Five-step scale from benign to pathogenic. This variant is uncertain. Select a step to read what it means.
This variantScenario
BenignPathogenic
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.
About FANCD2
FANCD2 encodes a DNA repair protein that works together with its partner FANCI to coordinate the cellular response to DNA damage and maintain genome stability. Inherited mutations in FANCD2 cause Fanconi anemia, a rare disorder characterized by congenital defects, bone marrow failure, and an elevated risk of cancer. FANCD2 acts as a tumor suppressor: loss of its function impairs DNA repair and can promote cancer development, and altered FANCD2 activity has been observed in several tumor types, including leukemias, breast cancer, melanoma, and colorectal cancer.
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 Not recorded. 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.
Met (supporting): variant is absent (AF 0) from gnomAD v2.1, v4.1, and gnomAD-Canada, below the <0.1% PM2 threshold.
Benign criteria applied · 0
No benign criterion marked as applied in the recorded assessment.
Recorded classification: Uncertain significance
Variant of Uncertain Significance: only PM2 (supporting) was met — the variant is absent (AF 0) from gnomAD — and one supporting criterion meets no Pathogenic, Likely Pathogenic, Benign, or Likely Benign threshold.
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
Not recorded
gnomAD v4.1 allele counts were not recorded for this variant.
ACMG/AMP criteria under Not recorded, with the data and sources behind each one.
28 criteria with recorded statuses: 1 met · 7 not met · 12 not assessed · 8 not applicable. Open any row for the full reasoning, the rule and the evidence behind it.
Governing framework: no ClinGen CSPEC/VCEP specification exists for FANCD2 (cspec found = false; vcep_materials empty), so the generic ACMG/AMP 2015 framework (PMID:25741868) applies.
Variant under review: NM_033084.4:c.3560+2T>A, a canonical +2 splice-donor variant in intron 35 of FANCD2.
The segregation/de novo evidence group cannot be scored: the case bundle contains no proband phenotype, no parental testing results, and no family segregation data for this variant.
No external evidence exists either: the variant is absent from ClinVar, absent from gnomAD v2.1/v4.1/Canada (frequency data, not segregation data), literature_pass found 0 PMIDs, fulltext_map is empty, and no publications directory content was available.
All four criteria (PS2, PM6, PP1, BS4) are returned as not_assessed because the required proband/family observations are entirely absent; absence of data must not be interpreted as either supporting or refuting any of these criteria.
Gene-context PVS1 reference papers present in the workspace (e.g., PMID:27285993, PMID:29376519) do not mention this variant and provide no segregation or de novo observations; they are not cited for this group.
No FANCD2 ClinGen CSPEC/VCEP specification was available (cspec found: false; framework_mode: generic_acmg), so PS3/BS3 were adjudicated under the generic ACMG/AMP 2015 rules (PMID 25741868).
No well-established functional assay evidence exists for NM_033084.4:c.3560+2T>A: literature pass returned 0 PMIDs, the fulltext map and publications/ directory are empty, and the variant is absent from ClinVar and COSMIC.
SpliceAI (max delta 0.99) is an in silico splice predictor and was considered but cannot be counted as functional evidence for PS3 or BS3; it neither establishes a deleterious functional effect in an assay nor demonstrates absence of damage.
The pvs1_gene_context supporting papers provide only gene-level germline loss-of-function context for FANCD2 (e.g., the gene's role in the FA pathway) and do not report functional assays of c.3560+2T>A, so they were not cited for PS3/BS3.
Both PS3 and BS3 are returned as not_assessed with the identified gap being the complete absence of variant-specific functional data; no functional criterion contributes to the classification.
Group-level: no criterion in this group (PS4, PP4, PP5, BP5, BP6) is met. PP5 and BP6 are not_met deterministically because the variant has no ClinVar record and therefore no exact-variant expert-panel assertion exists. PS4, PP4, and BP5 are not_assessed because the case bundle contains no case-control data, no proband phenotype/family history, and no proband-level genetic or diagnostic data, respectively. This group contributes no pathogenic or benign scoring evidence to the final classification, and its unresolved criteria do not alter the ACMG/AMP 2015 combination tallies.
No CSPEC/VCEP specification exists for FANCD2 (vcep_materials.json: 'no cspec url available'; final_classification_framework source=generic_acmg_fallback), so generic ACMG/AMP 2015 (PMID 25741868) governs all criteria in this group.
The variant NM_033084.4:c.3560+2T>A is a canonical splice donor variant in FANCD2 (FA complementation group D2, OMIM 613984), a gene in which biallelic loss-of-function causes autosomal recessive Fanconi anemia; PM3 is therefore the applicable in-trans criterion and BP2's cis clause is applicable in principle.
Both allelic criteria are returned as not_assessed: the case bundle contains no affected-proband observation, no second FANCD2 allele, and no phase (cis/trans) information. The variant is absent from ClinVar, gnomAD v2.1/v4.1/Canada, and COSMIC, and the literature pass identified zero PMIDs, so no source provides allele-configuration evidence.
Neither criterion contributes toward the final classification at this time; if proband/pedigree, segregation, or phase-resolved second-variant data become available, PM3 and BP2 must be re-adjudicated.
BayesDel score 0.66 and SpliceAI max delta 0.99 were noted but are irrelevant to allelic criteria and were not used.
Framework: no ClinGen CSPEC, no local gene-specific framework, and no VCEP materials were available for FANCD2 (vcep_materials.json: no cspec url, files_found empty, gene_vcep_dir null); generic ACMG/AMP 2015 (PMID 25741868) applies.
PP3 not met: SpliceAI max delta 0.99 strongly predicts splice disruption, but this canonical +2 donor variant's splice-impact evidence is evaluated under the ClinGen SVI PVS1 framework (PMC6185798) and must not be double-counted as PP3; no independent calibrated computational line (REVEL absent; BayesDel uncalibrated/not validly applicable) supports PP3 on its own merits.
BP4 not met: all available computational evidence indicates splice impact (SpliceAI 0.99 at the canonical donor); no calibrated benign-supporting computational evidence exists for this variant.
