TorchCraft framework-conditioned VHH design, raw output
A TorchCraft setting that holds a VHH framework sequence fixed, optimises the three CDR loops under a CDR-restricted antibody language-model prior, and returns the sequence without inverse-folding redesign.
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.
Model Capabilities & Hardware Requirements
Deployment facts recorded for TorchCraft framework-conditioned VHH design, raw output. Fields the record does not carry are marked not recorded.
Evaluated Checkpoint: As published in arXiv:2609.19770v1, on pretrained AlphaFold 3 weights through TorchFold
Evaluations and results
4 evaluations · 4 results. Different protocols are not a single leaderboard.
Protocol: Apparent dissociation constant of a selected VHH design by biolayer interferometry
Sorted by Dissociation constant (apparent KD of a framework-conditioned VHH, measured by biolayer interferometry with the VHH immobilised and fitted with a 1:1 binding model, for one displayed representative design from a filtered and ranked pool, not a campaign hit rate) (lower 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 | Dissociation constant (apparent KD of a framework-conditioned VHH, measured by biolayer interferometry with the VHH immobilised and fitted with a 1:1 binding model, for one displayed representative design from a filtered and ranked pool, not a campaign hit rate) |
|---|---|---|
| TorchCraft framework-conditioned VHH design, raw outputEvaluation | BHRF1 measurements of the VHH validation campaign | 24.3 (best in column) |
| TorchCraft framework-conditioned VHH design, raw outputEvaluation | EFNA5 measurements of the VHH validation campaign | 122 |
| TorchCraft framework-conditioned VHH design, raw outputEvaluation | PDGFRb measurements of the VHH validation campaign | 268 |
| TorchCraft framework-conditioned VHH design, raw outputEvaluation | S100A4 measurements of the VHH validation campaign | 354 |
All 4 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: TorchCraft framework-conditioned VHH design, raw output | Protocol: Apparent dissociation constant of a selected VHH design by biolayer interferometry Dataset subset: BHRF1 measurements of the VHH validation campaign | 24.3 Dissociation constant (apparent KD of a framework-conditioned VHH, measured by biolayer interferometry with the VHH immobilised and fitted with a 1:1 binding model, for one displayed representative design from a filtered and ranked pool, not a campaign hit rate) nanomolar · lower Uncertainty: Not reported by the source: No error, interval or replicate spread is printed for the KD values. Coverage: 1 displayed designs scored (One representative design; the tested denominator is not stated.) | Author-reported evaluation · Source checkedMethods, coverage and sourceMeasured binding of a selected TorchCraft VHH to BHRF1 Not reported Aggregation: Not reported TorchCraft: Unified binder design by inverting an all-atom structure predictor (arXiv:2609.19770v1) · Figure 3a caption (page 5), sentence "Apparent dissociation constants of the displayed binders were 24.3, 122, 268, and 354 nM, respectively."; value for BHRF1. The same number is printed in Section 2.3 and in the Figure 3a panel label "KD: 24.3nM". |
| Configuration: TorchCraft framework-conditioned VHH design, raw output | Protocol: Apparent dissociation constant of a selected VHH design by biolayer interferometry Dataset subset: EFNA5 measurements of the VHH validation campaign | 122 Dissociation constant (apparent KD of a framework-conditioned VHH, measured by biolayer interferometry with the VHH immobilised and fitted with a 1:1 binding model, for one displayed representative design from a filtered and ranked pool, not a campaign hit rate) nanomolar · lower Uncertainty: Not reported by the source: No error, interval or replicate spread is printed for the KD values. Coverage: 1 displayed designs scored (One representative design; the tested denominator is not stated.) | Author-reported evaluation · Source checkedMethods, coverage and sourceMeasured binding of a selected TorchCraft VHH to EFNA5 Not reported Aggregation: Not reported TorchCraft: Unified binder design by inverting an all-atom structure predictor (arXiv:2609.19770v1) · Figure 3a caption (page 5), sentence "Apparent dissociation constants of the displayed binders were 24.3, 122, 268, and 354 nM, respectively."; value for EFNA5. The same number is printed in Section 2.3 and in the Figure 3a panel label "KD: 122nM". |
