rewirebio.iobenchmarks
Task

antigen-antibody HIV neutralization prediction

Antibody neutralization classification is one of several distinct evaluations in the DeepInterAware paper.

SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

1 evaluation · 1 result

Overview

Datasets

The HIV dataset is used for neutralization prediction; other datasets address binding, binding sites and free-energy changes.

Metrics

AUROC, AUPRC and MCC are primary metrics for binding and neutralization classification.

Allowed inputs

Antibody–antigen pair representations.

SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
Evaluation procedure diagram
How it worksComputational evaluation flow
Computational evaluation flow1. Input: Antibody–antigen pair representations.. Then: 2. Evaluation: The HIV neutralization table separates antibody-unseen evaluation; exact partition construction is delegated to Supporting Information §S1.1.. Then: 3. Readout: AUROC, AUPRC and MCC are primary metrics for binding and neutralization classification.Computational evaluation flow1. Input: Antibody–antigen pair representations.. Then: 2. Evaluation: The HIV neutralization table separates antibody-unseen evaluation; exact partition construction is delegated to Supporting Information §S1.1.. Then: 3. Readout: AUROC, AUPRC and MCC are primary metrics for binding and neutralization classification.Computational evaluation flow1. Input: Antibody–antigen pair representations.. Then: 2. Evaluation: The HIV neutralization table separates antibody-unseen evaluation; exact partition construction is delegated to Supporting Information §S1.1.. Then: 3. Readout: AUROC, AUPRC and MCC are primary metrics for binding and neutralization classification.

Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol.

SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Source reviewed · Automated source review, 2026-09-16. All specifications and missing details

1 recorded evaluation, 1 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

1 evaluation · 1 result. Different protocols are not a single leaderboard.

Filter evaluations

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configurationProtocol and datasetFindingEvidence and details
Configuration: DeepInterAwareTask: antigen-antibody HIV neutralization prediction
Dataset: HIV neutralization
0.826 AUROC
unitless · unknown

Uncertainty: ± 0.017 (type not stated)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

DeepInterAware: antigen-antibody HIV neutralization prediction

Sequence-based interface-aware model, antibody-unseen split

Aggregation: Not reported

DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Table 2, Ab Unseen section, DeepInterAware row, AUROC column

Source checking is not independent reproduction. Release 2026-11-21-adc15377fab2.

Methods and evaluation design

Procedure, tasks and evaluated configurations

How it works

Evaluation methodology

The HIV dataset is used for neutralization prediction; other datasets address binding, binding sites and free-energy changes. The HIV neutralization table separates antibody-unseen evaluation; exact partition construction is delegated to Supporting Information §S1.1. AUROC, AUPRC and MCC are primary metrics for binding and neutralization classification. The HIV neutralization table compares DrugBAN, ESM2AbLang, ESM2AntiBERTy, ResPPI, PIPR, AbAgIntPre, MasonsCNN and DeepAAI. HIV pairs with antigen and antibody homology above 0.9 are filtered. Ab Unseen holds out 15% of antibodies; Ag Unseen holds out 20% of antigens. Ag&Ab Unseen additionally removes from training all pairs involving test antibodies. The other partner is deliberately shared in each single-partner holdout.

Sources (2)DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data; deepinteraware-2025__ADVS-12-2412533-s001.pdf · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions; Supporting Information S1.1, HIV dataset and three neutralization scenarios

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Run instructions

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

A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths supported by sources

No source-reviewed explanatory claims are recorded here yet.

Limitations and conditions

Profile review details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Stable record: reported-task-45ead9a1eddf8d

Specifications

Inputs, training, access and other details

Explanatory profile: source reviewed · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.

Data, procedure and scoring
PropertyDescription and evidence
DatasetsThe HIV dataset is used for neutralization prediction; other datasets address binding, binding sites and free-energy changes.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
SplitsThe HIV neutralization table separates antibody-unseen evaluation; exact partition construction is delegated to Supporting Information §S1.1.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
MetricsAUROC, AUPRC and MCC are primary metrics for binding and neutralization classification.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
BaselinesThe HIV neutralization table compares DrugBAN, ESM2AbLang, ESM2AntiBERTy, ResPPI, PIPR, AbAgIntPre, MasonsCNN and DeepAAI.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
Leakage controlsHIV pairs with antigen and antibody homology above 0.9 are filtered. Ab Unseen holds out 15% of antibodies; Ag Unseen holds out 20% of antigens. Ag&Ab Unseen additionally removes from training all pairs involving test antibodies. The other partner is deliberately shared in each single-partner holdout.
Sources (2)DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data; deepinteraware-2025__ADVS-12-2412533-s001.pdf · Supporting Information S1.1, HIV dataset and three neutralization scenarios
UncertaintyEach HIV scenario is repeated with five distinct random seeds and dataset splits. Supporting Information S1.1 calls the reported summaries means and variances; this wording is preserved rather than silently interpreting the table’s ± terms as 95% confidence intervals.
Sources (2)DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data; deepinteraware-2025__ADVS-12-2412533-s001.pdf · Supporting Information S1.1 HIV; Table 2
Entity typePaper-specific computational evaluation protocol.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
OrganismsHIV neutralization collection for this task.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
AssaysAntibody neutralization outcomes; other paper tasks use distinct binding/free-energy labels.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
Allowed inputsAntibody–antigen pair representations.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions
AdaptationSupervised antibody–antigen neutralization classification; the antibody-unseen test is identified separately from binding and transfer tasks.
SourcesDeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data · Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Evidence