No publications were reviewed by this group: literature_pass.json returned 0 PMIDs, fulltext_map.json is empty, and the publications/ directory contains no files; the evidence base is database lookups (SpliceAI, BayesDel, REVEL) and the named framework publications (PMID 25741868, PMID 30192042/PMC6185798, PMID 36413997, PMID 30661751) cited only as calibration/rule references.
Net contribution of this group: no supporting or benign computational evidence added (PP3 not met, BP4 not met), avoiding double-counting of the splice prediction that the PVS1 group evaluates.
No ClinGen CSPEC/VCEP specification exists for FANCD2 (cspec found=false, gene_vcep_dir=null), so generic ACMG/AMP 2015 (PMID 25741868) population criteria apply, with local gnomAD non-VCEP thresholds (PM2 <0.1%, BS1 >0.3%, BA1 >1%).
The variant NM_033084.4:c.3560+2T>A is absent from all three queried population datasets (gnomAD v2.1 exomes, gnomAD v4.1 exomes, gnomAD-Canada v1.0 genomes); no allele counts, frequencies, or homozygote observations were returned.
BA1 not met (AF = 0, far below >1% threshold); BS1 not met (AF = 0, far below >0.3% threshold, and Fanconi anemia is a very rare AR disorder with a low expected carrier frequency); BS2 not met (no homozygous observations; severe early-onset AR disease makes healthy adult homozygotes implausible).
PM2 met at supporting strength: absence across three independent population datasets (both GRCh37 and GRCh38 coordinates, exome and genome cohorts) is below the <0.1% threshold for recessive disorders, and the variant is a canonical splice-donor change well captured by these datasets.
Population evidence therefore contributes a single supporting pathogenic criterion (PM2) and no benign population-frequency criteria.
No criteria match.
Met · 1
PM2SupportingMet (supporting): variant is absent (AF 0) from gnomAD v2.1, v4.1, and gnomAD-Canada, below the <0.1% PM2 threshold.
Details ▾Hide ▴
Reasoning
PM2 (supporting) is met: the variant is absent (AF = 0) from gnomAD v2.1 exomes (GRCh37), gnomAD v4.1 exomes (GRCh38), and gnomAD-Canada v1.0 HostSeq genomes — well below the local gnomAD non-VCEP PM2 threshold of <0.1% for recessive disorders. Absence is consistent across three independent population datasets covering both genome builds and both exome and genome cohorts. The variant is a canonical +2 donor splice-site change (SpliceAI max delta 0.99), and splice-region variants are well captured in these datasets, so the absence is meaningful. No ClinVar record exists for this variant, and there is no conflicting population-frequency signal.
Rule
Generic ACMG/AMP 2015 (PMID 25741868) PM2: absent in controls or at extremely low frequency in recessive disorders (local gnomAD non-VCEP threshold <0.1%); observed AF = 0 in gnomAD v2.1/v4.1 and gnomAD-Canada, therefore met at supporting strength.
gnomAD v4.1 genome dataset was not separately queried (v4.1 exome dataset used); a genome-dataset confirmation would marginally strengthen the absence call.
Per-position coverage/depth at the intronic site was not captured; low coverage at this exact position cannot be formally excluded from the captured evidence.
PP3Not met: SpliceAI max delta 0.99 predicts near-certain splice impact, but this same prediction is counted under PVS1 and cannot be double-counted as PP3.
Details ▾Hide ▴
Reasoning
Variant NM_033084.4:c.3560+2T>A is a canonical +2 donor splice-site variant (intron 35; variant_exonic_positions 35i). SpliceAI predicts a near-certain splice impact (max delta = 0.99; DS_DL 0.99, DS_AL 0.61), far above the tool's published high-recall (0.2) and high-precision (0.5) thresholds (Jaganathan et al., Cell 2019, PMID 30661751). However, this same splice-effect prediction is the evidence evaluated under the ClinGen SVI PVS1 framework for canonical +1/+2 splice variants (PMC6185798, PMID 30192042), and stacking PP3 on the identical prediction would double-count it; the case's PVS1 scaffold likewise directs 'do not stack PP3 for the same splice-effect prediction evidence.' No independent, calibrated computational line exists to support PP3 on its own merits: REVEL returned no score for this variant (not found in local REVEL v1.3), and the BayesDel noAF score (0.66) is treated as insufficiently calibrated for PP3/BP4 (no verified published threshold with named publication/table is available to this pipeline) and in any case is a missense predictor applied to an intronic splice variant, which is not a valid application. The ClinGen SVI PP3/BP4 calibration paper (Pejaver et al. 2022, PMID 36413997) calibrated missense predictors and did not provide a PP3/BP4 calibration for splice predictors. PP3 is therefore not met; the computational splice evidence is owned by the PVS1 evaluation and must not be double-counted here.
Rule
Generic ACMG/AMP 2015 PP3 (Richards et al., PMID 25741868): 'Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.).' Caveats applied: algorithms sharing inputs are not independent lines; computational splice-impact evidence already counted under PVS1 for a canonical +1/+2 splice variant must not be double-counted as PP3 (ClinGen SVI PVS1 framework, PMC6185798).
Evidence considered
SpliceAI lookup (source_registry key: spliceai): max delta score 0.99 (DS_DL=0.99, DS_AL=0.61, DS_DG=0.18, DS_AG=0.0) for NM_033084.4:c.3560+2T>A; exceeds the SpliceAI publication's high-recall (0.2) and high-precision (0.5) thresholds (Jaganathan et al., Cell 2019; PMID 30661751). Score used descriptively only; not applied as PP3 because the same splice-effect prediction is subsumed by the PVS1 evaluation for canonical +1/+2 splice variants and double-counting the same prediction across criteria is prohibited.
BayesDel noAF lookup (source_registry key: bayesdel): score 0.66 for chr3:10130228 T>A. Treated as not_available for PP3/BP4: no verified published threshold with a nameable publication and table is available to this pipeline, and BayesDel is a missense pathogenicity predictor that is not validly applied to an intronic splice variant.
