dataset · audited-with-caveats · verified 2026-07-29

PPB-Affinity

A reusable protein-protein binding-affinity dataset with complex structures, measured affinities, receptor and ligand chains, and mutation annotations.

Audited with caveats: 2 field(s) are marked provisional or conflicted. Warnings are shown next to affected values and these claims are excluded from unqualified comparisons.

Benchmark definition

What is counted

Version
initial-release
Total
12062 (Whole-dataset unique-sample total explicitly reported by the source; the detailed uniqueness basis remains conflicted.)
Task formats
unclassified
Capabilities
Prediction
Modalities
Table3D structure

Version history

VersionStatusRelease / as-ofTotalFormal tracks
initial-release
ppb-affinity-initial-release-version
current2024-12-0312062 (Whole-dataset unique-sample total explicitly reported by the source; the detailed uniqueness basis remains conflicted.)None registered

Scientific Task Atlas

Scientific task classification

partial for initial-release. No Scientific Task claim passed independent high-confidence verification; task mapping remains pending a targeted official-source audit.

The source names only a broad direction; no more specific leaf task can be assigned without inference.

Relationship registry

How works use this benchmark

Partial claims, non-evaluation uses, and third-party summaries stay visible without entering model comparisons.

Partial evaluation claims

evaluation

explainable-protein-protein-binding-affinity-predictio-ppb-affinity-2-use

Partialsubset

Work: Explainable protein-protein binding affinity prediction via fine-tuning protein language models · source version explainable-protein-protein-binding-affinity-predictio-biorxiv-version-posted-2

Selection
formal subset · Random five-fold cross-validation with 80% training and 20% testing per fold
Metrics
RMSE
Linked runs
None

Not reported / unresolved: Benchmark version is not reported.; The exact realized random-split test n is not reported; only approximately 2,404 per fold is stated.; Model providers and versions are not explicitly reported.; benchmark version; realized n/scope

AI-assisted double-pass extraction; values are limited to independently supported claims.

Evidence
  • table: Table S2
    Supports: /relation_type
  • table: Table S1
    Supports: /benchmark_id
  • section: 4.2
    Supports: /scope
  • section: 4.2
    Supports: /scope
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /metric_labels

evaluation

explainable-protein-protein-binding-affinity-predictio-ppb-affinity-3-use

Partialsubset

Work: Explainable protein-protein binding affinity prediction via fine-tuning protein language models · source version explainable-protein-protein-binding-affinity-predictio-biorxiv-version-posted-2

Selection
formal subset · GroupKFold cold split grouping interactions by PDB ID
Metrics
RMSE
Linked runs
None

Not reported / unresolved: Benchmark version and exact realized test n are not reported.; Model providers and versions are not explicitly reported.; benchmark version; realized n/scope

AI-assisted double-pass extraction; values are limited to independently supported claims.

Evidence
  • table: Table S2
    Supports: /relation_type
  • section: 2.2
    Supports: /benchmark_id
  • section: 4.2
    Supports: /scope
  • section: 4.2
    Supports: /scope
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /metric_labels

evaluation

explainable-protein-protein-binding-affinity-predictio-ppb-affinity-4-use

Partialsubset

Work: Explainable protein-protein binding affinity prediction via fine-tuning protein language models · source version explainable-protein-protein-binding-affinity-predictio-biorxiv-version-posted-2

Selection
formal subset · K-mer Jaccard clustering with K=3, 30% similarity threshold, and pair exclusion
Metrics
RMSE
Linked runs
None

Not reported / unresolved: Benchmark version and exact realized test n are not reported.; Model providers and versions are not explicitly reported.; benchmark version; realized n/scope

AI-assisted double-pass extraction; values are limited to independently supported claims.

Evidence
  • table: Table S2
    Supports: /relation_type
  • section: 2.1
    Supports: /benchmark_id
  • section: 4.2
    Supports: /scope
  • section: 4.2
    Supports: /scope
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /model_ids
  • table: Table S2
    Supports: /metric_labels

evaluation

ppb-affinity-protein-protein-binding-affinity-dataset-ppb-affinity-2-use

Partialunknown

Work: PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · source version ppb-affinity-protein-protein-binding-affinity-dataset-2024-12-03

Selection
not reported
Metrics
Not reported / not applicable
Linked runs
None

Not reported / unresolved: The testing-subset size and selection procedure are not printed in the available text.; Repeats, seed, train-validation-test partition sizes, and model version are not reported in the available text.; Linked Figures 4–5 are unavailable locally, so additional labeled subgroup and source-partition results could not be inspected.; benchmark version; realized n/scope; metric; numeric result; prompt and tools; grader and repeats

Owner-reviewed conservative publication: the creator evaluation is retained only as a partial relationship; conflicted settings and outcomes are omitted pending manual reconciliation.

Evidence
  • section: Technical Validation > Whole dataset, paragraph 47
    Supports: /relation_type
  • section: Abstract, paragraph 1
    Supports: /benchmark_id
  • section: Author contributions; Technical Validation > A benchmark affinity prediction, paragraph 46
    Supports: /model_ids

Creation, training, validation, or model-selection uses

training

explainable-protein-protein-binding-affinity-predictio-ppb-affinity-1-use

Non-evaluationunknown

Work: Explainable protein-protein binding affinity prediction via fine-tuning protein language models · source version explainable-protein-protein-binding-affinity-predictio-biorxiv-version-posted-2

Selection
not applicable
Models
BALM-PPI
Metrics
Not reported / not applicable
Linked runs
None

Not reported / unresolved: Benchmark version is not reported.; Exact per-fold training sizes are not reported for cold and sequence-similarity splits.; BALM-PPI provider and version are not explicitly reported.