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

Papers and result coverage

Last literature check: 2026-09-17. Primary-paper discovery and source inspection. Source-checked results are not independently reproduced experiments.

Historical gaps recorded on 2026-09-17

The catalogue now holds 1 result rows for this benchmark. A note below about pending extraction describes the state on 2026-09-17 and may since have been answered by a later batch. The result rows and their sources are the current record.

  • Keep the three held-out-entity regimes separate. Section S1 contains detailed split/hyperparameter information; no sequence-design or biological optimization material is part of this extraction.
Search and extraction details

primary comparison table screened

Searches

  • "DeepInterAware" 2025

Evidence locations

  • Table 2
  • Experimental setting: model evaluation metrics

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.

21 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-adc15377fab2
Property and statementOriginal source and locationReview and provenance
Diagram caption
Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.diagram.caption

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Diagram steps
  • Input: Antibody–antigen pair representations.
  • Evaluation: The HIV neutralization table separates antibody-unseen evaluation; exact partition construction is delegated to Supporting Information §S1.1.
  • Readout: AUROC, AUPRC and MCC are primary metrics for binding and neutralization classification.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.diagram.steps

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Diagram title
Computational evaluation flow
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.diagram.title

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Datasets
The HIV dataset is used for neutralization prediction; other datasets address binding, binding sites and free-energy changes.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.facts.0.value

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Splits
The HIV neutralization table separates antibody-unseen evaluation; exact partition construction is delegated to Supporting Information §S1.1.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.facts.1.value

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Adaptation
Supervised antibody–antigen neutralization classification; the antibody-unseen test is identified separately from binding and transfer tasks.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.facts.10.value

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Metrics
AUROC, AUPRC and MCC are primary metrics for binding and neutralization classification.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.facts.2.value

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Baselines
The HIV neutralization table compares DrugBAN, ESM2AbLang, ESM2AntiBERTy, ResPPI, PIPR, AbAgIntPre, MasonsCNN and DeepAAI.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Experimental Setting—Datasets; Model Evaluation Metrics; cached text lines 85–93; matching task comparison table/ablation captions

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.facts.3.value

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Leakage controls
HIV pairs with antigen and antibody homology above 0.9 are filtered. Ab Unseen holds out 15% of antibodies; Ag Unseen holds out 20% of antigens. Ag&Ab Unseen additionally removes from training all pairs involving test antibodies. The other partner is deliberately shared in each single-partner holdout.
Individual claims
DeepInterAware: Deep Interaction Interface‐Aware Network for Improving Antigen‐Antibody Interaction Prediction from Sequence Data

Original source ↗

Supporting Information S1.1, HIV dataset and three neutralization scenarios

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: version of record
Retrieved: 2026-09-16T10:41:06Z

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.facts.4.value

Source artifact SHA-256: 25d3561934965f754d8712ec02b2052e9a3979e433b88ecebd5db14e930f17a1

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Leakage controls
HIV pairs with antigen and antibody homology above 0.9 are filtered. Ab Unseen holds out 15% of antibodies; Ag Unseen holds out 20% of antigens. Ag&Ab Unseen additionally removes from training all pairs involving test antibodies. The other partner is deliberately shared in each single-partner holdout.
Individual claims
deepinteraware-2025__ADVS-12-2412533-s001.pdf

Original source ↗

Supporting Information S1.1, HIV dataset and three neutralization scenarios

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: Retrieved 2026-09-16; sha256:3c66d0d9561d744ca0e42a5d33c020719e6f46d7ed27e37a25c19c2bd8d89345
Retrieved: 2026-09-16T21:05:56.618962+00:00

source checked

automated source review · 2026-09-16

Audit details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Field: attributes.profile.facts.4.value

Source artifact SHA-256: 3c66d0d9561d744ca0e42a5d33c020719e6f46d7ed27e37a25c19c2bd8d89345

Hash scope: Hash scope not separately documented; inspect source record

Archive member: ADVS-12-2412533-s001.pdf

Inspected artifact

How to read the evidence

Sources and history

Release 2026-11-21-adc15377fab2 · Record review: needs review

3 source records and release historyDownload this release (gzip)
Technical metadata and record identifiers

Technical metadata for this record is tracked under stable ID reported-task-45ead9a1eddf8d.

Related records

Suggest a correction