Protocol

Scaffold-Lab unconditional generation: Diversity (↑)

Share of distinct structural clusters among the generated backbones. Measured in the Scaffold-Lab unconditional generation pipeline: 100 backbones per length, 10 ProteinMPNN sequences each, refolded with ESMFold.

28 evaluations · 28 results

Overview

An explanatory profile has not yet been reviewed for this record.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

28 recorded evaluations, 28 metric rows. A comparison chart has not yet been validated for these results. The table retains the individual findings and their sources.

View coverage and remaining gaps across all benchmarks

Results

Results are available, but no reviewed comparison panel is linked in this release.

All evaluations

28 evaluations · 28 results. Different protocols are not a single leaderboard.

Protocol: Scaffold-Lab unconditional generation: Diversity (↑) · Dataset: Scaffold-Lab generated backbones (100 per length group)

Sorted by Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 100) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.

Tested configuration	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 100)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 50)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 200)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 300)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 400)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 500)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 1000)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 600)	Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 800)
Genie (Scaffold-Lab unconditional generation)Evaluation	0.83 (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	0.66	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
FrameDiff (Scaffold-Lab unconditional generation)Evaluation	0.46	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
FrameFlow (Scaffold-Lab unconditional generation)Evaluation	0.4	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	0.35	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	0.7	Not reported	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	0.88	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	0.84
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	0.48	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	0.67 (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	0.74	Not reported	Not reported	Not reported	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	0.84	Not reported	Not reported	Not reported
Chroma (Scaffold-Lab unconditional generation)Evaluation	Not reported	0.54	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
FrameDiff (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	0.39	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
FrameDiff (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	0.48	Not reported	Not reported	Not reported	Not reported	Not reported
FrameDiff (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	0.55	Not reported	Not reported	Not reported	Not reported
FrameDiff (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	0.59	Not reported	Not reported	Not reported
FrameDiff (Scaffold-Lab unconditional generation)Evaluation	Not reported	0.49	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
FrameFlow (Scaffold-Lab unconditional generation)Evaluation	Not reported	0.32	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Genie (Scaffold-Lab unconditional generation)Evaluation	Not reported	0.59 (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	1 (best in column)	Not reported	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	0.891 (best in column)	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	1 (best in column)
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	0.63 (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	0.67 (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	0.76 (best in column)	Not reported	Not reported	Not reported	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	Not reported	Not reported	Not reported	Not reported	0.87 (best in column)	Not reported	Not reported	Not reported
RFdiffusion (Scaffold-Lab unconditional generation)Evaluation	Not reported	0.12	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported

