Scaffold-Lab: Evaluation Results for Seven Protein Backbone Generation Methods

A dataset containing evaluation results from the Scaffold-Lab framework, which systematically assesses protein backbone generation methods. The data was created by Zhuoqi Zheng and last updated on May 22, 2026. It likely contains performance metrics for seven representative methods across categories like designability, novelty, diversity, efficiency, and structural properties.

Use Cases

• Benchmarking new protein design algorithms based on the standardized metrics described.
• Identifying methodological strengths and weaknesses for specific tasks like unconditional generation or motif reconstruction.
• Informing the selection of a backbone generation method for a specific project based on comparative performance data.

Strengths

• Provides a unified evaluation framework (Scaffold-Lab) for systematic comparison, as described.
• Assesses seven representative methods across multiple metrics, including designability and novelty.
• Dataset is openly licensed under CC-BY-4.0.

Limitations

• Dataset is very small at 5.5 KB, indicating limited scope or summary-level data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

figshare, authored by Zhuoqi Zheng.

Collection Method

Likely contains results from the proposed Scaffold-Lab evaluation framework applied to seven methods.

Time Range

null

Freshness

Last updated 2026-05-22 17:46:24; freshness should be verified.

Geography

null