Dataset

Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021

Antibody-antigen structures from SAbDab deposited after the AlphaFold3 training cutoff, filtered against earlier entries.

Research readiness

These checks assess whether the evidence supports a reproducible investigation. A source-checked score alone does not meet these requirements.

Release 2026-10-10-6e93f504adfc · Evidence verified: Not verified

Evidence incomplete

Replay metrics

Exact outcomes, predictions, identifiers and evaluator are connected.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing

Verified: Not verified

Evidence incomplete

Investigate discrepancies

Replay evidence includes annotations and an assessment of dependence. Unknown independence permits descriptive analysis only.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing
annotations: verification is missing
dependence: verification is missing

Verified: Not verified

Evidence incomplete

Run locally

A pinned recipe describes the inputs, environment and resource requirements.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
recipe pinned: verification is missing
resource estimate: verification is missing

Verified: Not verified

Evidence incomplete

Validate independently

Separate data and exposure records support an independent test.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
independent validation: verification is missing
overlap checked: verification is missing

Verified: Not verified

Readiness describes the evidence in this release. Availability on your computer is checked separately when an investigation runs. Existing data exposure can prevent independent validation even when files are available.

Artifacts and reproduction

No verified artifact manifest is connected to this record yet. The gaps above identify what is needed before analysis can begin.

Read reviewed discrepancy investigations

Evaluation results

9 evaluations · 27 results. Different protocols are not a single leaderboard.

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: AlphaFold3, 200 models per target, top model picked by aeiTM: ipTM recomputed from true aligned errors (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.508 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeiTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeiTM', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by aeiTM: ipTM recomputed from true aligned errors (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.873 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeiTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeiTM', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by aeiTM: ipTM recomputed from true aligned errors (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.562 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeiTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeiTM', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by aeRankConf: 0.2 aeTM + 0.8 aeiTM (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.511 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeRankConf on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeRankConf', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by aeRankConf: 0.2 aeTM + 0.8 aeiTM (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.879 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeRankConf on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeRankConf', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by aeRankConf: 0.2 aeTM + 0.8 aeiTM (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.595 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeRankConf on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeRankConf', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by aeTM: pTM recomputed from true aligned errors (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.516 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeTM', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by aeTM: pTM recomputed from true aligned errors (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.814 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeTM', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by aeTM: pTM recomputed from true aligned errors (oracle) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.665 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by aeTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'aeTM', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by DockQ against the experimental structure (oracle upper bound) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.544 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by DockQ on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'DockQ', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by DockQ against the experimental structure (oracle upper bound) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	1 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by DockQ on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'DockQ', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by DockQ against the experimental structure (oracle upper bound) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	1 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by DockQ on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'DockQ', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by ipSAE (Dunbrack 2025), PAE and distance cut-off 10 Å (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.353 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by ipSAE on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'ipSAE', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by ipSAE (Dunbrack 2025), PAE and distance cut-off 10 Å (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.75 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by ipSAE on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'ipSAE', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by ipSAE (Dunbrack 2025), PAE and distance cut-off 10 Å (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.247 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by ipSAE on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'ipSAE', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by AlphaFold3 ipTM (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.37 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by ipTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'ipTM', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by AlphaFold3 ipTM (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.707 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by ipTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'ipTM', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by AlphaFold3 ipTM (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.22 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by ipTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'ipTM', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by pDockQ2 (Zhu et al. 2023) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.352 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by pDockQ2 on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'pDockQ2', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by pDockQ2 (Zhu et al. 2023) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.603 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by pDockQ2 on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'pDockQ2', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by pDockQ2 (Zhu et al. 2023) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.184 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source AlphaFold3 top model by pDockQ2 on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'pDockQ2', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by AlphaFold3 pTM (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.369 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by pTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'pTM', column '<DockQ>'
Configuration: AlphaFold3, 200 models per target, top model picked by AlphaFold3 pTM (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.67 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by pTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'pTM', column 'R'
Configuration: AlphaFold3, 200 models per target, top model picked by AlphaFold3 pTM (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.2 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by pTM on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'pTM', column '<R>'
Configuration: AlphaFold3, 200 models per target, top model picked by AlphaFold3 ranking confidence (0.2 pTM + 0.8 ipTM) (Fromm et al. 2026)	Protocol: Picking the top AlphaFold3 model per antibody-antigen target with a confidence score Dataset: Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021	0.375 dockq unitless · higher Uncertainty: Not reported by the source Coverage: 110 scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaFold3 top model by RankConf on 110 post-cutoff antibody-antigen complexes (Fromm et al. 2026) structural-20261009-protocol-fromm2026-abag-model-selection Aggregation: Not reported Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Table 1, row 'RankConf', column '<DockQ>'

Source checking is not independent reproduction. Release 2026-10-10-6e93f504adfc.

Dataset and evaluation context

A dataset supplies biological observations. The evaluation protocol defines how those observations are split, used and scored.

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.

