanism of disease; use ATM PVS1 Decision Tree
Evidence considered
- Variant normalization: NM_000051.4:c.3080A>G -> NP_000042.3:p.(His1027Arg), consequence class = missense
- pvs1_variant_assessment: variant_bucket=other, apply_generic_pvs1_framework=false, rationale that variant is not nonsense/frameshift/canonical +-1,2 splice
- pvs1_gene_context: pvs1_gene_gate=eligible (LOF established mechanism for ATM via CSPEC/VCEP)
- SpliceAI: max delta score 0.05 (DS_AG 0.05, DS_DG 0.02) - no significant predicted splice impact
- ATM VCEP v1.5 PVS1 decision tree: applies to null variants only (nonsense, frameshift, canonical +/-1,2 splice sites, initiation codon, single/multi-exon deletion)
Sources
PS2 VCEPPS2 is not applicable under the ATM specification: informative de novo occurrences have not been observed, and de novo evidence is not informative for autosomal recessive ataxia-telangiectasia. No parental testing or de novo occurrence data are available in any case. Details ▾Hide ▴
Reasoning
The governing ClinGen HBOP ATM VCEP v1.5 (CSPEC doc 639508985, framework_mode=vcep) explicitly marks PS2 as Not applicable: 'Do not use for AD or AR disease: Informative de novo occurrences have not yet been observed and de novo AR conditions are unlikely to be informed by phase.' ATM-associated Ataxia Telangiectasia is an autosomal recessive disorder; de novo occurrence in a proband is not informative evidence under this specification, so PS2 is not applied regardless of case data. Additionally, no parental testing, maternity/paternity confirmation, or de novo occurrence data exists anywhere in the case evidence bundle.
VCEP rule
ATM HBOP VCEP v1.5 PS2: Not applicable. 'Do not use for AD or AR disease: Informative de novo occurrences have not yet been observed and de novo AR conditions are unlikely to be informed by phase.'
Sources
PM1 VCEPPM1 is not applicable because the ATM expert panel provides no hotspot- or domain-based PM1 rule for this gene. Consistently, residue 1027 is not a statistically significant hotspot (CancerHotspots) and no curated variant-specific functional evidence exists for p.His1027Arg. Details ▾Hide ▴
Reasoning
The ClinGen HBOP ATM VCEP v1.5 cspec (doc 639508985) declares PM1 'Not Applicable' for ATM - the panel provides no hotspot- or domain-based PM1 rule for this gene, and generic ACMG/AMP is not used because an applicable specification exists. This is consistent with the collected evidence: CancerHotspots reports ATM H1027 is not in a statistically significant hotspot (evidence.json hotspots result, query 'ATM H1027' = no_result), and OncoKB has no variant-specific curated functional evidence for H1027R. For domain context, the only domain guidance in the ATM VCEP concerns the FAT/PI3-K/FATC region (treated as one contiguous domain, PMID 28508083, in the PVS1 guide); residue 1027 lies N-terminal to that region and the spec does not define any PM1-style domain rule covering it. PM1 therefore cannot be applied at any strength.
VCEP rule
ClinGen HBOP ATM VCEP v1.5 PM1: Not Applicable
Evidence considered
- ATM VCEP v1.5 cspec PM1: applicability = Not Applicable
- CancerHotspots (evidence.json): ATM H1027 not in a statistically significant hotspot - no hotspot row found
- OncoKB (evidence.json): H1027R = Unknown Oncogenic Effect, no variant-specific reviewed functional evidence
- ATM PVS1 v1.5 guide domain note: FAT, PI3-K and FATC treated as one contiguous domain (PMID 28508083); residue 1027 is outside this C-terminal region and no domain-based PM1 rule is defined
Sources
cspec ↗vcep_atm_pvs1_1_5
PM4 N/APM4 is restricted to stop-loss variants under the ATM specification. This is a missense substitution that changes a single amino acid without altering protein length or disrupting the stop codon, so PM4 does not apply. Details ▾Hide ▴
Reasoning
The ATM VCEP v1.5 restricts PM4 to stop-loss variants only: 'Do not use for in-frame insertions or deletions less than a single exon; Use for stop-loss variants, only' (strength rule: 'Use for stop-loss variants'). The original ACMG/AMP 2015 scope is protein length change from in-frame deletions/insertions in a non-repeat region or stop-loss variants. NM_000051.4:c.3080A>G is a missense substitution (p.His1027Arg) that causes a single amino acid change with no alteration of protein length, is not an in-frame insertion/deletion, and does not disrupt the stop codon (ATM stop codon is far C-terminal at p.Arg3047* vicinity / position ~3056; c.3080 is an internal coding position). No stop-loss or length-change consequence exists, so PM4 is not applicable at any strength.
