Benchmark

PLINDER

PLINDER supplies annotated protein–ligand systems and evaluation resources for docking.

Sources

plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs

7 evaluations · 33 results

Overview

Datasets

Protein–ligand complexes with linked bound, unbound and predicted receptor structures.

Sources

plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs

Metrics

Pose scoring includes ligand lDDT-PLI, binding-site-superposed symmetry-corrected RMSD and pocket lDDT. System summaries retain mapped-chain fractions and optionally score receptor lDDT, oligomer interfaces and PoseBusters validity. A pose confidence score is optional input, not benchmark uncertainty.

Sources

plinder-config0 primary benchmark evidence · Pinned docs/evaluation.md: Write scores

Allowed inputs

Protein–ligand systems, receptor structures and curated metadata.

Sources

plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs

Evaluation procedure diagram

How it worksEvaluation procedure

Conceptual procedure. Task variants and protocol versions retain their separate scoring conditions.

Sources (2)

plinder-org/plinder official source; plinder-config0 primary benchmark evidence · Pinned README: Overview; Plinder versions; Known bugs; Pinned docs/evaluation.md: Write scores

limited source coverage · Automated source review, 2026-09-16. All specifications and missing details

Results

Each comparison retains its reviewed evaluation scope, dataset and metric. Results are shown without a pooled ranking.

Evaluation scope and datasetMetric and reported setup

Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · co-folding: protein_rmsd

protein_rmsd (angstrom) · Lower values are better.

Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · co-folding · Plinder-L95

Evidence origin: Independent external evaluation, Author-reported evaluation.

Improving Stereochemical Limitations in Protein–Ligand Complex Structure Prediction · Table 1 (XML tbl1); row AF3; column protein_rmsd; footnote a through Table 1 (XML tbl1); row Boltz R; column protein_rmsd; footnote a

Missing source cells and quarantined conflicts are recorded in acquisition and audit tables. Per-result scoring denominators may be unreported.

Comparison details and limitations

Complete selected source table is retained across source-order panels. These point estimates do not establish statistical significance or a universal ranking.

Source-specific evaluation. Input conditions: co-folding. No equivalence to other releases, protocols or model families is inferred.
Exact source-defined evaluation scope; reported scores are not rewire reproductions.

Automated source review: 2026-09-19. Numerical source review does not establish independent reproduction.

Dots show point estimates. Whiskers show only explicitly defined uncertainty (standard deviation, standard error or a labelled interval); their definitions remain in Table. Unresolved uncertainty is not plotted. Differences do not establish statistical significance.

Find a tested model or method

Showing 5 of 5 matching rows.

Zoom to observed range

Tested configuration

0.360.3740.3870.40.413

Reported score

Methods and evaluation design

Procedure, tasks and evaluated configurations

How it works

Evaluation methodology

PLINDER organizes protein–ligand complexes and their similarity relationships so a test set can be characterized by novelty of proteins, pockets, ligands and interactions. The evaluator compares predicted poses with reference systems and preserves matched-chain coverage. Dataset version, split and novelty stratum are essential parts of any reported result.

Sources

plinder-config0 primary benchmark evidence · Pinned docs/evaluation.md: per-pose scores and test stratification

Evaluation design

Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.

Protocols

These source-backed links do not make different protocols or scores interchangeable.

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 6 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.

Baseline status by linked protocol

Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · co-folding · 0/2 roles measured
Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · rigid-receptor docking · 0/2 roles measured
Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · docking with ground-truth-centred search box · 0/2 roles measured

Protocol coverage CSV · Model evaluation matrix · Source table · Release and checksums

Coverage is derived from release 2026-09-29-06401fd5b220. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run this benchmark

Choose a concrete protocol before running an evaluation. Its inputs, split and scoring rules determine which results can be compared.

Run this benchmark

Official data retrieval and package-install instructions are available, with separate release and iteration identifiers. Downloading PLINDER does not produce docking predictions or scores; pin the data release and then select the evaluation workflow and method. The README also records known data issues that matter to downstream benchmarking.

