Variant classification summary

NM_133509.4:c.226G>A

RAD51B · NP_598193.2:p.(Ala76Thr) · NM_133509.4

GRCh37: chr14:68301824 G>A · GRCh38: chr14:67835107 G>A

Gene: RAD51B Transcript: NM_133509.4

Final call

Likely Benign

BS1 strong benign BP4 supporting benign BP6 supporting benign

Variant details

Gene

RAD51B

Transcript

NM_133509.4

Protein

NP_598193.2:p.(Ala76Thr)

gnomAD AF

0.0003848685311408842 (v4.1)

ClinVar

Uncertain significance

OncoKB

Unknown Oncogenic Effect

Classification rationale

Interpretation summary

Generated evidence synthesis

1

NM_133509.4:c.226G>A (p.Ala76Thr) is a missense variant in exon 4 of RAD51B that is present in gnomAD v4.1 at a grpmax filtering allele frequency of 0.89%, exceeding the BS1 threshold of 0.3%, with 2 homozygotes observed in the general population.

2

Multiple in silico predictors consistently indicate a benign effect for this amino acid substitution: REVEL score 0.017, BayesDel score -0.558, and SpliceAI max delta 0.03, meeting BP4 at supporting benign strength.

3

Ambry Genetics, a clinical diagnostic laboratory, has classified this variant as Likely benign (ClinVar SCV005488250, criteria provided), meeting BP6 at supporting benign strength.

4

No pathogenic criteria are met. PVS1 is not applicable as the variant is missense. PS1-PS4, PM1-PM2, PM6, PP1-PP5, BS3-BS4, and BP1-BP2 are either not met or not applicable.

Final determination: Generic ACMG/AMP 2015 fallback rules support a Likely Benign classification because either one strong benign criterion plus one supporting benign criterion, or at least two supporting benign criteria, are present.

Criteria assessment

ACMG/AMP criteria review

Criteria shown when status is available

All criteria require review: For research and educational purposes only.