Rule
ClinGen HBOP ATM VCEP v1.5 PM4 (Pathogenic Moderate): use for stop-loss variants, only
Evidence considered
- Variant normalization: NM_000051.4:c.3080A>G -> NP_000042.3:p.(His1027Arg) (missense, single amino acid substitution)
- ATM VCEP v1.5 CSPEC PM4: instructionsToUse 'Do not use for in-frame insertions or deletions less than a single exon; Use for stop-loss variants, only'; strength rule 'Use for stop-loss variants'
- No stop-loss consequence: the variant is not within the stop codon and no protein length change is predicted
Sources
PM5 N/APM5 is not applicable under the ATM specification, which limits the rule to truncating and splice variants with premature termination codons upstream of p.Arg3047; a missense variant is outside that scope. Even under the classic reading, no other pathogenic missense at residue 1027 is documented (ClinVar lists only this variant, as Uncertain significance), so the outcome is unchanged. Details ▾Hide ▴
Reasoning
Under the governing ClinGen HBOP ATM VCEP v1.5 cspec, the PM5 rule (Supporting) is written with truncation-cutoff semantics: 'Apply to frameshifting or truncating variants with premature termination codons upstream of p.Arg3047. Apply to splice variants as with premature termination codons upstream of p.Arg3047 where PVS1_VS(RNA) is applied based on high quality observed splicing impact and must be NMD prone.' The VUA is a missense substitution (p.His1027Arg), which does not fall into any variant class covered by this rule, so PM5 is not applicable. The pipeline artifact pm5_candidates.json independently reached the same conclusion (pm5_mode = truncation_cutoff_gene_specific, eligible_for_classic_pm5_search = false, recommendation = not_applicable, 0 same-residue candidates seen). For completeness under the classic ACMG PM5 reading (novel missense at a residue where a different missense is established P/LP), no alternate pathogenic/likely-pathogenic missense at residue 1027 is documented anywhere in the bundle (ClinVar contains only the VUA as VUS; no same-residue comparator in any source), so PM5 would be not_met either way - the outcome is identical, but the anomalous rule text (PM5 carrying truncation-cutoff language normally associated with PVS1-style rules) is flagged for human review so the panel can confirm the intended PM5 semantics for ATM.
Rule
ClinGen HBOP ATM VCEP v1.5 PM5 (Pathogenic Moderate default; Supporting strength rule): apply to frameshifting or truncating variants with premature termination codons upstream of p.Arg3047; apply to splice variants as with PTCs upstream of p.Arg3047 where PVS1_VS(RNA) is applied (NMD prone) - truncation-cutoff semantics, not classic same-residue missense PM5
Evidence considered
- ATM VCEP v1.5 cspec PM5 rule (Supporting): applies to frameshifting/truncating variants with PTCs upstream of p.Arg3047 and to NMD-prone splice variants with PTCs upstream of p.Arg3047 (PVS1_VS(RNA)); the VUA is a missense and does not meet these variant-class conditions
- pm5_candidates.json: pm5_mode = truncation_cutoff_gene_specific, eligible_for_classic_pm5_search = false, candidate_summary total_candidates_seen = 0, recommendation = not_applicable
- ClinVar VCV000188327: only codon-1027 record is the VUA itself (VUS, 9 labs); no P/LP alternate missense at residue 1027
Still missing
- No alternate missense at residue 1027 (other than p.His1027Arg) was identified in the bundle; a dedicated ClinVar same-residue search was not run (pm5_candidates.clinvar_queries empty) - optional reviewer confirmation
- The ATM VCEP v1.5 cspec PM5 rule text uses truncation-cutoff (PTC upstream of p.Arg3047) semantics rather than classic same-residue missense PM5; needs panel/human confirmation of intended semantics (outcome unchanged for this missense variant)
Sources
PM6 VCEPPM6 is not applicable under the ATM specification: assumed-de-novo evidence is not weighted in autosomal recessive disease. No assumed de novo occurrence or parental testing data exist. Details ▾Hide ▴
Reasoning
The governing ClinGen HBOP ATM VCEP v1.5 explicitly marks PM6 as Not applicable with the identical instruction used for PS2: 'Do not use for AD or AR disease: Informative de novo occurrences have not yet been observed and de novo AR conditions are unlikely to be informed by phase.' ATM A-T is autosomal recessive; assumed-de-novo (unconfirmed parental testing) evidence is not weighted in AR disease under this specification. No case evidence of an assumed de novo occurrence or parental testing exists in the bundle.
VCEP rule
ATM HBOP VCEP v1.5 PM6: Not applicable. 'Do not use for AD or AR disease: Informative de novo occurrences have not yet been observed and de novo AR conditions are unlikely to be informed by phase.'
