Bennett VHH computational success rate under an AlphaFold3 filter; Nature 2025 version of record; Supplementary Methods Table 6

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.

6 recorded evaluations, 6 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

6 evaluations · 6 results. Different protocols are not a single leaderboard.

Protocol: Bennett VHH computational success rate under an AlphaFold3 filter

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)
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design target TcdB	27.5% (best in column)
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design target Influenza Hemagglutinin	5%
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design target Influenza Hemagglutinin	5%
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design target IL-7 Receptor α	5%
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design target SARS-CoV-2 RBD	0%
Fine-tuned RFdiffusion VHH design with fine-tuned RoseTTAFold2 filtering at hotspot proportion 1Evaluation	Bennett VHH design target RSV-F Site III	0.0*%

All 6 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 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 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 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 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 an AlphaFold3 filter Dataset: Bennett VHH design target RSV-F Site III	*0.0%** 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 RSV site 3, 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 'RSV site 3' (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 an AlphaFold3 filter Dataset: Bennett VHH design target TcdB	27.5% 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 TcdB, 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 'TcdB' (Designs screened = 96), column 'Success rate (%) (AF3)'

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

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

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.

Author-reported evaluations: 6

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 2026-10-22-1b7cf027e4f6. 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 2026-10-22-1b7cf027e4f6
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-protocol-bennett-insilico-af3-success

areas: proteins-complexes; protein-structure
contexts: research
limitations: An in silico filter pass rate, not a measured outcome.; Retrospective: AlphaFold3 was released after the design work in the study.; The caption states that 'in some cases, the cropped target used during design was not accurate re-predicted by AF3 ... and hence the success rates are very low', and that 'success rates would likely be higher with contiguous (non-cropped) targets'.; The RSV site 3 value carries an asterisk: 'RSV site 3 target epitope was not correctly predicted with AF3 (RMSD = 19A) in its cropped form used in the design methods'.
metric: success-rate
metric direction: higher
metric qualifier: computational filter pass rate, AlphaFold3, Ca RMSD below 5 A over the whole complex and ipTM above 0.5
protocol: The same design sets are predicted with AlphaFold3 and scored as successful where Ca RMSD over the whole complex is below 5 A and ipTM is above 0.5. Unlike the fine-tuned RoseTTAFold2 filter, AlphaFold3 does not fix the target, so a cropped target that AlphaFold3 re-predicts inaccurately drives the rate down.
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
scope limitations: An in silico filter pass rate, not a measured outcome, and retrospective: AlphaFold3 was released after the design work.
source locator: Supplementary Methods Table 6, column 'Success rate (%) (AF3)'
success threshold: criterion: Ca RMSD < 5 A over the whole complex and ipTM > 0.5; locator: Supplementary Methods Table 6 caption
unit: percent
version: Nature 2025 version of record; Supplementary Methods Table 6

Related records