7 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-10-6e93f504adfc
Property and statement	Original source and location	Review and provenance
attributes.population 110 antibody-antigen complexes deposited after 30 September 2021, X-ray or electron microscopy at 3.5 Å or better, under 80% sequence identity per chain to any entry before that date, then clustered at 80%; antibody-peptide complexes excluded Context-only references	Evaluating deep learning based structure prediction methods on antibody-antigen complexes Original source ↗ Section 2.1 (Dataset), paragraphs 1 to 3 Version: Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML Retrieved: 2026-10-09T21:21:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.population` Source artifact SHA-256: `a7320667ed6f7d8df440d90275d976b91d2fe98411172a79c6a304eb3f02275d` Hash scope: JATS XML parse (xml.etree), by extract/extract_structural.py Inspected artifact
attributes.scope_note No two antibodies are identical; one antigen appears twice with two antibodies. 15 antigen sequences occur in the training set. Concatenated CDR identity to the training set is at most 72%. Context-only references	Evaluating deep learning based structure prediction methods on antibody-antigen complexes Original source ↗ Section 2.1 (Dataset), paragraphs 1 to 3 Version: Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML Retrieved: 2026-10-09T21:21:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.scope_note` Source artifact SHA-256: `a7320667ed6f7d8df440d90275d976b91d2fe98411172a79c6a304eb3f02275d` Hash scope: JATS XML parse (xml.etree), by extract/extract_structural.py Inspected artifact
attributes.source_locator Section 2.1 (Dataset), paragraphs 1 to 3 Context-only references	Evaluating deep learning based structure prediction methods on antibody-antigen complexes Original source ↗ Section 2.1 (Dataset), paragraphs 1 to 3 Version: Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML Retrieved: 2026-10-09T21:21:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.source_locator` Source artifact SHA-256: `a7320667ed6f7d8df440d90275d976b91d2fe98411172a79c6a304eb3f02275d` Hash scope: JATS XML parse (xml.etree), by extract/extract_structural.py Inspected artifact
attributes.total 110 Context-only references	Evaluating deep learning based structure prediction methods on antibody-antigen complexes Original source ↗ Section 2.1 (Dataset), paragraphs 1 to 3 Version: Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML Retrieved: 2026-10-09T21:21:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.total` Source artifact SHA-256: `a7320667ed6f7d8df440d90275d976b91d2fe98411172a79c6a304eb3f02275d` Hash scope: JATS XML parse (xml.etree), by extract/extract_structural.py Inspected artifact
attributes.version SAbDab entries as of 23 October 2024, built with a modified AADaM; Zenodo 10.5281/zenodo.17978681 Context-only references	Evaluating deep learning based structure prediction methods on antibody-antigen complexes Original source ↗ Section 2.1 (Dataset), paragraphs 1 to 3 Version: Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML Retrieved: 2026-10-09T21:21:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.version` Source artifact SHA-256: `a7320667ed6f7d8df440d90275d976b91d2fe98411172a79c6a304eb3f02275d` Hash scope: JATS XML parse (xml.etree), by extract/extract_structural.py Inspected artifact
description Antibody-antigen structures from SAbDab deposited after the AlphaFold3 training cutoff, filtered against earlier entries. Context-only references	Evaluating deep learning based structure prediction methods on antibody-antigen complexes Original source ↗ Section 2.1 (Dataset), paragraphs 1 to 3 Version: Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML Retrieved: 2026-10-09T21:21:58Z	not individually reviewed No individual claim review recorded Audit details Field: `description` Source artifact SHA-256: `a7320667ed6f7d8df440d90275d976b91d2fe98411172a79c6a304eb3f02275d` Hash scope: JATS XML parse (xml.etree), by extract/extract_structural.py Inspected artifact
name Fromm et al. 2026 antibody-antigen benchmark: 110 complexes released after 30 September 2021 Context-only references	Evaluating deep learning based structure prediction methods on antibody-antigen complexes Original source ↗ Section 2.1 (Dataset), paragraphs 1 to 3 Version: Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML Retrieved: 2026-10-09T21:21:58Z	not individually reviewed No individual claim review recorded Audit details Field: `name` Source artifact SHA-256: `a7320667ed6f7d8df440d90275d976b91d2fe98411172a79c6a304eb3f02275d` Hash scope: JATS XML parse (xml.etree), by extract/extract_structural.py Inspected artifact

All evidence (JSONL) (gzip)How to read the evidence

Sources and history

Release 2026-10-10-6e93f504adfc · Record review: source checked

1 source records and release history

Evaluating deep learning based structure prediction methods on antibody-antigen complexes · Original source · Bioinformatics 42(4):btag136, 2026; PMC13061134 full-text XML

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: structural-20261009-data-fromm2026-abag-110

areas: proteins-complexes
contexts: research
version: SAbDab entries as of 23 October 2024, built with a modified AADaM; Zenodo 10.5281/zenodo.17978681
population: 110 antibody-antigen complexes deposited after 30 September 2021, X-ray or electron microscopy at 3.5 Å or better, under 80% sequence identity per chain to any entry before that date, then clustered at 80%; antibody-peptide complexes excluded
total: 110
scope note: No two antibodies are identical; one antigen appears twice with two antibodies. 15 antigen sequences occur in the training set. Concatenated CDR identity to the training set is at most 72%.
source locator: Section 2.1 (Dataset), paragraphs 1 to 3

Related records

Suggest a correction