Sources
PP2 VCEPPP2 is not applicable because the ATM specification does not evaluate the gene under the missense-mechanism premise of PP2: ATM-associated disease is driven predominantly by loss of function. Details ▾Hide ▴
Reasoning
The ClinGen HBOP ATM VCEP v1.5 cspec declares PP2 'Not Applicable' for ATM, and the applicable specification takes precedence over generic ACMG/AMP. This is mechanistically coherent: ATM-associated disease (Ataxia-telangiectasia, recessive; ATM-related cancer susceptibility) is driven predominantly by loss of function - the VCEP provides PVS1 guidance and the pipeline's pvs1_gene_context confirms lof_mechanism_supported=true - so the gene is not evaluated under the missense-mechanism premises of PP2 (low rate of benign missense variation + missense as a common disease mechanism). Additionally, no HCI prior / custom PP2 score was available for this variant (hci_prior found=false, mapp_score=null, custom_pp2_score=null). PP2 therefore cannot be applied at any strength.
VCEP rule
ClinGen HBOP ATM VCEP v1.5 PP2: Not Applicable
Evidence considered
- ATM VCEP v1.5 cspec PP2: applicability = Not Applicable
- pvs1_gene_context.json: lof_mechanism_supported = true (ATM germline LOF is an established disease mechanism per CSPEC/VCEP) - inconsistent with PP2's missense-mechanism premise
- case_summary hci_prior: found = false (no HCI prior/mapp/custom PP2 score available)
Sources
cspec ↗pvs1_gene_context
PP4 VCEPPP4 is not applicable because the ATM specification does not use phenotype specificity as an independent criterion; for this recessive condition, phenotype specificity is captured through the PM3 framework. Details ▾Hide ▴
Reasoning
PP4 (phenotype specificity) is marked Not Applicable in the governing ClinGen HBOP ATM VCEP v1.5 specification; the VCEP does not use phenotype specificity as an independent criterion for ATM variants (A-T phenotype specificity is captured through the PM3 points framework for recessive disease, and cancer phenotypes are not considered specific enough). No proband phenotype narratives are available in this case in any event.
VCEP rule
ATM HBOP VCEP v1.5: PP4 applicability = Not Applicable.
Evidence considered
- ATM HBOP VCEP v1.5 CSPEC lists PP4 applicability as Not Applicable.
- No proband phenotype narratives present in the case evidence bundle.
Sources
PP5 VCEPPP5 is not applicable because there is no expert-panel classification of Pathogenic or Likely pathogenic for this variant: ClinVar lists it as Uncertain significance from 9 ordinary clinical laboratory submissions, with no expert panel input. Details ▾Hide ▴
Reasoning
PP5 is marked Not Applicable in the ATM HBOP VCEP v1.5 specification, and the required trigger is absent: there is no exact-variant ClinVar expert-panel classification of Pathogenic/Likely pathogenic for c.3080A>G (ClinVar VCV000188327 is Uncertain significance from 9 ordinary clinical laboratory submissions, 0 expert panel submissions, review status 'criteria provided, single submitter'/2 stars). Per policy, ordinary laboratory submissions and aggregate ClinVar labels never trigger PP5.
VCEP rule
PP5 applied only for an exact-variant ClinVar expert-panel Pathogenic/Likely pathogenic classification; ATM HBOP VCEP v1.5 marks PP5 Not Applicable.
Evidence considered
- ClinVar VCV000188327: Uncertain significance; 9 clinical laboratory submissions, all VUS; expert_panels = {} (0 expert panel classifications).
- ATM HBOP VCEP v1.5 CSPEC lists PP5 applicability as Not Applicable.
Sources
BS2 VCEPBS2 is not applicable: the ATM expert panel specification explicitly marks it as not applicable, and the gene-specific specification takes precedence over the generic ACMG/AMP framework. Details ▾Hide ▴
Reasoning
The governing ClinGen HBOP ATM VCEP v1.5 specification (CSPEC doc 639508985) explicitly marks BS2 as 'Not Applicable'. Because an applicable gene-specific specification exists and governs this case (framework_mode=vcep, precedence_rank=1), generic ACMG/AMP BS2 is not evaluated under this framework.
VCEP rule
Not Applicable per ATM VCEP v1.5 (CSPEC doc 639508985)
Evidence considered
- ATM VCEP v1.5 CSPEC applicability field: BS2 = Not Applicable
Sources
BS4 VCEPBS4 is not applicable under the ATM specification: informative instances of lack of co-segregation in A-T families are considered too rare to be weighted, and no lack-of-segregation data exist. Details ▾Hide ▴
Reasoning
The governing ClinGen HBOP ATM VCEP v1.5 explicitly marks BS4 as Not applicable: 'AR Condition: Informative instances of lack of co-segregation in A-T families are too rare to be considered for weight at this time and can also be considered for BP2 if biallelic unaffected patients are observed in an A-T family.' Lack-of-segregation evidence therefore carries no weight in ATM/A-T under this specification (the biallelic-unaffected scenario routes to BP2, outside this group). No lack-of-segregation data exists in the case bundle in any case.
VCEP rule
ATM HBOP VCEP v1.5 BS4: Not applicable. 'AR Condition: Informative instances of lack of co-segregation in A-T families are too rare to be considered for weight at this time and can also be considered for BP2 if biallelic unaffected patients are observed in an A-T family.'