Criterion	Status	Rationale	Evidence used
PVS1	N/A	NM_133509.4:c.226G>A is a missense variant (p.Ala76Thr) and does not fall into any default PVS1 null-variant bucket (nonsense, frameshift, or canonical ±1,2 splice consensus). The generic PVS1 decision tree does not apply to this variant class.	pvs1_variant_assessment pvs1_generic_framework
PS1	Not met	No previously established pathogenic variant resulting in the same amino acid change (p.Ala76Thr) has been identified in ClinVar or the literature. The variant has not been classified as pathogenic by any submitter.	clinvar pm5_candidates
PS2	Not met	No de novo occurrence with confirmed paternity and maternity has been reported for this variant in the available literature or ClinVar submissions.	clinvar
PS3	Not met	No well-established functional studies demonstrating a damaging effect have been identified for this variant. In silico predictors (REVEL 0.017, BayesDel -0.558) consistently suggest a benign effect. OncoKB lists the variant as 'Unknown Oncogenic Effect' with no curated functional evidence.	revel bayesdel oncokb
PS4	Not met	No case-control or cohort data demonstrating statistically significant enrichment of this variant in affected individuals versus controls has been identified. The variant is observed in population databases at frequencies inconsistent with a rare pathogenic variant.	gnomad_v2 gnomad_v4 clinvar
PS5	N/A	PS5 is not a standard criterion in the generic ACMG/AMP 2015 framework (PMID:25741868). No VCEP framework exists for RAD51B to define a custom PS5 rule.	PMID:25741868
PM1	Not met	Residue Ala76 does not lie within a statistically significant mutational hotspot in RAD51B, nor within a well-established critical functional domain defined by a RAD51B-specific VCEP framework. The hotspot analysis did not identify this position as significant.
PM2	Not met	Although the overall allele frequency in gnomAD v2.1 (0.042%) and v4.1 (0.038%) is below the 0.1% PM2 threshold, gnomAD v4.1 reports 2 homozygotes and a grpmax filtering allele frequency of 0.89% (>0.3% BS1 threshold). The presence of homozygotes in a population database for a cancer predisposition gene contradicts the expectation that a pathogenic variant would be absent or extremely rare, and the elevated grpmax FAF exceeds the BS1 cutoff, further undermining PM2 applicability.	gnomad_v2 gnomad_v4
PM5	N/A	No same-residue pathogenic comparator variant at Ala76 was identified in ClinVar. The automated PM5 candidate search found no eligible variants for same-residue comparison.	pm5_candidates
PM6	Not met	No de novo occurrence (without confirmed paternity and maternity) has been reported for this variant in ClinVar submissions or the available literature.	clinvar
PP1	Not met	No co-segregation data with disease in multiple affected family members is available for this variant.
PP2	Not assessed	PP2 assessment requires gene-level constraint metrics (e.g., missense Z-score from gnomAD) to determine whether RAD51B has a low rate of benign missense variation and whether missense variants are a common mechanism of disease. These metrics are not available in the generic ACMG adjudication workflow for RAD51B.
PP3	Not met	Multiple in silico predictors consistently indicate a benign effect: REVEL score 0.017 (well below the 0.5 threshold for deleterious prediction), BayesDel score -0.558 (negative = benign), and SpliceAI max delta 0.03 (no predicted splicing impact). No computational tool supports a deleterious effect.	revel bayesdel spliceai
PP4	Not assessed	No patient phenotype or clinical history data are available for this variant. PP4 requires that the proband's phenotype or family history is highly specific for a disease with a single genetic etiology, which cannot be assessed without clinical data.
PP5	Not met	No reputable source has classified this variant as pathogenic. ClinVar reports one submission as Uncertain significance (GeneKor MSA) and one as Likely benign (Ambry Genetics). Neither submission is from an expert panel, and the only classification from a clinical testing laboratory with criteria provided is Likely benign.	clinvar
BA1	Not met	The overall allele frequency in gnomAD v4.1 is 0.038%, and the highest subpopulation frequency (Middle Eastern) is 1.11%. Neither approaches the BA1 threshold of >5%.	gnomad_v4
BS1	Met	The gnomAD v4.1 grpmax filtering allele frequency is 0.89%, exceeding the 0.3% BS1 threshold for generic non-VCEP ACMG adjudication. The Middle Eastern subpopulation allele frequency is 1.11% with 1 homozygote. In aggregate across v4.1, 2 homozygotes are observed in 1,613,538 alleles (621 total alleles). For a cancer predisposition gene where pathogenic variants are expected to be rare, this population frequency and the presence of homozygotes in ostensibly healthy population controls strongly supports a benign interpretation.	gnomad_v2 gnomad_v4
BS2	N/A	BS2 requires observation in a healthy adult individual for a disorder with full penetrance expected at an early age. RAD51B-associated cancer predisposition has incomplete penetrance and adult onset, and therefore does not meet the strict BS2 criteria for early-onset, fully penetrant disorders. The observed homozygotes in gnomAD are captured under BS1 instead.	PMID:25741868 gnomad_v4
BS3	Not met	No well-established functional studies demonstrating no damaging effect have been identified for this variant. While in silico predictors are consistently benign, these are captured under BP4 and do not independently constitute BS3-level functional evidence.	revel bayesdel
BS4	Not met	No family segregation data demonstrating lack of co-segregation with disease is available for this variant.
BP1	Not met	BP1 applies to missense variants in genes where only truncating variants are known to cause disease. While RAD51B loss-of-function via truncating variants is a reported disease mechanism (as confirmed in the PVS1 gene context), there is insufficient evidence to conclude that missense variants in RAD51B are never pathogenic. RAD51 paralogs (RAD51C, RAD51D) have both truncating and proven pathogenic missense variants, and the generic ACMG framework cannot confidently restrict RAD51B pathogenicity to truncating variants alone.	pvs1_gene_context
BP2	Not met	No observation of this variant in trans with a pathogenic variant for a fully penetrant dominant disorder, or in cis with a pathogenic variant for a recessive disorder, has been reported.
BP4	Met	Multiple lines of computational evidence consistently predict no impact on gene product: REVEL score 0.017 (well below 0.5 deleterious threshold), BayesDel score -0.558 (negative = benign), and SpliceAI max delta score 0.03 (no predicted splicing alteration). All available in silico tools concur in a benign assessment.	revel bayesdel spliceai
BP5	Not met	No observation of this variant in a case where an alternate molecular basis for disease has been identified.
BP6	Met	Ambry Genetics, a reputable clinical diagnostic laboratory, has classified this variant as Likely benign with criteria provided (ClinVar accession SCV005488250). While the review status is 'criteria provided, single submitter' and not an expert panel consensus, the classification from a clinical testing laboratory with established variant interpretation practices supports a benign interpretation.	clinvar
BP7	N/A	NM_133509.4:c.226G>A is a missense variant (p.Ala76Thr) resulting in an amino acid substitution, not a synonymous variant. BP7 applies exclusively to synonymous variants with no predicted splice impact.

Disclaimer: The content and results provided by LYFE Sciences are for research and educational purposes only and must not be used as a substitute for professional medical judgment, diagnosis, or treatment. Always consult a qualified healthcare professional before making any clinical decisions.