Configuration

ProQ3D (CAD-trained; rotameric optimization), reported by Pagès et al.

ProQ3D, a deep-learning model quality assessment method that uses structural information, Rosetta energy terms and sequence-derived evolutionary information. ATOM3D reuses scores reported by Pagès et al.

Identified as ProQ3D. The source table prints [Pagès et al., 2019].

Configuration: ProQ3D version trained on CAD-score with rotameric optimization, run by Pagès et al. with a 2013 UniRef database. ATOM3D reuses the CASP 11 stage 2 scores that Pagès et al. report.

ATOM3D App. F.7 takes its PSR comparison 'as reported in [Pagès et al., 2019]' and names 3DCNN, Ornate and Proq3D as the reported methods. All six ATOM3D Table 8 statistics (mean RP, RS, RK 0.444, 0.432, 0.304; global RP, RS, RK 0.772, 0.796, 0.594) equal the CASP 11 stage 2 Proq3D row of Pagès et al. Table 2; the Ornate and 3dCNN rows differ. The identity rests on ATOM3D's attribution to Pagès et al. together with a match on all six statistics to a single row. One matching value would not be enough.

ProQ3D version: Trained on CAD-score with rotameric optimization enabled, run with a 2013 UniRef sequence database.
Reported scores: Pagès et al. Table 2 correlations are against GDT-TS on CASP 11 stage 2 models.

Not established: Checkpoint. No ProQ3D software revision or checkpoint is identified. Test set equivalence. Neither source establishes that the Pagès et al. CASP 11 stage 2 model set is identical to the ATOM3D PSR test split.

ATOM3D: Tasks On Molecules in Three Dimensions; Protein model quality assessment using 3D oriented convolutional neural networks (HAL hal-01899468v1) · ATOM3D arXiv:2012.04035v4 Table 6, p.8; App. F.7, p.25; Table 8 PSR rows, p.28. Pagès et al., HAL hal-01899468v1 (preprint of Bioinformatics 35(18):3313, doi 10.1093/bioinformatics/btz122): Proq3D setup, PDF p.5; Table 2, CASP 11 stage 2 Proq3D row, PDF p.6.. AI-assisted source review on 2026-09-24; no human scientific review.

2 evaluations · 2 results

Overview

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

Evaluations and results

2 evaluations · 2 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: ProQ3D (CAD-trained; rotameric optimization), reported by Pagès et al. (cited as [Pagès et al., 2019])	Task: ATOM3D PSR-GLOBAL-RS: Protein structure ranking, global Spearman Dataset subset: ATOM3D PSR (ATOM3D split)	0.796 spearman_r correlation · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ProQ3D (CAD-trained; rotameric optimization), reported by Pagès et al. on ATOM3D PSR-GLOBAL-RS: Protein structure ranking, global Spearman Trained and scored under the ATOM3D split for this task. Asterisks in the paper mark a run whose training data differed. Aggregation: Not reported ATOM3D: Tasks On Molecules in Three Dimensions · Table 6, row(glob. RS), column([Pagès et al., 2019])
Configuration: ProQ3D (CAD-trained; rotameric optimization), reported by Pagès et al. (cited as [Pagès et al., 2019])	Task: ATOM3D PSR-MEAN-RS: Protein structure ranking, mean Spearman within a target Dataset subset: ATOM3D PSR (ATOM3D split)	0.432 spearman_r correlation · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ProQ3D (CAD-trained; rotameric optimization), reported by Pagès et al. on ATOM3D PSR-MEAN-RS: Protein structure ranking, mean Spearman within a target Trained and scored under the ATOM3D split for this task. Asterisks in the paper mark a run whose training data differed. Aggregation: Not reported ATOM3D: Tasks On Molecules in Three Dimensions · Table 6, row(PSR mean RS), column([Pagès et al., 2019])

Source checking is not independent reproduction. Release 2026-09-29-06401fd5b220.

Use this model

How it works, versions and access

Strengths, limitations and unresolved questions

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.

0 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

No evidence rows match these filters. Choose another scope or clear the search.

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: source checked

2 source records and release history

ATOM3D: Tasks On Molecules in Three Dimensions · Original source · Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Protein model quality assessment using 3D oriented convolutional neural networks (HAL hal-01899468v1) · Original source · hal-01899468v1

Download this release

Technical metadata and extraction receipts

Stable ID: atom3d-method-pag-s-et-al-2019

areas: molecular-interactions
source locator: Table 6, column([Pagès et al., 2019])
missing metadata: checkpoint revision: unreported; parameters: unextracted
source label: [Pagès et al., 2019]
source identity: status: resolved; label form: bracketed_citation; display name: ProQ3D (CAD-trained; rotameric optimization), reported by Pagès et al.; identity: ProQ3D; configuration: ProQ3D version trained on CAD-score with rotameric optimization, run by Pagès et al. with a 2013 UniRef database. ATOM3D reuses the CASP 11 stage 2 scores that Pagès et al. report.; basis: ATOM3D App. F.7 takes its PSR comparison 'as reported in [Pagès et al., 2019]' and names 3DCNN, Ornate and Proq3D as the reported methods. All six ATOM3D Table 8 statistics (mean RP, RS, RK 0.444, 0.432, 0.304; global RP, RS, RK 0.772, 0.796, 0.594) equal the CASP 11 stage 2 Proq3D row of Pagès et al. Table 2; the Ornate and 3dCNN rows differ. The identity rests on ATOM3D's attribution to Pagès et al. together with a match on all six statistics to a single row. One matching value would not be enough.; source ids: evidence-expansion-atom3d-92656c20; source-label-ornate-hal-91150854; source locator: ATOM3D arXiv:2012.04035v4 Table 6, p.8; App. F.7, p.25; Table 8 PSR rows, p.28. Pagès et al., HAL hal-01899468v1 (preprint of Bioinformatics 35(18):3313, doi 10.1093/bioinformatics/btz122): Proq3D setup, PDF p.5; Table 2, CASP 11 stage 2 Proq3D row, PDF p.6.; known details: label: ProQ3D version; value: Trained on CAD-score with rotameric optimization enabled, run with a 2013 UniRef sequence database.; source ids: source-label-ornate-hal-91150854; source locator: Pagès et al. HAL hal-01899468v1, PDF p.5; label: Reported scores; value: Pagès et al. Table 2 correlations are against GDT-TS on CASP 11 stage 2 models.; source ids: source-label-ornate-hal-91150854; source locator: Pagès et al. HAL hal-01899468v1 Table 2, PDF p.6; unknown: label: Checkpoint; note: No ProQ3D software revision or checkpoint is identified.; label: Test set equivalence; note: Neither source establishes that the Pagès et al. CASP 11 stage 2 model set is identical to the ATOM3D PSR test split.; original name: [Pagès et al., 2019]; original description: Published state-of-the-art structure ranking method that the ATOM3D authors compare against, named only by its citation in the table.; review: method: automated_source_review; date: 2026-09-24; note: AI-assisted review against the cited primary sources. No human scientific review. Values, locators and comparison conditions are unchanged.

Related records

Suggest a correction