Sources
BP1 VCEPBP1 is not applicable because loss of function is an established disease mechanism for ATM, which is the opposite of the premise BP1 requires. Independently, this is a missense variant, not a truncating one. Details ▾Hide ▴
Reasoning
The ClinGen HBOP ATM VCEP v1.5 cspec declares BP1 'Not Applicable' for ATM, and the applicable specification takes precedence over generic ACMG/AMP. BP1 (missense/truncating variant in a gene where LOF is not a mechanism of disease) cannot apply to ATM because germline loss of function is an established disease mechanism (pvs1_gene_context lof_mechanism_supported=true; VCEP PVS1 guidance applies). Independently, the VUA is a missense substitution (p.His1027Arg), not a truncating variant, so BP1's target variant class does not apply in any case. BP1 therefore cannot be applied at any strength.
VCEP rule
ClinGen HBOP ATM VCEP v1.5 BP1: Not Applicable (gene where LOF is not a mechanism - does not apply to ATM, where LOF is established)
Evidence considered
- ATM VCEP v1.5 cspec BP1: applicability = Not Applicable
- pvs1_gene_context.json: lof_mechanism_supported = true, pvs1_gene_gate = eligible - LOF is an established ATM disease mechanism, excluding BP1
- Variant normalization: NM_000051.4:c.3080A>G -> NP_000042.3:p.(His1027Arg) - missense, not a truncating variant
Sources
cspec ↗pvs1_gene_context
BP3 VCEPBP3 applies to in-frame insertions or deletions in repetitive regions; this is a missense single-nucleotide change, and the ATM specification explicitly disallows BP3, so it is not applicable. Details ▾Hide ▴
Reasoning
The ATM VCEP v1.5 marks BP3 as Not applicable with instructions 'Do not use.' Generic ACMG/AMP BP3 applies to in-frame deletions/insertions in a repetitive region without known function. NM_000051.4:c.3080A>G is a missense SNV (p.His1027Arg), not an in-frame indel, and there is no evidence the position lies in a repeat region; moreover the governing VCEP explicitly disallows BP3 for ATM. BP3 is therefore not applicable.
VCEP rule
ClinGen HBOP ATM VCEP v1.5 BP3: Not applicable (Do not use)
Evidence considered
- Variant normalization: NM_000051.4:c.3080A>G -> NP_000042.3:p.(His1027Arg) (missense SNV, not an in-frame indel)
- ATM VCEP v1.5 CSPEC BP3: applicability=Not applicable, instructionsToUse 'Do not use'
- No repeat-region/in-frame-indel context applies to a single-nucleotide missense substitution
Sources
BP5 VCEPBP5 is not applicable under the ATM specification, and no alternate molecular basis of disease is documented for a proband carrying this variant. Details ▾Hide ▴
Reasoning
BP5 is marked Not Applicable in the governing ATM HBOP VCEP v1.5 specification, and no case data document an alternate molecular basis of disease in a proband carrying this variant that would engage the criterion.
VCEP rule
ATM HBOP VCEP v1.5: BP5 applicability = Not Applicable.
Evidence considered
- ATM HBOP VCEP v1.5 CSPEC lists BP5 applicability as Not Applicable.
- No proband narratives or alternate-cause reports in the case evidence bundle.
Sources
BP6 VCEPBP6 is not applicable because there is no expert-panel classification of Benign or Likely benign for this variant: ClinVar lists it as Uncertain significance from 9 ordinary clinical laboratory submissions. Details ▾Hide ▴
Reasoning
BP6 is marked Not Applicable in the ATM HBOP VCEP v1.5 specification, and the required trigger is absent: there is no exact-variant ClinVar expert-panel classification of Benign/Likely benign for c.3080A>G (ClinVar VCV000188327 is Uncertain significance from 9 ordinary clinical laboratory submissions, 0 expert panel submissions). Ordinary laboratory submissions and aggregate ClinVar labels never trigger BP6.
VCEP rule
BP6 applied only for an exact-variant ClinVar expert-panel Benign/Likely benign classification; ATM HBOP VCEP v1.5 marks BP6 Not Applicable.
Evidence considered
- ClinVar VCV000188327: Uncertain significance; 9 clinical laboratory submissions, all VUS; expert_panels = {} (0 expert panel classifications); no benign assertion of any kind.
- ATM HBOP VCEP v1.5 CSPEC lists BP6 applicability as Not Applicable.
Sources
BP7 N/ABP7 applies to synonymous and deep intronic variants; this is a missense change (p.His1027Arg, exon 21), so it falls outside the criterion's scope. Details ▾Hide ▴
Reasoning
ATM HBOP VCEP v1.5 BP7 is defined for synonymous and deep intronic variants (further than +7 from donor and further than -21 from acceptor sites), with BP7(RNA) applicable to silent substitutions and intronic variants with observed lack of aberrant RNA. The variant under assessment is a missense change (NM_000051.4:c.3080A>G; p.His1027Arg, exon 21), not a synonymous or intronic variant, so BP7 is outside the criterion's scope and is not applicable. No RNA evidence was assessed for BP7.
