metadata
language: en
license: mit
pretty_name: AlphaFold Misinterpretation Classifier v2
task_categories:
- text-classification
- text-generation
tags:
- protein
- alphafold
- structural-biology
- uncertainty
- evaluation
size_categories:
- n<1K
AlphaFold Misinterpretation Classifier
Purpose
Detect common ways people overclaim from AlphaFold outputs.
Input fields
- protein_context
- alphafold_signals
- proposed_inference
Required output
Return one JSON object
- misinterpretation
yes or no - error_type
must match allowed list - correction
one sentence
Allowed error_type values
- no_error
- low_confidence_region_overtrust
- interdomain_orientation_overclaim
- disorder_as_structure
- loop_position_overtrust
- complex_negation_from_monomer
- interface_overconfidence
- cofactor_absence_misread
- low_complexity_as_domain
Scoring
- misinterpretation_accuracy
- error_type_accuracy
- correction_similarity
- format_pass_rate
Run scoring
python scorer.py --predictions predictions.jsonl --test_csv data/test.csv