All 28 result rows with coverage, uncertainty and sources

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.7 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 1000) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 1000 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Long' / 'Chroma' / length 1000, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.88 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 600) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 600 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Long' / 'Chroma' / length 600, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.84 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 800) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 800 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Long' / 'Chroma' / length 800, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.48 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 200) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 200 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'Chroma' / length 200, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.67 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 300) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 300 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'Chroma' / length 300, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.74 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 400) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 400 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'Chroma' / length 400, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.84 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 500) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 500 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'Chroma' / length 500, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.66 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 100) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 100 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'Chroma' / length 100, column 'Diversity (↑)'
Configuration: Chroma (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.54 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 50) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Chroma Diversity (↑) on Scaffold-Lab unconditional generation, length 50 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'Chroma' / length 50, column 'Diversity (↑)'
Configuration: FrameDiff (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.39 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 200) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameDiff Diversity (↑) on Scaffold-Lab unconditional generation, length 200 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'FrameDiff' / length 200, column 'Diversity (↑)'
Configuration: FrameDiff (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.48 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 300) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameDiff Diversity (↑) on Scaffold-Lab unconditional generation, length 300 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'FrameDiff' / length 300, column 'Diversity (↑)'
Configuration: FrameDiff (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.55 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 400) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameDiff Diversity (↑) on Scaffold-Lab unconditional generation, length 400 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'FrameDiff' / length 400, column 'Diversity (↑)'
Configuration: FrameDiff (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.59 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 500) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameDiff Diversity (↑) on Scaffold-Lab unconditional generation, length 500 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'FrameDiff' / length 500, column 'Diversity (↑)'
Configuration: FrameDiff (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.46 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 100) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameDiff Diversity (↑) on Scaffold-Lab unconditional generation, length 100 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'FrameDiff' / length 100, column 'Diversity (↑)'
Configuration: FrameDiff (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.49 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 50) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameDiff Diversity (↑) on Scaffold-Lab unconditional generation, length 50 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'FrameDiff' / length 50, column 'Diversity (↑)'
Configuration: FrameFlow (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.4 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 100) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameFlow Diversity (↑) on Scaffold-Lab unconditional generation, length 100 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'FrameFlow' / length 100, column 'Diversity (↑)'
Configuration: FrameFlow (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.32 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 50) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source FrameFlow Diversity (↑) on Scaffold-Lab unconditional generation, length 50 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'FrameFlow' / length 50, column 'Diversity (↑)'
Configuration: Genie (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.83 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 100) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Genie Diversity (↑) on Scaffold-Lab unconditional generation, length 100 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'Genie' / length 100, column 'Diversity (↑)'
Configuration: Genie (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.59 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 50) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source Genie Diversity (↑) on Scaffold-Lab unconditional generation, length 50 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Short' / 'Genie' / length 50, column 'Diversity (↑)'
Configuration: RFdiffusion (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	1 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 1000) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source RFdiffusion Diversity (↑) on Scaffold-Lab unconditional generation, length 1000 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Long' / 'RFdiffusion' / length 1000, column 'Diversity (↑)'
Configuration: RFdiffusion (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.891 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 600) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source RFdiffusion Diversity (↑) on Scaffold-Lab unconditional generation, length 600 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Long' / 'RFdiffusion' / length 600, column 'Diversity (↑)'
Configuration: RFdiffusion (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	1 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 800) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source RFdiffusion Diversity (↑) on Scaffold-Lab unconditional generation, length 800 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Long' / 'RFdiffusion' / length 800, column 'Diversity (↑)'
Configuration: RFdiffusion (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.63 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 200) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source RFdiffusion Diversity (↑) on Scaffold-Lab unconditional generation, length 200 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'RFdiffusion' / length 200, column 'Diversity (↑)'
Configuration: RFdiffusion (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.67 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 300) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source RFdiffusion Diversity (↑) on Scaffold-Lab unconditional generation, length 300 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'RFdiffusion' / length 300, column 'Diversity (↑)'
Configuration: RFdiffusion (Scaffold-Lab unconditional generation)	Protocol: Scaffold-Lab unconditional generation: Diversity (↑) Dataset: Scaffold-Lab generated backbones (100 per length group)	0.76 Proportion (unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing; length 400) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Table 3 Designability column: equation (1) defines a per-backbone success over 100 backbones per length, which gives two-decimal fractions, but the column prints three-decimal group means; the aggregation behind the column is not stated, and the criterion is sc-TM above 0.5, not the 2 Angstrom sc-RMSD in the store's designability concept. Do not compare these values with other designability results.Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, column 'Designability(↑)'; Metrics, equation (1) Methods, coverage and source RFdiffusion Diversity (↑) on Scaffold-Lab unconditional generation, length 400 lit-structure-20261011-scaffoldlab-protocol-diversity Aggregation: Not reported Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Table 3, row 'Medium' / 'RFdiffusion' / length 400, column 'Diversity (↑)'

Source checking is not independent reproduction. Release 2026-10-23-5347695500ed.

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Explore all linked results

Baseline coverage

Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.

0 of 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.

No execution recipe linked to this protocol. Recipe availability does not establish a completed evaluation.

External evaluations: 28

Literature evidence is not a Rewire measurement. Executed but unpublished runs and private review status are not included.

Null control

Proposed control: requires review

Select a task-valid null control after reviewing inputs and metric

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Conventional reference

Proposed control: requires review

Select an upstream conventional reference after reviewing the full protocol

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Protocol coverage CSV (gzip) · Model evaluation matrix (gzip) · Source table (gzip) · Release and checksums (gzip)

Coverage is derived from release 2026-10-23-5347695500ed. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run instructions

No runnable recipe has been reviewed for this protocol. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-23-5347695500ed
Property and statement	Original source and location	Review and provenance

No evidence rows match these filters. Choose another scope or clear the search.

How to read the evidence

Sources and history

Release 2026-10-23-5347695500ed · Record review: source checked

1 source record and release history

Scaffold-Lab: Critical evaluation and ranking of protein backbone generation methods in a unified framework · Original source · Version of record, PLoS Computational Biology 2026 (PMC13225634 JATS full text)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: lit-structure-20261011-scaffoldlab-protocol-diversity

areas: protein-structure
contexts: research
protocol: Per method and length, 100 backbones are generated, 10 sequences are designed for each with ProteinMPNN and refolded with ESMFold; the metrics are group means without data preprocessing. Ranking uses TOPSIS with CRITIC weighting.
metric: proportion
metric direction: higher
unit: fraction
metric qualifier: unique Foldseek-Cluster clusters divided by the number of backbones in the group, clustering TM-score threshold 0.5; group mean without data preprocessing
version: As published 2026
source locator: Table 3, column 'Diversity (↑)'; Metrics section
aggregation: Mean within each length group without data preprocessing
limitations: Group means only; the per-backbone values are not published in the article.
unit detail: unique Foldseek clusters per generated backbone

Related records

Suggest a correction