Rule
BP7 (Benign Supporting): Use for synonymous and deep intronic variants defined as further than (but not including) +7 and further than (but not including) -21 at donor and acceptor sites, respectively. BP7(RNA): Observed lack of aberrant RNA defect for silent substitutions and intronic variants.
Evidence considered
- Variant consequence class: missense (p.His1027Arg) per normalization and Sun et al. 2025 Table S1 (Variant_consequence='Missense', exon_21)
- ATM HBOP VCEP v1.5 BP7 rule from cspec (rule text quoted in 'rule') - scope limited to synonymous/deep intronic variants
Sources
Identifiers
- Gene
- ATM
- Transcript
- NM_000051.4
- Coding (HGVS c.)
- NM_000051.4:c.3080A>G
- Protein (HGVS p.)
- NP_000042.3:p.(His1027Arg)
- GRCh38
- chr11:108272534 A>G
- GRCh37
- chr11:108143261 A>G
- ClinVar
- 188327
Gene diagram
Every interpreted variant in ATM, mapped onto the exon structure of NM_000051.4. Select a row to find it on the plot.
ATMNM_000051.4
Fetching transcript structure from UCSC…
Exons
—
Transcript span
—
Strand
—
Variants mapped
—
Source
UCSC ncbiRefSeqCurated
All variants in ATM
—
Select a row to find it on the plot; use the link to open its page.
| Protein | Location | Classification | Link |
|---|
gnomAD v4.1
10 / 1,610,692 alleles
AF 6.21e-06 · 0 homozygotes · FAF 3.6e-06
gnomAD v2.1
1 / 251,156 alleles
AF 3.98e-06 · 0 homozygotes
gnomAD-Canada v1.0
Allele count data not recorded
By genetic ancestry
gnomAD v4.1
AF 6.2e-06 · 10 / 1,610,692 alleles · 0 hom · FAF 3.6e-06
| Remaining individuals1 / 62,392 | 1.6e-05 |
| European (non-Finnish)9 / 1,176,954 | 7.6e-06 |
0 alternate alleles recorded in 8 other groups: Admixed American, European (Finnish), Amish, East Asian, Middle Eastern, South Asian, Ashkenazi Jewish, African/African American.
gnomAD v2.1
AF 4e-06 · 1 / 251,156 alleles · 0 hom
| European (non-Finnish)1 / 113,570 | 8.8e-06 |
0 alternate alleles recorded in 7 other groups: African/African American, Admixed American, Ashkenazi Jewish, East Asian, European (Finnish), Remaining individuals, South Asian.
Statements
gnomAD v4.1
This variant is present in gnomAD v4.1 (AF= 6.20851e-06; MAF= 0.00062%, 10/1610692 alleles, homozygotes = 0) and has highest observed frequency in the Remaining individuals population (AF= 1.60277e-05; MAF= 0.00160%, 1/62392 alleles, homozygotes = 0); grpmax FAF= 3.6e-06.
gnomAD v2.1
This variant is present in gnomAD v2.1 (AF= 3.98159e-06; MAF= 0.00040%, 1/251156 alleles, homozygotes = 0) and has highest observed frequency in the European (non-Finnish) population (AF= 8.80514e-06; MAF= 0.00088%, 1/113570 alleles, homozygotes = 0).
gnomAD-Canada
Absent from gnomAD-Canada v1.0.
SpliceAI and Pangolin
Other predictors
- REVEL
- 0.349
- BayesDel
- -0.237
How the predictions were used
PP3 PP3 is not met on either computational path calibrated by the ATM specification: REVEL is 0.349, well below the >0.7333 pathogenic threshold, and SpliceAI max delta is 0.05, below the >=0.2 threshold. Secondary predictors (BayesDel, AlphaMissense, EVE, CADD) also do not indicate a damaging effect. Details ▾Hide ▴
Reasoning
Missense variant NM_000051.4:c.3080A>G (p.His1027Arg). ATM HBOP VCEP v1.5 PP3 (Pathogenic Supporting) has two sub-paths. Missense sub-path: REVEL > 0.7333 required; REVEL = 0.349, well below threshold -> not met. Splicing sub-path (which the VCEP applies to missense variants as well): SpliceAI >= 0.2 required; SpliceAI max delta = 0.05 (DS_AG 0.05, DS_DG 0.02) -> not met. Both VCEP-calibrated computational paths fail, so PP3 is not met on either mechanism. Secondary in-silico predictors (not VCEP-calibrated): BayesDel noAF -0.237229, AlphaMissense 0.2278, EVE 0.266, CADD phred 20.3 — none indicate damaging effect. Sun et al. 2025 Table S1 row for c.3080A>G lists Classification 'Functional' but with DeepATM_predicted=Yes and Confidence='Computational prediction', i.e. a trained-model extrapolation, not a functional measurement; per VCEP index guidance it can only serve as weak computational corroboration and does not support PP3. Literature papers indexed as PP3 candidates (PMID:25186627, PMID:32885271) were full-text searched and do not mention this variant. Decision is deterministic from VCEP thresholds.
