---
language:
  - en
license: mit
pretty_name: Variant–Evidence Integrity v0.1
dataset_name: variant-evidence-integrity-v0.1
tags:
  - clarusc64
  - genomics
  - variant-interpretation
  - evidence
  - causality
  - integrity
task_categories:
  - text-classification
size_categories:
  - 1K<n<10K
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train.csv
      - split: test
        path: data/test.csv
---

What this dataset tests  

Variant × evidence × claim.

Whether a genetic variant claim is grounded in real evidence.

Common error modes  

- pathogenic label without functional evidence  
- GWAS association treated as causality  
- in silico prediction treated as proof  

Required outputs  

- evidence binding map  
- evidence gaps  
- claim overreach flags  
- confidence calibration  

Suggested prompt wrapper  

System  

You bind variant claims to evidence.
You calibrate confidence to evidence strength.

User  

Variant  
{variant}

Evidence pack  
{evidence_pack}

Claim  
{claim}

Return  

- evidence binding map  
- evidence gaps  
- claim overreach flags  
- confidence calibration  

Citation  

ClarusC64 dataset family