REVEL: no score returned for this variant (prefetch.json revel.found=false; local REVEL v1.3 data lacks the record), so the ClinGen SVI-calibrated REVEL thresholds (Pejaver et al., Am J Hum Genet 2022; PMID 36413997) were not applied.
Generic ACMG/AMP 2015 rules (source_registry key: generic_acmg_combination_rules; Richards et al. 2015; PMID 25741868): PP3 requires multiple lines of computational evidence supporting a deleterious effect, with the caveat that in silico tools sharing inputs are not independent and must not be double-counted with each other or with overlapping functional/splice evidence.
ClinGen SVI PVS1 recommendations (source_registry key: pvs1_generic_framework; PMC6185798; Tayoun et al., Hum Mutat 2018; PMID 30192042): canonical +1/+2 splice variants are evaluated under the generic PVS1 framework, and the same splice-impact prediction must not be additionally counted as PP3.
Still missing
No independent second line of calibrated computational evidence: REVEL score not found for this variant (local REVEL v1.3 lookup returned no record).
BayesDel noAF score 0.66 cannot be used because no verified published PP3/BP4 threshold (named publication and table) is available to this pipeline.
SpliceAI has no ClinGen SVI-calibrated PP3/BP4 threshold (Pejaver et al. 2022, PMID 36413997 calibrated missense predictors only); its use here is descriptive and subsumed by the PVS1 splice-site evaluation.
PP5Not met: ClinVar has no record for this variant, so no expert-panel Pathogenic or Likely pathogenic classification exists to support PP5.
Details ▾Hide ▴
Reasoning
PP5 is applied only for an exact-variant ClinVar expert-panel classification of Pathogenic or Likely pathogenic (task-specified gate; ordinary laboratory submissions, aggregate labels, and non-expert assertions never trigger it). ClinVar has no record for NM_033084.4:c.3560+2T>A: exact-match search returned no_result, variation_ids is empty, and there are no submissions or expert-panel entries. Because no expert-panel Pathogenic/Likely pathogenic classification exists for this exact variant, PP5 is not met.
BA1Not met: variant is absent (AF 0) from population databases, far below the >1% BA1 threshold.
Details ▾Hide ▴
Reasoning
BA1 requires an allele frequency above the benign threshold in reference population databases (generic ACMG/AMP 2015, PMID 25741868; local gnomAD non-VCEP threshold >1%). The variant NM_033084.4:c.3560+2T>A (FANCD2) is absent (allele count 0, AF = 0) from gnomAD v2.1 exomes (GRCh37 3-10130228-T-A), gnomAD v4.1 exomes (GRCh38 3-10088544-T-A), and gnomAD-Canada v1.0 HostSeq genomes. Observed AF is far below any BA1 threshold, so BA1 is not met.
Rule
Generic ACMG/AMP 2015 (PMID 25741868) BA1: allele frequency >1% (local gnomAD non-VCEP threshold) in reference population databases; observed AF = 0, therefore not met.
Absence calls rest on gnomAD variant-page lookups; per-position coverage/depth metrics for this intronic site were not separately captured, though splice-region variants are generally well covered in gnomAD exome datasets.
BS1Not met: variant is absent (AF 0), far below the >0.3% BS1 threshold expected for this very rare recessive disorder.
Details ▾Hide ▴
Reasoning
BS1 requires an allele frequency greater than expected for the disorder. FANCD2-related Fanconi anemia (OMIM 613984) is a very rare autosomal recessive disorder, so the expected carrier allele frequency is far below the local gnomAD non-VCEP BS1 threshold of >0.3%. The variant is absent (AF = 0) from gnomAD v2.1, gnomAD v4.1, and gnomAD-Canada v1.0, i.e., far below any plausible BS1 threshold. BS1 is not met.
Rule
Generic ACMG/AMP 2015 (PMID 25741868) BS1: allele frequency greater than expected for the disorder (local gnomAD non-VCEP threshold >0.3%); observed AF = 0, therefore not met.
No disease-specific (Fanconi anemia) population carrier-frequency estimate was available in the evidence bundle to compute an explicit 'greater than expected for disorder' comparison; the observed AF of 0 makes the determination unambiguous regardless.
BS2Not met: no homozygous observations exist — the variant is absent from all queried population databases, and biallelic loss causes severe early-onset disease.
Details ▾Hide ▴
Reasoning
BS2 requires observation of the variant in homozygous state in a healthy adult. The variant is absent from all queried population databases (gnomAD v2.1, gnomAD v4.1, gnomAD-Canada v1.0), so no homozygous observations exist to support BS2. Moreover, biallelic FANCD2 loss-of-function causes a severe, early-onset autosomal recessive disorder, so healthy adult homozygotes would not be expected in population cohorts. BS2 is not met.
Rule
Generic ACMG/AMP 2015 (PMID 25741868) BS2: variant observed in homozygous state in a healthy adult; no homozygotes observed, therefore not met.
Evidence considered
gnomAD v2.1 (exome): variant absent, no homozygous observations
gnomAD v4.1 (exome): variant absent, no homozygous observations
gnomAD-Canada v1.0 (HostSeq genomes): variant absent, no homozygous observations
Absence of homozygotes is expected for a severe autosomal recessive disorder and cannot exclude rare homozygotes in unsampled populations; no targeted healthy-carrier cohort data for FANCD2 were available.
BP4Not met: SpliceAI max delta 0.99 predicts near-certain splice impact, the opposite of the no-impact evidence BP4 requires.
Details ▾Hide ▴
Reasoning
BP4 requires multiple lines of computational evidence suggesting no impact on the gene or gene product. The only splice-relevant computational evidence available points in the opposite direction: SpliceAI predicts near-certain donor-loss splicing impact (max delta = 0.99) at this canonical +2 donor position. No benign-supporting computational line exists: REVEL returned no score; the BayesDel noAF score (0.66) is not usable (no verified published PP3/BP4 threshold available to this pipeline; and as a missense predictor it is not validly applied to an intronic splice variant), and even directionally it is not benign-supporting. BP4 is therefore not met.