AI-assisted double-pass extraction; values are limited to independently supported claims.

Evidence
  • section: 2.1
    Supports: /relation_type
  • section: 2.1
    Supports: /benchmark_id
  • section: 4.1 Dataset
    Supports: /scope
  • section: 4.1 Dataset
    Supports: /scope
  • table: Table S1
    Supports: /scope
  • section: 4.1 Dataset
    Supports: /scope
  • section: 2.1
    Supports: /model_ids

benchmark creation

ppb-affinity-protein-protein-binding-affinity-dataset-ppb-affinity-1-use

Non-evaluationunknown

Work: PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · source version ppb-affinity-protein-protein-binding-affinity-dataset-2024-12-03

Selection
not applicable
Models
Not reported / not applicable
Metrics
Not reported / not applicable
Linked runs
None

Not reported / unresolved: No formal PPB-Affinity version label is printed.; The exact dataset release date is not printed; the data citation reports only 2024.; The licenses of the two GitHub repositories are not printed.; Figure 3 raster content is unavailable locally, preventing inspection of any additional labeled partition values.

AI-assisted double-pass extraction; values are limited to independently supported claims.

Evidence
  • section: Background & Summary, paragraph 4
    Supports: /relation_type
  • section: Background & Summary, paragraph 4
    Supports: /benchmark_id

training

ppb-affinity-protein-protein-binding-affinity-dataset-ppb-affinity-3-use

Non-evaluationunknown

Work: PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · source version ppb-affinity-protein-protein-binding-affinity-dataset-2024-12-03

Selection
not applicable
Metrics
Not reported / not applicable
Linked runs
None

Not reported / unresolved: The training percentages shown in Figure 5F cannot be inspected because the linked raster is unavailable locally.; Training sample counts, split procedure, repeats, seed, hyperparameters, and model version are not reported in the available text.

AI-assisted double-pass extraction; values are limited to independently supported claims.

Evidence
  • section: Technical Validation > Source datasets, paragraph 50
    Supports: /relation_type
  • section: Abstract, paragraph 1; Technical Validation > Source datasets, paragraph 50
    Supports: /benchmark_id
  • section: Technical Validation > Source datasets, paragraph 50
    Supports: /scope
  • section: Author contributions; Technical Validation > A benchmark affinity prediction, paragraph 46; Source datasets, paragraph 50
    Supports: /model_ids

Evaluation registry

Works and run settings

A setting change—scope, prompt, tools, budget, grader, or repeats—creates a separate run. Charts never cross a comparability group.

No normalized evaluation run is published yet. Creator evidence is still attached below.

Evidence and change history

Source locators remain visible; expand an item to inspect the exact Registry fields it supports.

PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Data Record, paragraph 24 · Supports 1 field

Open source →

  • /access/artifacts
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Data Record, paragraph 24; Usage Notes, paragraph 53 · Supports 1 field

Open source →

  • /access/level
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Data Record, paragraph 24 · Supports 1 field

Open source →

  • /access/license
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Usage Notes > Potential uses of the dataset, paragraph 54 · Supports 1 field

Open source →

  • /capabilities
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Abstract, paragraph 1 · Supports 1 field

Open source →

  • /domains
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Abstract, paragraph 1 · Supports 1 field

Open source →

  • /kind
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Data Record, paragraph 24 · Supports 1 field

Open source →

  • /modalities
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Background & Summary, paragraph 4 · Supports 1 field

Open source →

  • /name
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Front matter > Author affiliations · Supports 1 field

Open source →

  • /organizations
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Methods > Data processing and quality assessment, paragraph 6; Data Record, paragraph 24 · Supports 1 field

Open source →

  • /summary
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · other: Crossref bibliographic metadata (Resolved from the canonical paper identifier during intake.) · Supports 1 field

Open source →

  • /release_date
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Methods > Data processing and quality assessment > Protein chains, paragraphs 16–18; Data Record > Metadata, paragraph 43 · Supports 2 fields

Open source →

  • /task_counts/total
  • /versions/0/task_counts
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Code availability · Supports 2 fields

Open source →

  • /resources
  • /implementations
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Data Record, paragraph 24 · Supports 1 field

Open source →

  • /resources/0
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Data Record, paragraph 24 · Supports 2 fields

Open source →

  • /latest_version
  • /versions/0
PPB-Affinity: Protein-Protein Binding Affinity dataset for AI-based protein drug discovery · section: Methods > Data processing and quality assessment > Protein chains, paragraphs 16–18; Data Record > Metadata, paragraph 43 · Supports 2 fields

Open source →

  • /task_counts/basis
  • /task_counts/subsets

Unresolved field claims

  • /task_counts/basisConflicted · high — The owner approved the explicit root-total value after the extractor reported it and the verifier independently located it at high confidence; the detailed uniqueness basis remains conflicted and all subcounts are excluded from publication.
    Evidence: ppb-affinity-automated-count-conflict-evidence
  • /task_counts/subsetsConflicted · high — The owner approved the explicit root-total value after the extractor reported it and the verifier independently located it at high confidence; the detailed uniqueness basis remains conflicted and all subcounts are excluded from publication.
    Evidence: ppb-affinity-automated-count-conflict-evidence

View source-level modification history on GitHub →