Rule
PP3 (Pathogenic Supporting): Missense: REVEL > .7333; Splicing: Predicted impact via splicing (SpliceAI >= 0.2) for silent, missense/in-frame and for intronic variants outside of donor and acceptor 1,2 sites.
Evidence considered
- REVEL score 0.349 (missense path; VCEP PP3 threshold >0.7333 not met)
- SpliceAI max delta 0.05, DS_AG 0.05, DS_DG 0.02 (splice path; VCEP PP3 threshold >=0.2 not met)
- ATM HBOP VCEP v1.5 PP3 rule from cspec (rule text quoted in 'rule')
- Sun et al. 2025 (PMID 40580951) Table S1 row c.3080A>G: REVEL 0.349, AlphaMissense 0.2278, EVE_scores_ASM 0.266, CADD.phred 20.3, boostDM 0.081, Classification 'Functional', DeepATM_predicted=Yes (computational prediction) - secondary corroboration only
- BayesDel noAF score -0.237229 (not VCEP-calibrated; benign-leaning, no PP3 support)
Still missing
- No RNA/minigene splice assay data for this variant (not required for the PP3 computational sub-paths, but would be the only route to a splice-impact finding given SpliceAI < 0.2).
Sources
BP4 BP4 is not met: under the rule as applied, missense variants are evaluated only on the REVEL threshold (<= 0.249), and REVEL is 0.349 - above the threshold and inside the specification's gray zone. Although the SpliceAI score is low, the splicing sub-path does not apply to missense variants because it only rules out one damaging mechanism, so BP4 remains not met. Details ▾Hide ▴
Reasoning
Missense variant. Under the lab's BP4 curator override (stated in vcep_materials gene_notes), exactly one mechanism is evaluated per variant: for missense variants ONLY the missense sub-path applies (REVEL <= 0.249). REVEL = 0.349 > 0.249, and 0.349 falls in the VCEP gray zone (0.249-0.7333); per the override, BP4 is not_met for this variant regardless of SpliceAI. OVERRIDE CHANGES OUTCOME vs. the literal ATM VCEP v1.5 text: the VCEP's splicing sub-path (SpliceAI <= 0.1) would be met here (max delta 0.05 <= 0.1), but the override explicitly forbids applying the splicing sub-path to a missense variant because a clean SpliceAI score only rules out one damaging mechanism (splicing) and does not clear the variant's own protein-level effect, which REVEL is calibrated to assess. The Sun et al. 2025 row's model-derived 'Functional' classification is DeepATM_predicted=Yes (computational prediction) and does not satisfy the REVEL <= 0.249 missense threshold; it is weak computational corroboration only. BP4 not_met.
Rule
BP4 (Benign Supporting): Missense: REVEL score <= .249; Splicing: No predicted impact via splicing (SpliceAI <= 0.1). Lab curator override for missense variants: missense sub-path only (REVEL <= 0.249); splicing sub-path not applied; if REVEL is in the 0.249-0.7333 gray zone or unavailable, BP4 is not_met/not_assessed regardless of SpliceAI.
Evidence considered
- REVEL score 0.349 (missense sub-path; VCEP BP4 threshold <=0.249 not met; gray zone 0.249-0.7333)
- SpliceAI max delta 0.05 (<=0.1 would satisfy the VCEP splicing sub-path but is suppressed for missense variants by the lab BP4 curator override)
- ATM HBOP VCEP v1.5 BP4 rule from cspec plus lab curator BP4 missense/splicing override from vcep_materials gene_notes
- Sun et al. 2025 (PMID 40580951) Table S1 row c.3080A>G: Classification 'Functional', DeepATM_predicted=Yes (computational prediction) - secondary corroboration only, does not meet REVEL threshold
- BayesDel noAF score -0.237229 (not VCEP-calibrated; benign-leaning direction, secondary only)
Still missing
- REVEL 0.349 sits in the VCEP gray zone (0.249-0.7333): the missense sub-path cannot meet BP4 and no functional (kinase/radiosensitivity) or RNA assay result is available to resolve the protein-level effect of p.His1027Arg.
Sources
BP7 N/ABP7 applies to synonymous and deep intronic variants; this is a missense change (p.His1027Arg, exon 21), so it falls outside the criterion's scope. Details ▾Hide ▴
Reasoning
ATM HBOP VCEP v1.5 BP7 is defined for synonymous and deep intronic variants (further than +7 from donor and further than -21 from acceptor sites), with BP7(RNA) applicable to silent substitutions and intronic variants with observed lack of aberrant RNA. The variant under assessment is a missense change (NM_000051.4:c.3080A>G; p.His1027Arg, exon 21), not a synonymous or intronic variant, so BP7 is outside the criterion's scope and is not applicable. No RNA evidence was assessed for BP7.
