Why use it

The universal gold standard for variant interpretation in diagnostic and research molecular genetics. Ensures consistent, reproducible classification across laboratories and clinicians. Required by CAP, CLIA, and ISO 15189 accreditation. Accepted by insurance payers and clinical actionability frameworks. The CMV sub-classification helps laboratories and clinicians prioritize VUS variants for re-analysis and family follow-up.

About

The ACMG-AMP 2015 guidelines provide the global standard framework for classifying sequence variants (SNVs and small indels) into five tiers: Pathogenic (P), Likely Pathogenic (LP), Variant of Uncertain Significance (VUS), Likely Benign (LB), or Benign (B). The classification combines evidence from four domains: (1) Population Data — allele frequency in control databases (gnomAD/ExAC/1000G); (2) Computational/Predictive Data — in silico tools and conservation; (3) Functional Data — biochemical or cell-based assays; (4) Segregation/Observational Data — co-segregation in families and co-occurrence with other variants. Each evidence item is weighted by strength: Very Strong (PVS), Strong (PS/BS), Moderate (PM), Supporting (PP/BP), or Stand-alone (BA). This enhanced version includes Clinically Meaningful Variant (CMV) sub-classification of VUS into VUS-Favor Pathogenic (Hot VUS), VUS (Indeterminate), and VUS-Favor Benign (Cold VUS), per ClinGen SVI recommendations and Bayesian point-based refinements (Tavtigian et al. 2018). The CMV layer provides a finer-grained triage of VUS variants for clinical follow-up prioritization.

When to use

Evaluating a novel or rare variant identified on clinical sequencing (WES, WGS, gene panel)
Re-evaluating a historic VUS in light of new functional data, population data, or segregation evidence
Classifying variants during family cascade testing after identification of a likely pathogenic variant
Determining actionability of secondary/incidental findings per ACMG SF v3.2 list
Laboratory classification meetings and multidisciplinary variant review
Prioritizing VUS variants for re-contact or periodic re-review using CMV sub-classification

Formula

PATHOGENIC requires (any one of): ≥1 PVS + ≥1 PS or ≥2 PM or (1 PM + ≥1 PP) or ≥2 PP ≥2 PS 1 PS + (≥3 PM or [2 PM + ≥2 PP] or [1 PM + ≥4 PP]) LIKELY PATHOGENIC requires (any one of): 1 PVS + 1 PM 1 PS + (1–2 PM) 1 PS + ≥2 PP ≥3 PM 2 PM + ≥2 PP 1 PM + ≥4 PP BENIGN requires: ≥1 BA, OR ≥2 BS LIKELY BENIGN requires: 1 BS + 1 BP, OR ≥2 BP CONFLICT → VUS (default when evidence is insufficient or contradictory) BAYESIAN POINT SYSTEM (Tavtigian et al. 2018): PVS1 = 8 pts, PS = 4 pts each, PM = 2 pts each, PP = 1 pt each BA1 = stand-alone benign, BS = -4 pts each, BP = -1 pt each Pathogenic ≥ 10 | Likely Path 6–9 | VUS-FP 4–5 | VUS 1–3 | VUS-FB 0 | Likely Benign -1 to -6 | Benign ≤ -7

Pearls & pitfalls

BA1 (allele frequency > 5% in gnomAD) is a stand-alone rule that OVERRIDES nearly all pathogenic criteria for rare Mendelian diseases. Always check gnomAD first.
PVS1 (null variant) requires that Loss-of-Function is the established disease mechanism for that gene. Do not apply to dominant-negative genes (e.g., COL1A1 structural domains).
PM2 (absent/ultra-rare in gnomAD) is one of the most commonly applied criteria — but gnomAD v4 has population-specific subsets; use appropriate ancestry-matched population.
A VUS is not a negative result — it means "insufficient or conflicting evidence." It is the DEFAULT classification when criteria do not clearly tip toward P/LP or B/LB.
Conflicting evidence (e.g., one P criterion AND one B criterion met) automatically results in VUS regardless of additional evidence on either side.
PP5/BP6 (reports from databases like ClinVar without underlying evidence) carry reduced weight per updated SVI recommendations (2020 ClinGen SVI guidance). Some expert panels now recommend NOT applying PP5/BP6 at all.
Revised ClinGen SVI framework (2019–2024) provides refined rules for many criteria — this tool implements the original 2015 framework with Bayesian point-based extensions. It applies every criterion at its ORIGINAL 2015 strength. When a ClinGen VCEP specification modulates a strength (PM2 at supporting, PVS1_Strong, a calibrated PP3), the 2015 combining rules cannot express it — use the companion "ClinGen SVI / VCEP Points" calculator instead.
BA1 is stand-alone and is treated as such here: if BA1 is applied the result is Benign even when pathogenic criteria are also selected. That combination is flagged, because it usually means the BA1 frequency threshold chosen was not the right one for the condition.
PS2/PM6 (de novo) requires careful verification: PS2 requires confirmed maternity AND paternity via molecular methods.
The Bayesian point-based system (Tavtigian et al. 2018) assigns: PVS = 8 pts, PS = 4 pts, PM = 2 pts, PP = 1 pt, BS = -4 pts, BP = -1 pt, BA = stand-alone benign. Total ≥10 = Pathogenic, 6–9 = Likely Pathogenic, 0–5 = VUS, -1 to -6 = Likely Benign, ≤-7 = Benign.
CMV sub-classification: VUS-Favor Pathogenic = Bayesian points 4–5; VUS Indeterminate = points 1–3; VUS-Favor Benign = points 0 or negative (but not meeting LB/B threshold).

Next steps

Pathogenic / Likely Pathogenic: Use for clinical decision-making, cascade family testing, and management per gene-specific guidelines. VUS-Favor Pathogenic (Hot VUS): Consider as a strong candidate for additional investigations — functional studies, segregation analysis, or re-review in 6–12 months. Do NOT use for predictive testing in relatives. VUS (Indeterminate): Do NOT use for clinical decision-making. Do NOT offer predictive testing to relatives. File for periodic re-review. VUS-Favor Benign (Cold VUS): Low priority for re-review; unlikely to be reclassified as pathogenic. Likely Benign / Benign: Do NOT use for clinical decision-making. Consider reporting as "not contributory."

Evidence

Richards S et al. 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. Genet Med. 2015;17(5):405-424. Tavtigian SV et al. Modeling the ACMG/AMP variant classification guidelines as a Bayesian classification framework. Genet Med. 2018;20(9):1054-1060. ClinGen SVI Working Group ongoing refinements (2019–2024). Bayesian point-based system adopted by multiple ClinGen Variant Curation Expert Panels (VCEPs).

References

Richards S et al. Standards and guidelines for the interpretation of sequence variants. Genet Med. 2015;17(5):405-424. — link
Tavtigian SV et al. Modeling the ACMG/AMP variant classification guidelines as a Bayesian classification framework. Genet Med. 2018;20(9):1054-1060. — link
Biesecker LG, Harrison SM. The ACMG/AMP reputable source criteria for the interpretation of sequence variants. Genet Med. 2018;20(12):1687-1688. — link
Pejaver V et al. Calibration of computational tools for missense variant pathogenicity classification and ClinGen recommendations for PP3/BP4 criteria. Am J Hum Genet. 2022;109(12):2163-2177. — link
ClinGen SVI Recommendation for de novo Criteria (PS2/PM6). 2019. — link

ACMG-AMP 2015 Variant Pathogenicity Classifier

Calculator