Rule
Generic ACMG/AMP 2015 BP4 (Richards et al., PMID 25741868): 'Multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.).' Caveat applied: algorithms sharing inputs are not independent lines; no computational line here suggests no impact.
Evidence considered
SpliceAI lookup (source_registry key: spliceai): max delta score 0.99 (DS_DL=0.99) indicates donor-loss splice impact at c.3560+2T>A — evidence AGAINST no-impact, so it cannot support BP4.
BayesDel noAF lookup (source_registry key: bayesdel): score 0.66; treated as not_available for PP3/BP4 (no verified published threshold with nameable publication/table) and not validly applicable to an intronic splice variant; directional value is not benign-supporting in any case.
REVEL: no score returned for this variant (prefetch.json revel.found=false), so no benign-threshold (BP4) evaluation was possible.
Generic ACMG/AMP 2015 rules (source_registry key: generic_acmg_combination_rules; Richards et al. 2015; PMID 25741868): BP4 requires multiple lines of computational evidence suggesting no impact; absent any such line, BP4 is not met.
Still missing
No calibrated benign-supporting computational evidence was available: REVEL score not found; BayesDel unverifiable/uncalibrated for PP3/BP4; no SVI-calibrated splice-predictor BP4 threshold exists (Pejaver et al. 2022, PMID 36413997).
No population-frequency or functional benign evidence falls within this group's scope; BP4 rests solely on in silico predictions, and none support no impact.
BP6Not met: ClinVar has no record for this variant, so no expert-panel Benign or Likely benign classification exists to support BP6.
Details ▾Hide ▴
Reasoning
BP6 is applied only for an exact-variant ClinVar expert-panel classification of Benign or Likely benign (task-specified gate; ordinary laboratory submissions, aggregate labels, and non-expert assertions never trigger it). ClinVar has no record for NM_033084.4:c.3560+2T>A: exact-match search returned no_result, variation_ids is empty, and there are no submissions or expert-panel entries. Because no expert-panel Benign/Likely benign classification exists for this exact variant, BP6 is not met.
PVS1Not assessed: insufficient evidence was available to evaluate whether this splice variant causes loss of function.
Details ▾Hide ▴
Reasoning
PVS1 could not be assessed because its agent did not produce valid output.
Still missing
attempt 1: invalid criterion strength: pvs1
attempt 2: invalid criterion strength: pvs1
attempt 3: evidence_refs must contain only source_registry keys
PS2Not assessed: no proband or parental-testing data were available to confirm or exclude a de novo origin.
Details ▾Hide ▴
Reasoning
PS2 requires a confirmed de novo occurrence (maternity and paternity confirmed) in a proband with a consistent phenotype. The case evidence bundle contains no proband phenotype description and no parental testing results: ClinVar has no entry for NM_033084.4:c.3560+2T>A, the literature pass returned 0 PMIDs, and no variant-specific publication or full text was retrieved (fulltext_map empty). De novo origin cannot be confirmed or refuted from any available source, so PS2 is not assessable.
Rule
PS2: de novo (maternity and paternity confirmed) - Richards et al. 2015 ACMG/AMP framework (PMID:25741868)
Still missing
No proband genotype/phenotype data available in this case.
No parental testing results (neither parent genotyped for the variant); de novo occurrence cannot be confirmed or excluded.
No variant-specific publication reporting this variant as de novo was found (literature_pass total_pmids = 0; ClinVar absent).
PS3Not assessed: no functional assay evidence was available; the only relevant datum was an in silico splice prediction, not a functional assay.
Details ▾Hide ▴
Reasoning
No VCEP/CSPEC exists for FANCD2 (cspec not found; framework_mode = generic_acmg), so PS3 is evaluated under the generic ACMG/AMP 2015 rule: well-established functional studies showing a deleterious effect (PMID 25741868). For NM_033084.4:c.3560+2T>A (canonical +2 donor of intron 35), no well-established functional assay evidence is available in this case: literature_pass.json returned 0 PMIDs, the fulltext_map is empty, the publications/ directory contains no articles, ClinVar has no record for this variant, and COSMIC has no entry. The pvs1_gene_context supporting papers (PMID:27285993, PMID:29659569, PMID:41440239, PMID:29376519, PMID:31346352) provide only gene-level germline loss-of-function context for FANCD2 and do not report functional assays of c.3560+2T>A; they cannot be used as PS3 evidence for this specific variant. The only variant-level functional-adjacent datum is the SpliceAI in silico prediction (max delta 0.99, DS_DL 0.99), which is a computational splice predictor, not a well-established in vitro/in vivo functional assay, and per the functional-criteria calibration it cannot substitute for PS3. PS3 is therefore not assessable due to absent functional evidence.
Rule
PS3: Well-established functional studies show a deleterious effect (ACMG/AMP 2015, Richards et al., PMID 25741868; generic fallback, no FANCD2 VCEP/CSPEC available)
Evidence considered
SpliceAI lookup (source key: spliceai): max delta 0.99 (DS_DL 0.99) predicts disruption of the canonical donor site; reviewed and excluded as PS3 evidence because it is an in silico prediction without a validated functional assay component.
Case-level negative evidence: variant absent from ClinVar and COSMIC; literature_pass.json total_pmids = 0; fulltext_map.json empty; publications/ directory empty; no functional study of this variant identified in any source.
Still missing
No in vitro or in vivo functional studies (e.g., patient-cell FANCD2 monoubiquitination/ID2 complex assay, MMC/cisplatin sensitivity assay, or minigene/RT-PCR splicing assay) of c.3560+2T>A are available; such studies demonstrating aberrant splicing or loss of FANCD2 function would be required to apply PS3.
PS4Not assessed: no case-control or cohort data reported this variant — it is absent from ClinVar, gnomAD, and the literature.