Rule
BP7 (Benign Supporting): Use for synonymous and deep intronic variants defined as further than (but not including) +7 and further than (but not including) -21 at donor and acceptor sites, respectively. BP7(RNA): Observed lack of aberrant RNA defect for silent substitutions and intronic variants.
Evidence considered
- Variant consequence class: missense (p.His1027Arg) per normalization and Sun et al. 2025 Table S1 (Variant_consequence='Missense', exon_21)
- ATM HBOP VCEP v1.5 BP7 rule from cspec (rule text quoted in 'rule') - scope limited to synonymous/deep intronic variants
Sources
Statement
In silico
SpliceAI predicts no significant splice impact for this variant (max delta score = 0.05). REVEL score = 0.349. BayesDel score = -0.237229.
8 paper records listed. No publication is explicitly linked to an applied criterion in the recorded metadata. None of the listed paper records explicitly indicates that the variant is named.
Recorded publications
-
PMID 25741868 ↗Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.ACMG/AMP 2015 standards and guidelines (Richards et al.), the generic framework from which the ATM VCEP rules derive. Defines PS1 as the same amino acid change produced by a different nucleotide change, which can be assumed pathogenic once a missense at that change is known to be pathogenic, particularly when the mechanism is altered protein function. Also enumerates the criterion weighting system (PS1-4 strong; PM1-6 moderate; PP1-5 supporting; benign BP1-6 supporting). Consulted as baseline definitional reference for PS1 (and, implicitly, the classic PM1/PM5/PP2/BP1 semantics) because the ATM VCEP rule text references standard ACMG/AMP criteria; the full-text extraction available locally contains the PS1 definition and the criteria weighting table but not the individual PM1/PM5/PP2/BP1 definition rows.Does not name this variant
Quoted passage
PS1 Same amino acid change In most cases, when one missense variant is known to be pathogenic, a different nucleotide change that results in the same amino acid [e.g. c.34G>C (p.Val12Leu) and c.34G>T (p.Val12Leu)] can also be assumed to be pathogenic, particularly if the mechanism of pathogenicity is through altered protein function.
Full-text extraction (publications/25741868.txt), criteria definitions section around Tables 3-4 (PS1 definition at approx. char 29259)
-
PMID 28779002 ↗Rare, protein-truncating variants in ATM, CHEK2 and PALB2, but not XRCC2, are associated with increased breast cancer risks.Decker B et al. · 2017Decker et al. 2017, case-control sequencing study of rare protein-truncating variants in ATM, CHEK2 and PALB2 and breast cancer risk (13,087 cases / 5,488 controls, East Anglia). Tagged 'functional' theme in the literature pass, but it is an association study, not a functional assay. Fulltext checked for c.3080A>G / p.His1027Arg / H1027R: the variant is not mentioned, and no functional (PS3/BS3-relevant) data for this variant are reported. Any case/control evidence it might carry belongs to PS4/PS2 groups, not functional.Does not name this variant
Quoted passage
No quote: variant absent from fulltext (0 occurrences of c.3080A>G, H1027, or His1027).
Fulltext searched throughout (variant absent); study covers protein-truncating variants only, whereas c.3080A>G is missense.
-
PMID 34242744 ↗Customizing local and systemic therapies for women with early breast cancer: the St. Gallen International Consensus Guidelines for treatment of early breast cancer 2021.No paper-specific finding is recorded.
-
PMID 17508274 ↗Risk assessment and genetic counseling for hereditary breast and ovarian cancer: recommendations of the National Society of Genetic Counselors.Berliner JL et al. · 2007National Society of Genetic Counselors professional recommendations for risk assessment and genetic counseling for hereditary breast and ovarian cancer. It is a counseling-practice guideline with no variant-level data; NM_000051.4:c.3080A>G (p.His1027Arg) is not mentioned, and it cannot support PS4 or PP5.Does not name this variant
Quoted passage
These cancer genetic counseling recommendations describe the medical, psychosocial and ethical implications of identifying at-risk individuals for hereditary breast and ovarian cancer (HBOC) through cancer risk assessment, with or without genetic susceptibility testing.
Abstract; variant absent from full text (searched c.3080A>G, p.His1027Arg, H1027R, 3080, 1027).
