Configuration

Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1

The VHH design campaign configuration: 200 fine-tuned RFdiffusion backbones per target, the best of eight ProteinMPNN sequences per backbone, filtered by fine-tuned RoseTTAFold2 with all interface hotspots provided.

30 evaluations · 36 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.

Evaluations and results

30 evaluations · 36 results. Different protocols are not a single leaderboard.

Sorted by Success rate (AlphaFold3 filter) (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	Protocol and dataset	Success rate (AlphaFold3 filter)	Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0)	Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1)	Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 1)	Count (SPR-confirmed binders; initial screening method SPR)	Count (SPR-confirmed binders; initial screening method YSD)	Hit rate (SPR-confirmed binders out of designs screened; initial screening method SPR)	Hit rate (SPR-confirmed binders out of designs screened; initial screening method YSD)	Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0; value printed in parentheses because it repeats the 96-design row from the same design campaign)	Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1; value printed in parentheses because it repeats the 96-design row from the same design campaign)	Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 1; value printed in parentheses because it repeats the 96-design row from the same design campaign)
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under an AlphaFold3 filter, Bennett VHH design target TcdB	27.5% (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under an AlphaFold3 filter, Bennett VHH design target Influenza Hemagglutinin	5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under an AlphaFold3 filter, Bennett VHH design target Influenza Hemagglutinin	5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under an AlphaFold3 filter, Bennett VHH design target IL-7 Receptor α	5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under an AlphaFold3 filter, Bennett VHH design target SARS-CoV-2 RBD	0%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under an AlphaFold3 filter, Bennett VHH design target RSV-F Site III	0.0*%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design, E. coli expression and single-concentration SPR screen of 96 designs, Bennett VHH design target Influenza Hemagglutinin	Not reported	Not reported	Not reported	Not reported	1	Not reported	1%	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target Influenza Hemagglutinin	Not reported	3.5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target Influenza Hemagglutinin	Not reported	Not reported	7.5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target Influenza Hemagglutinin	Not reported	Not reported	Not reported	11.5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation, Bennett VHH design target Influenza Hemagglutinin	Not reported	Not reported	Not reported	Not reported	Not reported	20 (best in column)	Not reported	0.2% (best in column)	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target Influenza Hemagglutinin	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	(3.5)%	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target Influenza Hemagglutinin	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	(7.5)%	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target Influenza Hemagglutinin	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	(11.5)%
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design, E. coli expression and single-concentration SPR screen of 96 designs, Bennett VHH design target IL-7 Receptor α	Not reported	Not reported	Not reported	Not reported	0	Not reported	0%	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target IL-7 Receptor α	Not reported	12.5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target IL-7 Receptor α	Not reported	Not reported	27%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target IL-7 Receptor α	Not reported	Not reported	Not reported	34.5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation, Bennett VHH design target SARS-CoV-2 RBD	Not reported	Not reported	Not reported	Not reported	Not reported	5	Not reported	0.06%	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target SARS-CoV-2 RBD	Not reported	11%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target SARS-CoV-2 RBD	Not reported	Not reported	23.5%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target SARS-CoV-2 RBD	Not reported	Not reported	Not reported	28%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation, Bennett VHH design target RSV-F Site III	Not reported	Not reported	Not reported	Not reported	Not reported	1	Not reported	0.01%	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target RSV-F Site III	Not reported	1%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target RSV-F Site III	Not reported	Not reported	3%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target RSV-F Site III	Not reported	Not reported	Not reported	25%	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design, E. coli expression and single-concentration SPR screen of 96 designs, Bennett VHH design target TcdB	Not reported	Not reported	Not reported	Not reported	2 (best in column)	Not reported	2% (best in column)	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target TcdB	Not reported	27.5% (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target TcdB	Not reported	Not reported	33.5% (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter, Bennett VHH design target TcdB	Not reported	Not reported	Not reported	41% (best in column)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported

