ACMG Criteria Calculator
Estimate a variant classification using standard ACMG/AMP evidence combination logic. Enter the number of pathogenic and benign criteria currently met, then calculate the predicted classification outcome.
Evidence Inputs
Use whole numbers only. This calculator summarizes evidence strength categories, not individual code names. For example, PS1, PS2, PS3, and PS4 all belong to the strong pathogenic tier.
Classification Result
Important: This page is a structured ACMG criteria calculator for educational and workflow support. Final variant interpretation should incorporate gene and disease specific specifications, laboratory SOPs, and expert review.
Expert Guide to Using an ACMG Criteria Calculator
An ACMG criteria calculator is a practical decision support tool used in clinical genetics, molecular pathology, and genomic curation workflows to translate evidence into a standardized variant classification. In most laboratories, the phrase refers to a calculator based on the American College of Medical Genetics and Genomics and the Association for Molecular Pathology framework, commonly shortened to ACMG/AMP. The goal is simple: organize available evidence, grade its strength, and convert that evidence into one of five familiar categories: pathogenic, likely pathogenic, variant of uncertain significance, likely benign, or benign.
Although the framework is conceptually straightforward, real-world interpretation can become complex very quickly. Population frequency, disease mechanism, segregation data, functional assay quality, transcript relevance, phenotype match, and case-level evidence all matter. A well-designed ACMG criteria calculator does not replace expert judgment. Instead, it makes the combination rules transparent and reproducible. That is especially useful for quality assurance, peer review, case conferences, and educational training.
What the calculator actually measures
The calculator above asks for counts of evidence strengths rather than specific code names. That makes it easier to understand the underlying logic. On the pathogenic side, the ACMG/AMP framework contains one very strong category, four strong criteria, six moderate criteria, and five supporting criteria. On the benign side, there is one stand-alone benign criterion, four strong benign criteria, and seven supporting benign criteria. These counts are not arbitrary. They reflect the original ACMG/AMP architecture for weighing different types of genomic evidence.
| Evidence side | Tier | Number of criteria in the ACMG/AMP framework | Typical interpretation meaning |
|---|---|---|---|
| Pathogenic | Very Strong | 1 | Highest pathogenic weight, classically PVS1-level loss-of-function evidence when disease mechanism is established. |
| Pathogenic | Strong | 4 | Robust case, functional, or de novo style evidence with substantial support. |
| Pathogenic | Moderate | 6 | Meaningful but not decisive evidence such as domain localization, rarity, or trans observations. |
| Pathogenic | Supporting | 5 | Helpful lower-weight evidence including segregation and phenotype consistency. |
| Benign | Stand-alone | 1 | High enough benign evidence to classify a variant as benign by itself. |
| Benign | Strong | 4 | Powerful evidence against pathogenicity such as compelling population or functional findings. |
| Benign | Supporting | 7 | Weaker evidence against pathogenicity, useful when combined with stronger benign observations. |
In total, the classic ACMG/AMP framework includes 28 evidence criteria, split into 16 pathogenic-side and 12 benign-side rules. This is one reason calculators are so helpful. The evidence landscape is broad enough that manual combination can be error-prone, especially when multiple analysts or rotating staff are involved.
How ACMG combination logic works
The calculator uses the standard idea that stronger evidence can compensate for smaller amounts of additional support. For example, one very strong pathogenic criterion plus one strong criterion is enough to reach a pathogenic result. A different path to pathogenicity is two strong criteria without any very strong evidence. Likely pathogenic requires slightly less evidence, such as one very strong plus one moderate criterion, or three moderate criteria. On the benign side, one stand-alone benign criterion or two strong benign criteria support a benign classification, while lesser combinations can support likely benign.
Where things become clinically important is in the gray zone. A variant can have some pathogenic evidence and some benign evidence at the same time. This is common in modern curation. You might have a rare missense variant in a suspicious domain, but also population frequency or functional data that argues against disease relevance. In those situations, calculators often return a conflict-aware outcome that points the analyst back to detailed review. That is usually safer than overcalling a result.
| Classification | Example qualifying combinations | How to think about it in practice |
|---|---|---|
| Pathogenic | 1 Very Strong + 1 Strong; 2 Strong; 1 Strong + 3 Moderate | Evidence is sufficiently convincing that clinical actionability may be high, assuming disease context fits. |
| Likely Pathogenic | 1 Very Strong + 1 Moderate; 1 Strong + 1 to 2 Moderate; 3 Moderate | Evidence favors disease causality, but residual uncertainty remains. |
| VUS | Insufficient evidence, mixed evidence, or conflicting benign and pathogenic signals | Do not over-interpret. Reassessment with new data is often appropriate. |
| Likely Benign | 1 Strong Benign + 1 Supporting Benign; 2 Supporting Benign | Evidence leans away from pathogenicity, but not enough for the full benign label. |
| Benign | 1 Stand-alone Benign; 2 Strong Benign | Evidence strongly excludes clinical pathogenicity in the intended disease context. |
Why laboratories and analysts use an ACMG criteria calculator
There are four major advantages. First, calculators improve consistency. When multiple scientists interpret the same evidence set, standardized combination rules reduce variability. Second, calculators improve speed. Analysts can focus on the quality of evidence instead of repeatedly checking threshold tables. Third, calculators support documentation. A transparent output that lists the triggered rule set is easier to defend in audits and multidisciplinary review. Fourth, calculators support education. Trainees learn more quickly when they can see how moving one criterion from supporting to moderate changes the result.