Details ▾Hide ▴
Reasoning
PS4 (Richards et al. 2015, PMID:25741868) requires statistically significant enrichment of the exact variant in affected cases versus controls. No case-control or cohort data report NM_033084.4:c.3560+2T>A: the variant is absent from ClinVar (search_status=no_result; no record, no submissions), absent from gnomAD v2.1, v4.1, and gnomAD-Canada, and the literature pass returned 0 PMIDs (literature_pass.json note: 'No PMIDs found across evidence sources'). With zero affected-case observations and no comparison cohort, variant-level case enrichment cannot be evaluated. Absence of population-frequency data (gnomAD) is prevalence context only and does not substitute for case-control enrichment. No governing VCEP/CSPEC exists for FANCD2 (final_classification_framework found=false, generic ACMG fallback applies).
Rule
PS4 (Richards et al. 2015, PMID:25741868): case-control enrichment of the exact variant in affected individuals
Evidence considered
ClinVar exact-variant search: search_status=no_result, variation_ids empty, no submissions (evidence.json clinvar)
PM3Not assessed: no proband, second-allele, or phase data were available to test for a pathogenic trans counterpart.
Details ▾Hide ▴
Reasoning
No governing CSPEC/VCEP exists for FANCD2, so generic ACMG/AMP 2015 (PMID 25741868) applies: PM3 requires detection of the variant in trans with a pathogenic variant in an affected proband for a recessive disorder. FANCD2 (FA complementation group D2, OMIM 613984) causes autosomal recessive Fanconi anemia, so PM3 is in principle applicable. However, the case bundle contains no proband/pedigree data, no second allele on the other chromosome, and no phase information: the variant is absent from ClinVar (no submissions, no reported allele observations), gnomAD v2.1/v4.1/Canada, and COSMIC, and the literature pass identified zero PMIDs (literature_pass.json total_pmids=0; fulltext_map.json empty). With no affected-proband trans-observation available, PM3 cannot be evaluated and is returned as not_assessed pending proband/pedigree and phase data.
Rule
PM3 (Richards et al. 2015, PMID 25741868): For recessive disorders, detected in trans with a pathogenic variant. Not evaluable because no affected-proband trans-observation of a second pathogenic allele is reported.
Evidence considered
case_summary.json: framework_mode=generic_acmg; no CSPEC/VCEP for FANCD2
case_summary.json clinvar section: variant absent from ClinVar, no submissions, no criterion leads, no PMIDs
evidence.json clinvar: search_status=no_result; evidence_sentence 'This variant is absent from ClinVar.'
evidence.json gnomad (v2.1, v4.1, Canada): variant absent in all three datasets
literature_pass.json: total_pmids=0, criteria_index empty; no proband-level allelic observations identified
No proband/pedigree data (affected status, family history) provided in the case bundle
No second FANCD2 allele identified in the proband to establish trans configuration
No phase information (cis/trans) available for this variant relative to any other variant
No ClinVar submissions or literature reports documenting an affected individual carrying this variant
No evidence the proband has biallelic FANCD2 loss of function consistent with recessive Fanconi anemia
Sources
generic_acmg_combination_rules
PM6Not assessed: no proband or parental testing data were available to evaluate an assumed de novo occurrence.
Details ▾Hide ▴
Reasoning
PM6 applies to an assumed de novo occurrence without confirmation of parental identity. No proband or parental testing data are present in the evidence bundle, and no variant-specific literature or ClinVar submission reports this variant as assumed de novo. Phenotype consistency with FANCD2-related Fanconi anemia also cannot be evaluated because no clinical phenotype for the proband was provided. PM6 is therefore not assessable.
Rule
PM6: assumed de novo, without confirmation of paternity and maternity - Richards et al. 2015 ACMG/AMP framework (PMID:25741868)
Still missing
No proband testing data and no parental testing results available.
No evidence that the variant arose de novo (assumed or confirmed) from any source.
Proband phenotype not provided, so phenotype-consistency requirement cannot be checked.
No variant-specific publication or ClinVar entry documenting this variant exists.
PP1Not assessed: no pedigree or family segregation data were available.
Details ▾Hide ▴
Reasoning
PP1 requires co-segregation of the variant with disease in multiple affected family members in a gene definitively known to cause the disease. The evidence bundle contains no pedigree or segregation data: no affected relatives were genotyped and no family study was reported. ClinVar has no entry for this variant and no variant-specific publication was retrieved. With no segregation observations at all, PP1 cannot be scored.
Rule
PP1: co-segregation with disease in affected family members in a gene definitively known to cause the disease - Richards et al. 2015 ACMG/AMP framework (PMID:25741868)
Still missing
No family segregation data (no affected relatives genotyped for the variant).
No pedigree information in the case bundle.
No variant-specific publication reporting segregation of c.3560+2T>A with disease.
PP4Not assessed: no proband phenotype or family history data were available to assess phenotype specificity.
Details ▾Hide ▴
Reasoning
PP4 (Richards et al. 2015, PMID:25741868) requires the proband's phenotype or family history to be highly specific for a single genetic etiology (for FANCD2, Fanconi anemia features such as congenital anomalies, bone marrow failure, and cancer predisposition). The evidence bundle contains no proband phenotype, clinical presentation, or family history data, so phenotype specificity for the carrier of this variant cannot be evaluated. Gene-level knowledge that FANCD2 loss of function causes Fanconi anemia (pvs1_gene_context) documents the disease association but does not establish that any carrier of c.3560+2T>A exhibits that phenotype; PP4 cannot be inferred from gene-level disease context alone, and no case report describing a carrier of this exact variant was retrieved.
Rule
PP4 (Richards et al. 2015, PMID:25741868): phenotype or family history highly specific for a gene with a single genetic etiology
Evidence considered
Case bundle contains no proband phenotype or family-history fields (case_summary.json compact_evidence; evidence.json)
pvs1_gene_context.json: germline disease context found (Fanconi anemia) via literature-only sources; no carrier-phenotype data for this variant
BS3Not assessed: no well-established functional study demonstrating a benign effect was available for this variant.
Details ▾Hide ▴
Reasoning
BS3 is evaluated under the generic ACMG/AMP 2015 rule: well-established functional studies showing no damaging effect (PMID 25741868), because no FANCD2 VCEP/CSPEC is available. No functional assay evidence exists for NM_033084.4:c.3560+2T>A in this case: 0 PMIDs in the literature pass, empty fulltext map, empty publications/ directory, and no ClinVar/COSMIC record. The available in silico SpliceAI prediction (max delta 0.99) in fact predicts a strong splicing impact, which conflicts with any no-damaging-effect hypothesis and in any case is not a functional assay; it therefore cannot support BS3. BS3 is not assessable because no well-established functional study demonstrating normal/benign function of this variant has been identified.