All 36 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: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, E. coli expression and single-concentration SPR screen of 96 designs Dataset: Bennett VHH design target Influenza Hemagglutinin	1 Count (SPR-confirmed binders; initial screening method SPR) count · higher Uncertainty: Not applicable: An exact count. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against H1 stem, E. coli expression and single-concentration SPR uc-binder-20261022-protocol-bennett-vhh-ecoli-spr-96 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 96), column 'SPR confirmed binders'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, E. coli expression and single-concentration SPR screen of 96 designs Dataset: Bennett VHH design target Influenza Hemagglutinin	1% Hit rate (SPR-confirmed binders out of designs screened; initial screening method SPR) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against H1 stem, E. coli expression and single-concentration SPR uc-binder-20261022-protocol-bennett-vhh-ecoli-spr-96 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 96), column 'Success rate by SPR'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under an AlphaFold3 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	5% Success rate (AlphaFold3 filter) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, AlphaFold3 filter uc-binder-20261022-protocol-bennett-insilico-af3-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 96), column 'Success rate (%) (AF3)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	3.5% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 96), column 'Success rate (%) (RF2, hotspot proportion = 0.0)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	7.5% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 96), column 'Success rate (%) (RF2, hotspot proportion = 0.1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	11.5% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 1) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, fine-tuned RoseTTAFold2 filter, hotspot proportion 1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 96), column 'Success rate (%) (RF2, hotspot proportion = 1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation Dataset: Bennett VHH design target Influenza Hemagglutinin	20 Count (SPR-confirmed binders; initial screening method YSD) count · higher Uncertainty: Not applicable: An exact count. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against H1 stem, yeast surface display with SPR confirmation uc-binder-20261022-protocol-bennett-vhh-yeast-display-9000 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 9000), column 'SPR confirmed binders'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation Dataset: Bennett VHH design target Influenza Hemagglutinin	0.2% Hit rate (SPR-confirmed binders out of designs screened; initial screening method YSD) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against H1 stem, yeast surface display with SPR confirmation uc-binder-20261022-protocol-bennett-vhh-yeast-display-9000 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 9000), column 'Success rate by SPR'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under an AlphaFold3 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	5% Success rate (AlphaFold3 filter) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, AlphaFold3 filter uc-binder-20261022-protocol-bennett-insilico-af3-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 9000), column 'Success rate (%) (AF3)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	(3.5)% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0; value printed in parentheses because it repeats the 96-design row from the same design campaign) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 9000), column 'Success rate (%) (RF2, hotspot proportion = 0.0)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	(7.5)% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1; value printed in parentheses because it repeats the 96-design row from the same design campaign) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 9000), column 'Success rate (%) (RF2, hotspot proportion = 0.1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target Influenza Hemagglutinin	(11.5)% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 1; value printed in parentheses because it repeats the 96-design row from the same design campaign) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against H1 stem, fine-tuned RoseTTAFold2 filter, hotspot proportion 1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'H1 stem' (Designs screened = 9000), column 'Success rate (%) (RF2, hotspot proportion = 1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, E. coli expression and single-concentration SPR screen of 96 designs Dataset: Bennett VHH design target IL-7 Receptor α	0 Count (SPR-confirmed binders; initial screening method SPR) count · higher Uncertainty: Not applicable: An exact count. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against IL7R, E. coli expression and single-concentration SPR uc-binder-20261022-protocol-bennett-vhh-ecoli-spr-96 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'IL7R' (Designs screened = 96), column 'SPR confirmed binders'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, E. coli expression and single-concentration SPR screen of 96 designs Dataset: Bennett VHH design target IL-7 Receptor α	0% Hit rate (SPR-confirmed binders out of designs screened; initial screening method SPR) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against IL7R, E. coli expression and single-concentration SPR uc-binder-20261022-protocol-bennett-vhh-ecoli-spr-96 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'IL7R' (Designs screened = 96), column 'Success rate by SPR'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under an AlphaFold3 filter Dataset: Bennett VHH design target IL-7 Receptor α	5% Success rate (AlphaFold3 filter) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against IL7R, AlphaFold3 filter uc-binder-20261022-protocol-bennett-insilico-af3-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'IL7R' (Designs screened = 96), column 'Success rate (%) (AF3)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target IL-7 Receptor α	12.5% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against IL7R, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'IL7R' (Designs screened = 96), column 'Success rate (%) (RF2, hotspot proportion = 0.0)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target IL-7 Receptor α	27% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against IL7R, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'IL7R' (Designs screened = 96), column 'Success rate (%) (RF2, hotspot proportion = 0.1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target IL-7 Receptor α	34.5% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 1) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against IL7R, fine-tuned RoseTTAFold2 filter, hotspot proportion 1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'IL7R' (Designs screened = 96), column 'Success rate (%) (RF2, hotspot proportion = 1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation Dataset: Bennett VHH design target SARS-CoV-2 RBD	5 Count (SPR-confirmed binders; initial screening method YSD) count · higher Uncertainty: Not applicable: An exact count. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against SARS-CoV-2 RBD, yeast surface display with SPR confirmation uc-binder-20261022-protocol-bennett-vhh-yeast-display-9000 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'SARS-CoV-2 RBD' (Designs screened = 9000), column 'SPR confirmed binders'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation Dataset: Bennett VHH design target SARS-CoV-2 RBD	0.06% Hit rate (SPR-confirmed binders out of designs screened; initial screening method YSD) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against SARS-CoV-2 RBD, yeast surface display with SPR confirmation uc-binder-20261022-protocol-bennett-vhh-yeast-display-9000 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'SARS-CoV-2 RBD' (Designs screened = 9000), column 'Success rate by SPR'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under an AlphaFold3 filter Dataset: Bennett VHH design target SARS-CoV-2 RBD	0% Success rate (AlphaFold3 filter) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against SARS-CoV-2 RBD, AlphaFold3 filter uc-binder-20261022-protocol-bennett-insilico-af3-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'SARS-CoV-2 RBD' (Designs screened = 9000), column 'Success rate (%) (AF3)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target SARS-CoV-2 RBD	11% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against SARS-CoV-2 RBD, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.0 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'SARS-CoV-2 RBD' (Designs screened = 9000), column 'Success rate (%) (RF2, hotspot proportion = 0.0)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target SARS-CoV-2 RBD	23.5% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against SARS-CoV-2 RBD, fine-tuned RoseTTAFold2 filter, hotspot proportion 0.1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'SARS-CoV-2 RBD' (Designs screened = 9000), column 'Success rate (%) (RF2, hotspot proportion = 0.1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH computational success rate under the fine-tuned RoseTTAFold2 filter Dataset: Bennett VHH design target SARS-CoV-2 RBD	28% Success rate (fine-tuned RoseTTAFold2 filter, hotspot proportion 1) percent · higher Uncertainty: Not reported by the source: No interval is printed. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH computational success rate against SARS-CoV-2 RBD, fine-tuned RoseTTAFold2 filter, hotspot proportion 1 uc-binder-20261022-protocol-bennett-insilico-rf2-success Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'SARS-CoV-2 RBD' (Designs screened = 9000), column 'Success rate (%) (RF2, hotspot proportion = 1)'
Configuration: Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1	Protocol: Bennett VHH design, yeast surface display screen of 9,000 designs with SPR confirmation Dataset: Bennett VHH design target RSV-F Site III	1 Count (SPR-confirmed binders; initial screening method YSD) count · higher Uncertainty: Not applicable: An exact count. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The main text says 'at lower throughput with Escherichia coli expression and single-concentration surface plasmon resonance (SPR; 95 designs per target; TcdB, IL-7Ra and influenza haemagglutinin)', but Supplementary Methods Table 6 prints 96 designs screened for TcdB, IL7R and H1 stem. Table 6's value is recorded because it is the chosen table; the main text figure is 95.Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Results paragraph 2; Supplementary Methods Table 6, column 'Designs screened' Methods, coverage and source Fine-tuned RFdiffusion VHH design against RSV site 3, yeast surface display with SPR confirmation uc-binder-20261022-protocol-bennett-vhh-yeast-display-9000 Aggregation: Not reported Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6; Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Supplementary Methods Table 6, row 'RSV site 3' (Designs screened = 9000), column 'SPR confirmed binders'