In many genomic laboratories, the classification process is no longer just a one-time event. Variants may be reinterpreted as new population datasets, functional assays, and clinical correlations become available. An ACMG criteria calculator becomes even more useful in that setting because it lets teams compare the previous and updated evidence structures in a reproducible way.
Best practices before entering evidence into the calculator
- Confirm the transcript and gene-disease relationship. Criteria should be applied only in the correct biological context. A truncating variant does not automatically satisfy a very strong criterion unless loss of function is an established disease mechanism for that gene.
- Check for gene-specific rule modifications. Many expert groups refine the original ACMG/AMP framework. Those specifications can change evidence strength levels and should take priority over generic counting.
- Validate population frequency thresholds. Benign evidence can depend heavily on inheritance model, disease prevalence, penetrance, and ancestry-specific frequency.
- Review the quality of functional evidence. Not all assays are equally informative. Well-validated assays are more persuasive than exploratory experiments.
- Keep case-level evidence independent. Avoid double counting the same observation under multiple criteria.
Common pitfalls that can distort calculator results
- Double counting evidence. The same case series, segregation pattern, or functional result should not be reused under multiple criteria unless guidance clearly allows it.
- Ignoring disease mechanism. A nonsense or frameshift variant is not automatically strong evidence if the gene is not loss-of-function sensitive for the relevant phenotype.
- Overusing computational evidence. In silico predictions can be helpful, but by themselves they are rarely decisive.
- Applying population rules without ancestry context. Population frequency interpretation should consider subpopulation data and coverage quality.
- Failing to flag conflict. When benign and pathogenic evidence coexist, an automatic result without human review can be misleading.
Practical takeaway: The best ACMG criteria calculator is not merely a scoring box. It is a structured checkpoint that forces clear evidence assignment, highlights combinations, and makes uncertainty visible. That last feature is especially valuable because many clinically encountered variants belong in the uncertain category until stronger data arrive.
How to interpret each output category
Pathogenic means the evidence strongly supports disease causality in the relevant gene-disease context. This is the category most likely to influence diagnosis, cascade testing, and risk management. Likely pathogenic is close, but it still carries residual uncertainty. Different institutions may vary in how they communicate that uncertainty to clinicians and patients. Variant of uncertain significance, often abbreviated VUS, is not a negative result and not a positive result. It means the evidence is currently insufficient or contradictory. Likely benign and benign indicate that the variant is not expected to explain the disease indication being assessed.
One of the most important educational points is that a VUS should generally not drive major medical decisions. A calculator can help reinforce that discipline by showing exactly why the evidence did not cross a threshold. That transparency matters in discussions with referring physicians, molecular tumor boards, inherited disease boards, and internal quality teams.
Where authoritative guidance comes from
If you want to go deeper than a quick calculator result, the most useful references are the primary guideline framework and the federal or academic resources that support variant interpretation. For foundational background, see the National Center for Biotechnology Information resource on the ACMG/AMP interpretation framework at NCBI Bookshelf. For broader genomic medicine context, the National Human Genome Research Institute provides educational material at genome.gov. For public variant data sharing and interpretation records, ClinVar at the National Library of Medicine is essential: ClinVar. Public health context for genomics and testing can also be explored through the CDC Office of Genomics and Precision Public Health.
How this calculator fits into a modern variant interpretation workflow
In a typical workflow, an analyst first reviews the variant, transcript, zygosity, inheritance pattern, disease indication, and gene validity. Next comes evidence gathering: population databases, case reports, segregation data, functional studies, and any internal laboratory observations. After evidence is mapped to ACMG/AMP strengths, the analyst enters the counts into an ACMG criteria calculator. The output then serves as a structured draft classification, which is reviewed against gene-specific specifications, literature nuance, and any conflicting data. Finally, the result is signed out with explanatory language tailored to the test indication.
This means the calculator should be viewed as the middle of the process, not the beginning and not the end. It is strongest when the evidence entering it is high quality and when the result leaving it is still checked by expert humans.
Final thoughts
An ACMG criteria calculator is valuable because it converts a complex framework into something consistent, auditable, and teachable. It helps prevent threshold errors, keeps evidence weighting visible, and supports repeatable variant assessment across teams. At the same time, variant interpretation remains a clinical science problem, not just a counting exercise. Disease mechanism, gene-specific rules, assay quality, and phenotype fit still determine whether the assigned evidence is valid in the first place.
If you use this tool as intended, it can save time and improve consistency while still preserving the central role of professional judgment. That balance is exactly what good genomic interpretation requires.