| Configuration: TorchCraft framework-conditioned VHH design, raw output | Protocol: Apparent dissociation constant of a selected VHH design by biolayer interferometry Dataset subset: PDGFRb measurements of the VHH validation campaign | 268 Dissociation constant (apparent KD of a framework-conditioned VHH, measured by biolayer interferometry with the VHH immobilised and fitted with a 1:1 binding model, for one displayed representative design from a filtered and ranked pool, not a campaign hit rate) nanomolar · lower Uncertainty: Not reported by the source: No error, interval or replicate spread is printed for the KD values. Coverage: 1 displayed designs scored (One representative design; the tested denominator is not stated.) | Author-reported evaluation · Source checkedMethods, coverage and sourceMeasured binding of a selected TorchCraft VHH to PDGFRb Not reported Aggregation: Not reported TorchCraft: Unified binder design by inverting an all-atom structure predictor (arXiv:2609.19770v1) · Figure 3a caption (page 5), sentence "Apparent dissociation constants of the displayed binders were 24.3, 122, 268, and 354 nM, respectively."; value for PDGFRb. The same number is printed in Section 2.3 and in the Figure 3a panel label "KD: 268nM". |
| Configuration: TorchCraft framework-conditioned VHH design, raw output | Protocol: Apparent dissociation constant of a selected VHH design by biolayer interferometry Dataset subset: S100A4 measurements of the VHH validation campaign | 354 Dissociation constant (apparent KD of a framework-conditioned VHH, measured by biolayer interferometry with the VHH immobilised and fitted with a 1:1 binding model, for one displayed representative design from a filtered and ranked pool, not a campaign hit rate) nanomolar · lower Uncertainty: Not reported by the source: No error, interval or replicate spread is printed for the KD values. Coverage: 1 displayed designs scored (One representative design; the tested denominator is not stated.) | Author-reported evaluation · Source checkedMethods, coverage and sourceMeasured binding of a selected TorchCraft VHH to S100A4 Not reported Aggregation: Not reported TorchCraft: Unified binder design by inverting an all-atom structure predictor (arXiv:2609.19770v1) · Figure 3a caption (page 5), sentence "Apparent dissociation constants of the displayed binders were 24.3, 122, 268, and 354 nM, respectively."; value for S100A4. The same number is printed in Section 2.3 and in the Figure 3a panel label "KD: 354nM". |
Source checking is not independent reproduction. Release 2026-11-21-c8bea25e5abb.
Use this configuration
How it works, versions and access
Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.
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
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
Sources and history
Release 2026-11-21-c8bea25e5abb · Record review: source checked
1 source record and release history
- TorchCraft: Unified binder design by inverting an all-atom structure predictor (arXiv:2609.19770v1) · Original source · arXiv:2609.19770v1, submitted 17 September 2026
Technical metadata and record identifiers
Technical metadata for this record is tracked under stable ID torchcraft-20261115-config-vhh-raw.
Source locator: Supplementary Table S1 row 'VHH'; Appendix A.1.3, A.1.4 and A.2.1; Section 2.3
Related records
- configuration of: TorchCraft
- uses model: AlphaFold 3
- uses model: TorchFold on pretrained AlphaFold 3 parameters
- uses model: IgLM
- system: Measured binding of a selected TorchCraft VHH to BHRF1
- system: Measured binding of a selected TorchCraft VHH to EFNA5
- system: Measured binding of a selected TorchCraft VHH to PDGFRb
- system: Measured binding of a selected TorchCraft VHH to S100A4
How to cite
TorchCraft framework-conditioned VHH design, raw output. Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/torchcraft-20261115-config-vhh-raw/
Release 2026-11-21-c8bea25e5abb. Rankings and values change between releases, so keep the release id in the citation. Cite the original publication as well for any value you reuse: each result links its source.
The compilation and annotations made by Rewire in this release are offered under CC BY 4.0; values and text taken from cited sources stay under the terms of those sources and are not covered by this licence. Data licence