A maintained rewire runner has not been verified for this benchmark. Check data access, weights, licences, dependencies and hardware in the linked official documentation; requirements have not been fully extracted.

plinder-org/plinder / README.md · README.md lines 47–78 and 96–135 (Versions, known bugs and Getting Started)

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths and considerations

Similarity annotations make protein/ligand overlap inspectable when choosing splits.
Sources
plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs

Limitations and conditions

A split can be novel by one molecular similarity measure and familiar by another. The documented evaluator produces scores and strata; it does not certify every model’s training history or impose a common uncertainty protocol.
Sources
plinder-config0 primary benchmark evidence · Pinned docs/evaluation.md: per-pose scores and test stratification

Profile review details

Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction.

Stable record: discovery-benchmark-plinder

Specifications

Inputs, training, access and other details

Explanatory profile: limited source coverage · 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
Property	Description and evidence
Datasets	Protein–ligand complexes with linked bound, unbound and predicted receptor structures. Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs
Splits	Train/validation/test splits can be tuned to the learning task. Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs
Metrics	Pose scoring includes ligand lDDT-PLI, binding-site-superposed symmetry-corrected RMSD and pocket lDDT. System summaries retain mapped-chain fractions and optionally score receptor lDDT, oligomer interfaces and PoseBusters validity. A pose confidence score is optional input, not benchmark uncertainty. Sources plinder-config0 primary benchmark evidence · Pinned docs/evaluation.md: Write scores
Baselines	The official release history identifies a dataset version used to retrain DiffDock and points to the companion Moving Beyond Memorization study. Baseline identity must include the particular PLINDER release/split; a dataset entry does not define one universal reference score. Sources plinder-readme primary benchmark evidence · README: dataset versions and Moving Beyond Memorization reference
Leakage controls	Similarity annotations support task-dependent splitting; a specific split is required before claiming overlap exclusion. Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs
Uncertainty	The inspected evaluator documentation defines per-pose metrics, system averages and similarity strata but no universal bootstrap or repeated-training interval. An uncertainty estimate must be attached to a specific evaluated model and dataset release. · Not reported in inspected sources Sources plinder-config0 primary benchmark evidence · Pinned docs/evaluation.md: per-pose scores and test stratification
Entity type	Protein–ligand dataset and task-dependent split resource. Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs
Organisms	Molecular systems define the collection; it is not tied to one organism. · Not applicable Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs
Assays	Protein–ligand structural data with bound, unbound and predicted receptor states. Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs
Allowed inputs	Protein–ligand systems, receptor structures and curated metadata. Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs
Adaptation	The chosen downstream model and split determine fitting; the data resource imposes no single adaptation scheme. Sources plinder-org/plinder official source · Pinned README: Overview; Plinder versions; Known bugs

Applicable tests and references

Applicability is distinct from a completed evaluation.

Classical molecular docking · proposed
AlphaFold 3 · Proposed association
Boltz · Proposed association

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. Dated primary-source discovery and protocol/table screening. Source checking does not mean experimental reproduction. Only separately extracted and independently reviewed numeric batches are publishable.

Paper or primary resource	Version	Reference
PLINDER: Protein Ligand INteraction Dataset and Evaluation Resource — official repository documentation	Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256	Read source

Historical gaps recorded on 2026-09-17

The catalogue now holds 33 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.

complete comparable numeric result batch: Specific candidate tables and protocol boundaries are documented; no graph values or incomplete winner-only selection are converted into publishable rows.

Search and extraction details

primary protocol reviewed

Searches

PLINDER protein ligand benchmark paper dataset

Evidence locations

Official pinned README, dataset releases and evaluation links; original paper DOI 10.1101/2024.07.17.603955

