Protein Function Test Data with GO Annotations and Interactions

Evaluation dataset for the BioReason-Pro model, containing proteins with Gene Ontology term annotations, GO-GPT predictions, InterPro domains, STRING protein-protein interactions, and protein metadata. The dataset was created by wanglab and follows the CAFA framework's temporal holdout methodology. It was last updated in March 2026.

Use Cases

• Benchmarking protein function prediction models using the provided GO term annotations as ground truth labels.
• Training multimodal models that integrate protein metadata, InterPro domains, and STRING protein-protein interaction data.
• Evaluating the accuracy of GO-GPT predictions against the held-out temporal annotations.
• Analyzing the relationship between protein-protein interaction networks from STRING and specific Gene Ontology functions.

Strengths

• Designed for evaluation following the established CAFA framework for temporal holdout.
• Integrates multiple biological data modalities including annotations, predictions, domains, and interactions.

Limitations

• Specific row counts, column details, and sample size are unknown.
• Dataset scope is limited to test/evaluation data, not a full training corpus.

Provenance

Source

wanglab on Hugging Face.

Collection Method

Constructed as an evaluation dataset for the BioReason-Pro model, following CAFA temporal holdout protocols.

Freshness

Last updated on 2026-03-20.