-
PMID 20301317 ↗Hereditary Ataxia Overview.GeneReviews 'Hereditary Ataxia Overview' - a gene-level disease review resource. Full text is not available in the case bundle, and no variant-specific data for NM_000051.4:c.3080A>G (p.His1027Arg) can be verified; it is not citable for any criterion in this group.Does not name this variant
-
PMID 25186627 ↗Frequency of mutations in individuals with breast cancer referred for BRCA1 and BRCA2 testing using next-generation sequencing with a 25-gene panel.Tung N et al. · 2015Tung et al. 2015 (frequency of mutations in individuals with breast cancer referred for BRCA1/2 testing using a 25-gene NGS panel). This paper is cited by ClinVar submissions for the VUA (SCV001363782, SCV000276398, SCV000219029) and was screened as the most likely literature source for a codon-1027 observation. Full-text screening found no mention of c.3080A>G, c.3082A>G, residue 1027, or p.His1027Arg, so it provides no same-amino-acid or same-residue comparator evidence and is not cited for any criterion.Does not name this variant
Quoted passage
No text matching c.3080A>G, c.3082A>G, residue 1027, or p.His1027Arg is present in the full-text extraction (grep for '3080|3082|1027|His1027' across publications/ returned no matches) - negative screen result.
publications/25186627.txt (full-text extraction)
-
PMID 26389210 ↗Genetics of Breast and Gynecologic Cancers (PDQ®): Health Professional Version.PDQ 'Genetics of Breast and Gynecologic Cancers' - a gene-level cancer genetics summary resource. Full text is not available in the case bundle, and no variant-specific data for NM_000051.4:c.3080A>G (p.His1027Arg) can be verified; it is not citable for any criterion in this group.Does not name this variant
-
PMID 32885271 ↗Multigene panel testing for hereditary breast and ovarian cancer in the province of Ontario.Lerner-Ellis J et al. · 2021Lerner-Ellis et al. 2021 (multigene panel testing for hereditary breast and ovarian cancer in Ontario). Cited by ClinVar submissions for the VUA (SCV000276398, SCV001354067) and screened for any residue-1027 observation. Full-text screening found no mention of c.3080A>G, c.3082A>G, residue 1027, or p.His1027Arg, so it provides no same-amino-acid or same-residue comparator evidence and is not cited for any criterion.Does not name this variant
Quoted passage
No text matching c.3080A>G, c.3082A>G, residue 1027, or p.His1027Arg is present in the full-text extraction (grep for '3080|3082|1027|His1027' across publications/ returned no matches) - negative screen result.
publications/32885271.txt (full-text extraction)
Screenshots represent source records captured during evidence collection. Select one to view the recorded source content.
gnomAD v4.1
10 / 1,610,692 recorded alleles
Open source ↗
gnomAD v2.1
Captured gnomAD v2.1 source record
Open source ↗
SpliceAI Lookup
Recorded maximum delta 0.050
Open source ↗
ClinVar
9 Uncertain significance
Open source ↗
COSMIC
Captured COSMIC source record
Open source ↗
Cancer hotspots
Captured Cancer Hotspots source record
Open source ↗
OncoKB
Captured OncoKB source record
Open source ↗
ClinGen CSpec
Captured ClinGen CSpec source record
Open source ↗
All source links
- ClinGen CSpec ↗cspec.genome.network/cspec/ui/svi/doc/639508985
- ClinVar ↗www.ncbi.nlm.nih.gov/clinvar/variation/188327/
- gnomAD v2.1 ↗gnomad.broadinstitute.org/variant/11-108143261-A-G?dataset=gnom…
- gnomAD v4.1 ↗gnomad.broadinstitute.org/variant/chr11-108272534-A-G?dataset=g…
- SpliceAI Lookup ↗spliceailookup.broadinstitute.org/#variant=NM_000051.4%3Ac.3080…
- OncoKB ↗www.oncokb.org/gene/ATM/H1027R
- COSMIC ↗cancer.sanger.ac.uk/cosmic/search?q=ATM+c.3080A%3EG
- Cancer hotspots ↗www.cancerhotspots.org/
Statements for reports
The evidence statements recorded by the interpretation pipeline. They have not been reinterpreted by the page.
gnomAD v4.1
This variant is present in gnomAD v4.1 (AF= 6.20851e-06; MAF= 0.00062%, 10/1610692 alleles, homozygotes = 0) and has highest observed frequency in the Remaining individuals population (AF= 1.60277e-05; MAF= 0.00160%, 1/62392 alleles, homozygotes = 0); grpmax FAF= 3.6e-06.
gnomAD v2.1
This variant is present in gnomAD v2.1 (AF= 3.98159e-06; MAF= 0.00040%, 1/251156 alleles, homozygotes = 0) and has highest observed frequency in the European (non-Finnish) population (AF= 8.80514e-06; MAF= 0.00088%, 1/113570 alleles, homozygotes = 0).
gnomAD-Canada
Absent from gnomAD-Canada v1.0.
ClinVar
This variant has been reported in ClinVar as Uncertain significance (9 clinical laboratories). (ClinVarID = 188327)
In silico
SpliceAI predicts no significant splice impact for this variant (max delta score = 0.05). REVEL score = 0.349. BayesDel score = -0.237229.
Functional
OncoKB did not identify variant-specific reviewed functional evidence for this variant; gene-level curated context is available for reviewer follow-up. ATM, a kinase involved in the DNA damage response, is mutated in various solid and hematologic malignancies.
Somatic
This variant does not lie in a statistically significant hotspot. This variant has not previously been reported in somatic cancers (COSMIC).
Cancer Hotspots
This variant does not lie in a statistically significant hotspot.