Rule
BS3: Well-established functional studies show no damaging effect (ACMG/AMP 2015, Richards et al., PMID 25741868; generic fallback, no FANCD2 VCEP/CSPEC available)
Evidence considered
SpliceAI lookup (source key: spliceai): max delta 0.99 (DS_DL 0.99) predicts donor-site disruption; reviewed and excluded as BS3 evidence because it is in silico (not a functional assay) and its direction opposes a no-damaging-effect conclusion.
Case-level negative evidence: no functional study of this variant identified in ClinVar, COSMIC, literature_pass.json (0 PMIDs), fulltext_map.json (empty), or publications/ (empty).
Still missing
No functional assay demonstrating that c.3560+2T>A retains normal FANCD2 splicing or function (e.g., RT-PCR/minigene showing wild-type transcript, or normal FANCD2 monoubiquitination and DNA-damage response) is available; such evidence would be required to apply BS3.
BS4Not assessed: no family segregation testing was performed, so lack of segregation could not be observed.
Details ▾Hide ▴
Reasoning
BS4 requires observed lack of segregation of the variant with disease in affected family members. No segregation testing of any family member is present in the case bundle, so there is no observation of non-segregation. Absence of data must not be treated as evidence of non-segregation; BS4 cannot be applied.
Rule
BS4: lack of segregation in affected family members - Richards et al. 2015 ACMG/AMP framework (PMID:25741868)
Still missing
No genotyping of affected family members available; non-segregation has not been observed.
No pedigree or family testing data in the case bundle.
Absence of the variant from ClinVar and gnomAD does not constitute segregation evidence and cannot support BS4.
BP2Not assessed: no second variant or phase-resolved observation was available to evaluate a cis/trans configuration.
Details ▾Hide ▴
Reasoning
Under the governing generic ACMG/AMP 2015 framework (PMID 25741868; no FANCD2 CSPEC/VCEP exists), BP2 applies when the variant is observed in trans with a pathogenic variant for a fully penetrant dominant disorder, or observed in cis with a pathogenic variant in any inheritance pattern. FANCD2-associated Fanconi anemia is autosomal recessive, so the dominant-trans clause is not relevant, but the cis clause could in principle apply. However, the case bundle contains no variant-variant observations whatsoever: no second FANCD2 variant is reported, no phase (cis/trans) is established for any observation, the variant is absent from ClinVar (no submissions), gnomAD v2.1/v4.1/Canada, and COSMIC, and literature_pass found zero PMIDs. With no observed allele configuration, BP2 cannot be evaluated and is returned as not_assessed pending phase-resolved observations.
Rule
BP2 (Richards et al. 2015, PMID 25741868): Observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder; or observed in cis with a pathogenic variant in any inheritance pattern. Not evaluable because no phase-resolved observation of this variant with a second variant exists.
Evidence considered
case_summary.json: framework_mode=generic_acmg; no CSPEC/VCEP for FANCD2
case_summary.json clinvar section: variant absent from ClinVar, no submissions, no PMIDs
evidence.json clinvar: search_status=no_result; no allele-configuration observations
evidence.json gnomad (v2.1, v4.1, Canada): variant absent in all three datasets
evidence.json cosmic: no somatic record; no cis/trans allele-pair evidence
literature_pass.json: total_pmids=0; no publications reporting this variant in any phase configuration
Still missing
No observation of this variant in cis with any pathogenic variant (no compound-allele or single-haplotype data)
No phase information (cis/trans) available from any proband, family, or population source
No ClinVar submissions or literature reports describing this variant together with a second FANCD2 variant
No proband/pedigree data provided to permit allele-configuration assessment
No population-level observation of this variant at all (absent from gnomAD v2.1/v4.1/Canada)
Sources
generic_acmg_combination_rules
BP5Not assessed: no proband genotype or diagnostic data were available to evaluate an alternative molecular cause.
Details ▾Hide ▴
Reasoning
BP5 (Richards et al. 2015, PMID:25741868) applies when a case has an established alternate molecular basis of disease (e.g., a pathogenic variant in another gene fully explaining the phenotype). No proband-level genetic or diagnostic data are present in the evidence bundle: there is no documented alternative pathogenic variant, no segregation data, and no established alternate diagnosis. No publication reporting this exact variant was retrieved (literature_pass total_pmids=0). Without proband genotype/diagnostic information, the alternate-cause determination cannot be evaluated.
Rule
BP5 (Richards et al. 2015, PMID:25741868): case with an established alternate molecular basis of disease
Evidence considered
Case bundle contains no proband genotype, segregation, or diagnostic fields (case_summary.json compact_evidence; evidence.json)
ClinVar exact-variant search: no_result, no record (evidence.json clinvar)
Not applicable · 8· recorded as not applicableShow reasons ▾Hide reasons ▴PS1N/ANot applicable: as a splice-site variant producing no altered amino acid, no established pathogenic missense change exists to compare.
Details ▾Hide ▴
Reasoning
PS1 (Richards et al. 2015, PMID:25741868) requires the same amino acid change as a previously established pathogenic variant, evaluated regardless of the underlying nucleotide change. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, no altered amino acid exists at this position to compare against any previously established pathogenic missense variant, so PS1's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: canonical_splice (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
PM1N/ANot applicable: as a splice-site variant producing no altered residue, no mutational hotspot or critical-domain membership can be evaluated.
Details ▾Hide ▴
Reasoning
PM1 (Richards et al. 2015, PMID:25741868) applies to a missense variant located in a mutational hot spot and/or a critical, well-established functional domain without benign variation. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, no altered residue exists to evaluate for mutational hot-spot or critical-domain membership, so PM1's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: canonical_splice (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
PM4N/ANot applicable: as a splice-site variant, no protein length change occurs for this criterion to evaluate.