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 2026-09-29-06401fd5b220
Property and statement	Original source and location	Review and provenance
Diagram caption Conceptual procedure. Task variants and protocol versions retain their separate scoring conditions. Individual claims	plinder-config0 primary benchmark evidence Original source ↗ Pinned README: Overview; Plinder versions; Known bugs; Pinned docs/evaluation.md: Write scores Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4:docs/evaluation.md Retrieved: 2026-09-16T21:11:45.036654+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `23ad37a97187942840416e240d7ee99d0a677a4964a75beed3e8be0082b9375d` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram caption Conceptual procedure. Task variants and protocol versions retain their separate scoring conditions. Individual claims	plinder-org/plinder official source Original source ↗ Pinned README: Overview; Plinder versions; Known bugs; Pinned docs/evaluation.md: Write scores Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4 Retrieved: 2026-09-16T10:30:23.659162+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `1d53c3b89030dc4651d3e7bf4749256b7e579330fc7a660cffaa992d646da34a` Hash scope: Hash scope not separately documented; inspect source record
Diagram steps Allowed inputs: Protein–ligand systems, receptor structures and curated metadata. Splits: Train/validation/test splits can be tuned to the learning task. Metrics: Pose scoring includes ligand lDDT-PLI, binding-site-superposed symmetry-corrected RMSD and pocket lDDT. System summaries retain mapped-chain fractions and optionally score receptor lDDT, oligomer interfaces and PoseBusters validity. A pose confidence score is optional input, not benchmark uncertainty. Individual claims	plinder-config0 primary benchmark evidence Original source ↗ Pinned README: Overview; Plinder versions; Known bugs; Pinned docs/evaluation.md: Write scores Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4:docs/evaluation.md Retrieved: 2026-09-16T21:11:45.036654+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `23ad37a97187942840416e240d7ee99d0a677a4964a75beed3e8be0082b9375d` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps Allowed inputs: Protein–ligand systems, receptor structures and curated metadata. Splits: Train/validation/test splits can be tuned to the learning task. Metrics: Pose scoring includes ligand lDDT-PLI, binding-site-superposed symmetry-corrected RMSD and pocket lDDT. System summaries retain mapped-chain fractions and optionally score receptor lDDT, oligomer interfaces and PoseBusters validity. A pose confidence score is optional input, not benchmark uncertainty. Individual claims	plinder-org/plinder official source Original source ↗ Pinned README: Overview; Plinder versions; Known bugs; Pinned docs/evaluation.md: Write scores Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4 Retrieved: 2026-09-16T10:30:23.659162+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `1d53c3b89030dc4651d3e7bf4749256b7e579330fc7a660cffaa992d646da34a` Hash scope: Hash scope not separately documented; inspect source record
Diagram title Evaluation procedure Individual claims	plinder-config0 primary benchmark evidence Original source ↗ Pinned README: Overview; Plinder versions; Known bugs; Pinned docs/evaluation.md: Write scores Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4:docs/evaluation.md Retrieved: 2026-09-16T21:11:45.036654+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `23ad37a97187942840416e240d7ee99d0a677a4964a75beed3e8be0082b9375d` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Evaluation procedure Individual claims	plinder-org/plinder official source Original source ↗ Pinned README: Overview; Plinder versions; Known bugs; Pinned docs/evaluation.md: Write scores Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4 Retrieved: 2026-09-16T10:30:23.659162+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `1d53c3b89030dc4651d3e7bf4749256b7e579330fc7a660cffaa992d646da34a` Hash scope: Hash scope not separately documented; inspect source record
Datasets Protein–ligand complexes with linked bound, unbound and predicted receptor structures. Individual claims	plinder-org/plinder official source Original source ↗ Pinned README: Overview; Plinder versions; Known bugs Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4 Retrieved: 2026-09-16T10:30:23.659162+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.0.value` Source artifact SHA-256: `1d53c3b89030dc4651d3e7bf4749256b7e579330fc7a660cffaa992d646da34a` Hash scope: Hash scope not separately documented; inspect source record
Splits Train/validation/test splits can be tuned to the learning task. Individual claims	plinder-org/plinder official source Original source ↗ Pinned README: Overview; Plinder versions; Known bugs Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4 Retrieved: 2026-09-16T10:30:23.659162+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.1.value` Source artifact SHA-256: `1d53c3b89030dc4651d3e7bf4749256b7e579330fc7a660cffaa992d646da34a` Hash scope: Hash scope not separately documented; inspect source record
Adaptation The chosen downstream model and split determine fitting; the data resource imposes no single adaptation scheme. Individual claims	plinder-org/plinder official source Original source ↗ Pinned README: Overview; Plinder versions; Known bugs Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4 Retrieved: 2026-09-16T10:30:23.659162+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.10.value` Source artifact SHA-256: `1d53c3b89030dc4651d3e7bf4749256b7e579330fc7a660cffaa992d646da34a` Hash scope: Hash scope not separately documented; inspect source record
Metrics Pose scoring includes ligand lDDT-PLI, binding-site-superposed symmetry-corrected RMSD and pocket lDDT. System summaries retain mapped-chain fractions and optionally score receptor lDDT, oligomer interfaces and PoseBusters validity. A pose confidence score is optional input, not benchmark uncertainty. Individual claims	plinder-config0 primary benchmark evidence Original source ↗ Pinned docs/evaluation.md: Write scores Version: 85b3f1cb1763530a6cfd934f4263a1777c41afa4:docs/evaluation.md Retrieved: 2026-09-16T21:11:45.036654+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.2.value` Source artifact SHA-256: `23ad37a97187942840416e240d7ee99d0a677a4964a75beed3e8be0082b9375d` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

