Why this evidence?
It is linked to a specific model version, scientific task, dataset, split, metric and source. That makes the claim inspectable rather than a detached marketing score.
DiffDock: Protein–ligand pose prediction
Peer-reviewed top-ranked docking-pose evaluation.
Model versionVersion history not yet curated
TaskProtein–ligand pose prediction
DatasetPDBBind
SplitPDBBind held-out complexes
MetricTop-ranked docking pose
Replicationindependent-benchmarks-exist
Reported byModel developers
Review statuscurated
Evidence confidence · strong
Confidence is multidimensional, not a universal model score.
Evidence completenessstrongModel version, task, dataset, split, metric, source and provenance fields.
Independent validationstrongindependent-benchmarks-exist
Source qualitystrongPeer-reviewed
Reproducibility evidencestrongReflects documented replication status, not a universal reproducibility score.
Version specificitystrongVersion history not yet curated
Context applicabilitystrongDepends on task, split and explicit caveats; users must still validate their own context.
Contradiction reviewclearNo direct contradiction signal is currently queued.
BioAtlas reports evidence dimensions separately so a strong source cannot hide weak applicability, incomplete replication or unresolved contradiction.
Why?
Why should this evidence influence a decision?
Why not a universal score?
Performance can change with dataset, split, preprocessing, metric and context of use. BioAtlas therefore keeps confidence dimensions separate.
What could change the conclusion?
Independent replication, a better matched prospective dataset, a version change, a contradictory result or a more relevant validation protocol can reopen this evidence record.
Evidence boundary
What this claim does not prove.
- Random and temporal splits can produce materially different estimates.
- Pose accuracy is not binding-affinity accuracy.
BioAtlas groups benchmark claims only when task, dataset, split, metric and protocol context align. This record is not a universal model score.
Model context
DiffDock treats docking as generative diffusion over the space of ligand poses rather than an optimization search, giving faster, more accurate blind docking. It became a template for generative approaches to protein–ligand interaction and ships inside NVIDIA BioNeMo.
Full evidence passport →Known model limitations
- Performance depends on the evaluation dataset and operating conditions.
- Task-specific benchmark results should not be compared across unlike domains.
- Outputs require task-specific scientific and experimental validation.