> For the complete documentation index, see [llms.txt](https://docs.omics.ai/products/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.omics.ai/products/workbench/variant-interpreter/classifying-variants.md).

# Classifying variants

Classification is the interpretive step between reviewing a variant and putting it on the report. You assign each variant a clinical significance on the five-tier ACMG/AMP scale, backed by the specific evidence criteria you applied. Only classified variants are eligible to become report findings.

{% hint style="info" %}
Classifying variants requires the **Classify Variants** capability. Viewers can see classifications but not change them. See [Workspaces & roles](/products/workbench/variant-interpreter/workspaces-and-roles.md).
{% endhint %}

## The five-tier scale

Every classification resolves to one category on the ACMG/AMP pathogenicity scale, shown as a visual gradient in the classification panel:

| Category              | Meaning                            |
| --------------------- | ---------------------------------- |
| **Benign**            | Not disease-causing.               |
| **Likely benign**     | Probably not disease-causing.      |
| **VUS**               | Variant of uncertain significance. |
| **Likely pathogenic** | Probably disease-causing.          |
| **Pathogenic**        | Disease-causing.                   |

## Automated vs. Your classification

A variant can carry two kinds of classification, and the panel distinguishes them:

* **Automated classification** — a starting suggestion computed by the platform from the annotations and criteria it can evaluate on its own.
* **Your classification** — the classification you commit as the interpreter. Your classification is what counts for the report; the automated one is only a starting point you can accept, adjust, or override.

## Applying evidence criteria

You justify a classification by applying **ACMG criteria**. Each applied criterion has:

* a **direction** — whether it argues toward **pathogenic** or **benign**;
* a **strength** — Standalone, Very strong, Strong, Moderate, or Supporting; and
* an **evidence note** — free text recording *why* the criterion applies to this variant.

The set of criteria offered depends on the **variant type** — small variants, structural variants, tandem repeats, and paralogs each have their own criteria appropriate to how that variant type is assessed. Each criterion has a default strength, which you can override for a specific variant when the evidence warrants; overridden criteria are marked so a reviewer can see where you departed from the default.

## Recording a classification

1. Open the variant and go to its **Classification** panel.
2. Review the automated suggestion, if present.
3. Apply the criteria that the evidence supports, setting strength and an evidence note on each.
4. Select the resulting ACMG/AMP category.
5. Add any overall **notes** for the variant.

Your classification is saved against the variant and becomes available to the report.

## From classification to report

The report's candidate findings are derived **live** from your classifications: any variant with an active classification is eligible to be included as a finding. You curate which of those actually appear on the report from the report itself — see [Generating a report](/products/workbench/variant-interpreter/generating-a-report.md). There is no separate "add to report" flag stored on the variant; inclusion is decided on the report.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.omics.ai/products/workbench/variant-interpreter/classifying-variants.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.