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

Sources and history

View linked audit checks and correction history

Release 2026-09-29-06401fd5b220 · Record review: discovered

5 source records and release history

plinder-org/plinder official source · Original source · 85b3f1cb1763530a6cfd934f4263a1777c41afa4
PLINDER: Protein Ligand INteraction Dataset and Evaluation Resource — official repository documentation · Original source · Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
plinder-org/plinder / README.md · Original source · Git commit 85b3f1cb1763530a6cfd934f4263a1777c41afa4
plinder-config0 primary benchmark evidence · Original source · 85b3f1cb1763530a6cfd934f4263a1777c41afa4:docs/evaluation.md
plinder-readme primary benchmark evidence · Original source · 85b3f1cb1763530a6cfd934f4263a1777c41afa4

Download this release

Technical metadata and extraction receipts

Stable ID: discovery-benchmark-plinder

areas: molecular-interactions
entity level: suite
scope note: Specialist molecular or omics evaluation; protocol details require review before numerical comparison.
task: Protein-ligand interaction evaluation
version: Not reported
benchmark research: review date: 2026-09-17; status: primary_protocol_reviewed; primary sources: evidence-expansion-plinder-official-1d53c3b8; inspected locators: Official pinned README, dataset releases and evaluation links; original paper DOI 10.1101/2024.07.17.603955; searched queries: PLINDER protein ligand benchmark paper dataset; gaps: complete comparable numeric result batch: Specific candidate tables and protocol boundaries are documented; no graph values or incomplete winner-only selection are converted into publishable rows.; claim scope: Dated primary-source discovery and protocol/table screening. Source checking does not mean experimental reproduction. Only separately extracted and independently reviewed numeric batches are publishable.
historical missing metadata: dataset release: unextracted; metric implementation: unextracted; split manifest: unextracted; version: unextracted
metadata review scope: historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
entity classification: review date: 2026-09-17; rationale: The cited profile describes a collection of evaluation tasks or protocols; retain it as the top-level benchmark suite. Its datasets and individual protocols remain separate records.; source ids: src-discovery-plinder-org-plinder; source locator: Pinned README: Overview; Plinder versions; Known bugs; ambiguities: None recorded
run documentation: record id: discovery-benchmark-plinder; source ids: run-doc-plinder-readme-md-85b3f1cb; status: official_documentation_linked; summary: Official data retrieval and package-install instructions are available, with separate release and iteration identifiers. Downloading PLINDER does not produce docking predictions or scores; pin the data release and then select the evaluation workflow and method. The README also records known data issues that matter to downstream benchmarking.; source locator: README.md lines 47–78 and 96–135 (Versions, known bugs and Getting Started)

Related records

part of: Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · co-folding
part of: Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · rigid-receptor docking
part of: Paper-specific Plinder-L95 stereochemistry assessment; Table 1 all entries · Plinder-L95 · docking with ground-truth-centred search box
applicable to: Classical molecular docking
applicable to: AlphaFold 3
applicable to: Boltz

Suggest a correction