Details ▾Hide ▴
Reasoning
PM4 (Richards et al. 2015, PMID:25741868) applies to protein length changes: in-frame insertions/deletions in a non-repetitive region, or stop-loss variants that extend the protein past the reference stop codon. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, no change in protein length occurs, so there is nothing for this criterion to evaluate, so PM4's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
PM5N/ANot applicable: as a splice-site variant producing no missense change, no pathogenic missense comparator exists at this residue.
Details ▾Hide ▴
Reasoning
PM5 (Richards et al. 2015, PMID:25741868) requires a novel missense change at an amino acid residue where a different missense change has previously been determined to be pathogenic. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, no missense change exists at this residue to compare against a different pathogenic missense change at the same position, so PM5's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: canonical_splice (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
PP2N/ANot applicable: as a splice-site variant producing no missense change, the gene's missense-constraint properties are not relevant.
Details ▾Hide ▴
Reasoning
PP2 (Richards et al. 2015, PMID:25741868) applies to a missense variant in a gene with a low rate of benign missense variation, where missense variants are a common mechanism of disease. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, the gene's missense-constraint properties are irrelevant because no missense change is present to evaluate, so PP2's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: canonical_splice (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
BP1N/ANot applicable: as a splice-site variant producing no missense change, the truncating-variant mechanism criterion does not apply.
Details ▾Hide ▴
Reasoning
BP1 (Richards et al. 2015, PMID:25741868) applies to a missense variant in a gene for which primarily truncating variants are known to cause disease. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, the gene's truncating-variant mechanism is irrelevant because no missense change is present to evaluate, so BP1's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_missense
Evidence considered
Deterministic consequence classification: canonical_splice (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
BP3N/ANot applicable: as a splice-site variant, no in-frame length change in a repetitive region occurs for this criterion to evaluate.
Details ▾Hide ▴
Reasoning
BP3 (Richards et al. 2015, PMID:25741868) applies to in-frame insertions/deletions located in a repetitive region without a known function. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, the variant does not alter protein length within a repeat region, so there is nothing for this criterion to evaluate, so BP3's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
BP7N/ANot applicable: this is not a synonymous variant — it alters the encoded protein sequence.
Details ▾Hide ▴
Reasoning
BP7 (Richards et al. 2015, PMID:25741868) is defined for a synonymous (silent) variant for which splicing prediction algorithms predict no impact to the splice consensus sequence and the nucleotide is not highly conserved. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, the encoded protein sequence is altered, so the criterion's silent-variant premise does not hold regardless of any splicing prediction, so BP7's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_synonymous
Evidence considered
Deterministic consequence classification: canonical_splice (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
Identifiers
Gene
FANCD2
Transcript
NM_033084.4
Coding (HGVS c.)
NM_033084.4:c.3560+2T>A
Protein (HGVS p.)
NP_149075.2:p.?
GRCh38
chr3:10088544 T>A
GRCh37
chr3:10130228 T>A
Gene diagram
Every interpreted variant in FANCD2, mapped onto the exon structure of NM_033084.4. Select a row to find it on the plot.
FANCD2NM_033084.4
Fetching transcript structure from UCSC…
Exons
—
Transcript span
—
Strand
—
Variants mapped
—
Source
UCSC ncbiRefSeqCurated
All variants in FANCD2—Select a row to find it on the plot; use the link to open its page.
Protein
Location
Classification
Link
gnomAD v4.1Allele count data not recorded
gnomAD v2.1Allele count data not recorded
gnomAD-Canada v1.0Allele count data not recorded
Statements
gnomAD v4.1
Absent from gnomAD v4.1.
gnomAD v2.1
Absent from gnomAD v2.1.
gnomAD-Canada
Absent from gnomAD-Canada v1.0.
SpliceAI and Pangolin
SpliceAI acceptor lossat -95 bp
0.610
SpliceAI acceptor gainat +66 bp
0.000
SpliceAI donor lossat -2 bp
0.990
SpliceAI donor gainat +110 bp
0.180
Pangolin splice lossmagnitude shown
-0.850
Pangolin splice gain
0.080
Other predictors
REVEL
Not available
BayesDel
0.660
How the predictions were used
PP3Not met: SpliceAI max delta 0.99 predicts near-certain splice impact, but this same prediction is counted under PVS1 and cannot be double-counted as PP3.
Details ▾Hide ▴
Reasoning
Variant NM_033084.4:c.3560+2T>A is a canonical +2 donor splice-site variant (intron 35; variant_exonic_positions 35i). SpliceAI predicts a near-certain splice impact (max delta = 0.99; DS_DL 0.99, DS_AL 0.61), far above the tool's published high-recall (0.2) and high-precision (0.5) thresholds (Jaganathan et al., Cell 2019, PMID 30661751). However, this same splice-effect prediction is the evidence evaluated under the ClinGen SVI PVS1 framework for canonical +1/+2 splice variants (PMC6185798, PMID 30192042), and stacking PP3 on the identical prediction would double-count it; the case's PVS1 scaffold likewise directs 'do not stack PP3 for the same splice-effect prediction evidence.' No independent, calibrated computational line exists to support PP3 on its own merits: REVEL returned no score for this variant (not found in local REVEL v1.3), and the BayesDel noAF score (0.66) is treated as insufficiently calibrated for PP3/BP4 (no verified published threshold with named publication/table is available to this pipeline) and in any case is a missense predictor applied to an intronic splice variant, which is not a valid application. The ClinGen SVI PP3/BP4 calibration paper (Pejaver et al. 2022, PMID 36413997) calibrated missense predictors and did not provide a PP3/BP4 calibration for splice predictors. PP3 is therefore not met; the computational splice evidence is owned by the PVS1 evaluation and must not be double-counted here.
Rule
Generic ACMG/AMP 2015 PP3 (Richards et al., PMID 25741868): 'Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.).' Caveats applied: algorithms sharing inputs are not independent lines; computational splice-impact evidence already counted under PVS1 for a canonical +1/+2 splice variant must not be double-counted as PP3 (ClinGen SVI PVS1 framework, PMC6185798).