Source checking is not independent reproduction. Release 2026-10-22-1b7cf027e4f6.

Use this configuration

How it works, versions and access

Strengths, limitations and unresolved questions

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 2026-10-22-1b7cf027e4f6
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 2026-10-22-1b7cf027e4f6 · Record review: source checked

2 source records and release history

Atomically accurate de novo design of antibodies with RFdiffusion (Bennett et al., Nature 2025) · Original source · Nature, version of record; Europe PMC JATS full text for PMC12727541
Bennett 2025 Supplementary Information, including Supplementary Methods Tables 5 and 6 · Original source · Supplementary Information PDF 41586_2025_9721_MOESM1_ESM.pdf, 129 pages, InDesign 17.4, ModDate 2026-09-15

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: uc-binder-20261022-config-bennett-rfdiffusion-vhh

areas: proteins-complexes; protein-structure
contexts: research
method types: specialist
access reuse: Released as RosettaCommons/RFantibody.
foundation model eligible: false
limitations: The filter is the authors' own fine-tuned RoseTTAFold2, so design and selection are not independent. The paper states the filter's power is limited and that this 'probably accounts for the low experimental success rates'.
method role: design
parameters: 200 backbones designed per target; the best of eight ProteinMPNN sequences per backbone by the filtering method; fine-tuned RoseTTAFold2 as the filter with hotspot proportion = 1, which the Table 6 caption notes 'is likely not the most predictive setting'. Filter success was defined as Ca RMSD < 2 A and pAE interaction < 10.
reported name: fine-tuned RFdiffusion
review: artifact sha256: d7876bc73c47f2cb3fcb0fa1ef917c89c9b67750a22f6402bf515f978cd1564a; method: source-hash-verification; independent-cell-check; ai-assisted-source-review; method note: Re-downloaded from Europe PMC (HTTP 200, 169,840 bytes) and the SHA-256 recomputed and matched. Own flattening of the JATS XML. The main text's '95 designs per target' was confirmed verbatim and the string '96 designs' does not occur, so the conflict with Supplementary Methods Table 6 is real.; note: Reviewed in docs/reviews/use-cases/binder-2026-10-22.md. Automated AI review of the source transcription only; it is not human scientific review and nothing here was reproduced experimentally.; retrieval url: https://www.ebi.ac.uk/europepmc/webservices/rest/PMC12727541/fullTextXML; reviewed at: 2026-10-22T15:00:00Z; reviewer: claude; reviewer note: Claude review agent, independent of the extraction agent; no human review claimed
selection: Designs passing the fine-tuned RoseTTAFold2 filter at hotspot proportion 1.
source locator: Supplementary Methods Table 6 caption; Section 6.8
version: Nature 2025 version of record

Related records

Suggest a correction