> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bunkerbio.com/llms.txt
> Use this file to discover all available pages before exploring further.

# GenoScan-LLM: Whole-Genome Variant Calling

> GenoScan-LLM is a foundation model for whole-genome variant calling. Trained on 200K+ genomes, it flags pathogenic variants at clinical-grade sensitivity.

GenoScan-LLM is an 800M parameter foundation model for variant calling. Given aligned reads or a target region, it identifies SNVs, small indels, and structural variants, then classifies them by pathogenicity with clinical-grade sensitivity.

## At a glance

| Field         | Value                                             |
| ------------- | ------------------------------------------------- |
| Category      | Genomics                                          |
| Parameters    | 800M                                              |
| Version       | v2.0                                              |
| Endpoint      | `POST /v1/models/genoscan-llm/predict`            |
| Training data | 200,000+ genomes across ancestry-balanced cohorts |

## Install

```bash theme={null}
pip install bunker-fold
```

See [Installation](/installation) for full setup, including `uv` and `poetry`.

## When to use

* Call variants from whole-genome or targeted panel data.
* Flag pathogenic variants in a clinical reporting workflow.
* Reprioritize a VCF against updated pathogenicity evidence.

## Example

<CodeGroup>
  ```bash cURL theme={null}
  curl -X POST https://api.bunkerbio.com/v1/models/genoscan-llm/predict \
    -H "Authorization: Bearer $BUNKER_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "region": "chr17:41244000-41277500",
      "gene": "BRCA1",
      "coverage": 42,
      "format": "CRAM",
      "input_url": "s3://my-bucket/sample.cram"
    }'
  ```

  ```python Python theme={null}
  import requests, os

  r = requests.post(
      "https://api.bunkerbio.com/v1/models/genoscan-llm/predict",
      headers={"Authorization": f"Bearer {os.environ['BUNKER_API_KEY']}"},
      json={
          "region": "chr17:41244000-41277500",
          "gene": "BRCA1",
          "coverage": 42,
          "format": "CRAM",
          "input_url": "s3://my-bucket/sample.cram",
      },
  )
  print(r.json())
  ```
</CodeGroup>

```json Response theme={null}
{
  "variants_detected": 3,
  "variants": [
    {
      "id": "rs80357906",
      "hgvs": "c.5266dupC",
      "classification": "pathogenic",
      "acmg_class": 5,
      "confidence": 0.994
    },
    { "id": "rs1799966", "classification": "benign" },
    { "id": "rs16942", "classification": "benign" }
  ]
}
```

## Parameters

<ParamField path="region" type="string" required>
  Genomic region in `chrN:start-end` format.
</ParamField>

<ParamField path="input_url" type="string" required>
  Signed URL or S3 path to a BAM, CRAM, or VCF file.
</ParamField>

<ParamField path="format" type="string" required>
  One of `BAM`, `CRAM`, or `VCF`.
</ParamField>

<ParamField path="gene" type="string">
  Optional gene symbol used to prioritize functional annotation.
</ParamField>

<ParamField path="coverage" type="number">
  Average coverage depth. Used to calibrate confidence thresholds.
</ParamField>

## Response fields

<ResponseField name="variants_detected" type="integer">
  Total number of variants called in the region.
</ResponseField>

<ResponseField name="variants" type="array">
  List of variants with identifier, HGVS notation, and pathogenicity classification.

  <Expandable title="variant fields">
    <ResponseField name="id" type="string">
      dbSNP identifier where available.
    </ResponseField>

    <ResponseField name="hgvs" type="string">
      HGVS coding notation.
    </ResponseField>

    <ResponseField name="classification" type="string">
      One of `pathogenic`, `likely_pathogenic`, `uncertain`, `likely_benign`, `benign`.
    </ResponseField>

    <ResponseField name="acmg_class" type="integer">
      ACMG classification from 1 (benign) to 5 (pathogenic).
    </ResponseField>

    <ResponseField name="confidence" type="number">
      Model confidence in the classification, between 0 and 1.
    </ResponseField>
  </Expandable>
</ResponseField>