Evidence considered
SpliceAI lookup (source_registry key: spliceai): max delta score 0.99 (DS_DL=0.99, DS_AL=0.61, DS_DG=0.18, DS_AG=0.0) for NM_033084.4:c.3560+2T>A; exceeds the SpliceAI publication's high-recall (0.2) and high-precision (0.5) thresholds (Jaganathan et al., Cell 2019; PMID 30661751). Score used descriptively only; not applied as PP3 because the same splice-effect prediction is subsumed by the PVS1 evaluation for canonical +1/+2 splice variants and double-counting the same prediction across criteria is prohibited.
BayesDel noAF lookup (source_registry key: bayesdel): score 0.66 for chr3:10130228 T>A. Treated as not_available for PP3/BP4: no verified published threshold with a nameable publication and table is available to this pipeline, and BayesDel is a missense pathogenicity predictor that is not validly applied to an intronic splice variant.
REVEL: no score returned for this variant (prefetch.json revel.found=false; local REVEL v1.3 data lacks the record), so the ClinGen SVI-calibrated REVEL thresholds (Pejaver et al., Am J Hum Genet 2022; PMID 36413997) were not applied.
Generic ACMG/AMP 2015 rules (source_registry key: generic_acmg_combination_rules; Richards et al. 2015; PMID 25741868): PP3 requires multiple lines of computational evidence supporting a deleterious effect, with the caveat that in silico tools sharing inputs are not independent and must not be double-counted with each other or with overlapping functional/splice evidence.
ClinGen SVI PVS1 recommendations (source_registry key: pvs1_generic_framework; PMC6185798; Tayoun et al., Hum Mutat 2018; PMID 30192042): canonical +1/+2 splice variants are evaluated under the generic PVS1 framework, and the same splice-impact prediction must not be additionally counted as PP3.
Still missing
No independent second line of calibrated computational evidence: REVEL score not found for this variant (local REVEL v1.3 lookup returned no record).
BayesDel noAF score 0.66 cannot be used because no verified published PP3/BP4 threshold (named publication and table) is available to this pipeline.
SpliceAI has no ClinGen SVI-calibrated PP3/BP4 threshold (Pejaver et al. 2022, PMID 36413997 calibrated missense predictors only); its use here is descriptive and subsumed by the PVS1 splice-site evaluation.
BP4Not met: SpliceAI max delta 0.99 predicts near-certain splice impact, the opposite of the no-impact evidence BP4 requires.
Details ▾Hide ▴
Reasoning
BP4 requires multiple lines of computational evidence suggesting no impact on the gene or gene product. The only splice-relevant computational evidence available points in the opposite direction: SpliceAI predicts near-certain donor-loss splicing impact (max delta = 0.99) at this canonical +2 donor position. No benign-supporting computational line exists: REVEL returned no score; the BayesDel noAF score (0.66) is not usable (no verified published PP3/BP4 threshold available to this pipeline; and as a missense predictor it is not validly applied to an intronic splice variant), and even directionally it is not benign-supporting. BP4 is therefore not met.
Rule
Generic ACMG/AMP 2015 BP4 (Richards et al., PMID 25741868): 'Multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.).' Caveat applied: algorithms sharing inputs are not independent lines; no computational line here suggests no impact.
Evidence considered
SpliceAI lookup (source_registry key: spliceai): max delta score 0.99 (DS_DL=0.99) indicates donor-loss splice impact at c.3560+2T>A — evidence AGAINST no-impact, so it cannot support BP4.
BayesDel noAF lookup (source_registry key: bayesdel): score 0.66; treated as not_available for PP3/BP4 (no verified published threshold with nameable publication/table) and not validly applicable to an intronic splice variant; directional value is not benign-supporting in any case.
REVEL: no score returned for this variant (prefetch.json revel.found=false), so no benign-threshold (BP4) evaluation was possible.
Generic ACMG/AMP 2015 rules (source_registry key: generic_acmg_combination_rules; Richards et al. 2015; PMID 25741868): BP4 requires multiple lines of computational evidence suggesting no impact; absent any such line, BP4 is not met.
Still missing
No calibrated benign-supporting computational evidence was available: REVEL score not found; BayesDel unverifiable/uncalibrated for PP3/BP4; no SVI-calibrated splice-predictor BP4 threshold exists (Pejaver et al. 2022, PMID 36413997).
No population-frequency or functional benign evidence falls within this group's scope; BP4 rests solely on in silico predictions, and none support no impact.
BP7N/ANot applicable: this is not a synonymous variant — it alters the encoded protein sequence.
Details ▾Hide ▴
Reasoning
BP7 (Richards et al. 2015, PMID:25741868) is defined for a synonymous (silent) variant for which splicing prediction algorithms predict no impact to the splice consensus sequence and the nucleotide is not highly conserved. NM_033084.4:c.3560+2T>A in FANCD2 is a variant at a canonical ±1/2 splice-consensus position predicted to produce NP_149075.2:p.?. As a result, the encoded protein sequence is altered, so the criterion's silent-variant premise does not hold regardless of any splicing prediction, so BP7's structural prerequisite is not met by this variant class. This determination was made deterministically from the variant's consequence class, prior to LLM adjudication, because no structured VCEP criteria-combination framework governs this case (framework_complete=false); the generic ACMG/AMP 2015 fallback applies.
Rule
deterministic_prefilter: requires_synonymous
Evidence considered
Deterministic consequence classification: canonical_splice (prefetch.json normalized protein consequence)
Sources
generic_acmg_combination_rules
Statement
In silico
SpliceAI predicts possible splice impact for this variant (max delta score = 0.99). BayesDel score = 0.66.
0 paper records listed. No publication is explicitly linked to an applied criterion in the recorded metadata.
Screenshots represent source records captured during evidence collection. Select one to view the recorded source content.
This variant was interpreted by pipeline 7.1.0.
The current version is 11.0.07.
Re-running queues a fresh interpretation; the result replaces this page when complete.
Re-runs are not free. Blank draws on today's public interpretation allowance; with a code, it uses one of that code's uses.