GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
GlycanGT study: kingdom Accuracy. Scored with Accuracy on SugarBase taxonomy kingdom; GlycanML official motif split. Taxonomy: 13,209 glycans total across eight levels, 4–1,737 classes per level. Official fixed GlycanML motif-based train/validation/test splits (8:1:1). Section 2.5 reports class-balanced classifiers, train ∪ validation hyperparameter selection by randomized search with 3-fold cross-validation, followed by one evaluation on the held-out test set; complete procedure repeated with three random seeds, reporting mean and standard deviation. GlycanGT large model pretrained with 35% masking provides [Graph] embeddings to SVM/LightGBM; the selected classifier for each S4 row is not identified. Section 2.6 states that all four graph baselines were trained and evaluated on the same datasets/splits; it does not establish that each baseline used the GlycanGT downstream classifier search. SVM search: 10 iterations, RBF/linear, C logU(1e-3,1e2), gamma logU(1e-4,1e-1); LightGBM: 15 randomized iterations. Do not combine with separate original GlycanML-paper protocols.
Overview
GlycanGT study: kingdom Accuracy. Scored with Accuracy on SugarBase taxonomy kingdom; GlycanML official motif split. Taxonomy: 13,209 glycans total across eight levels, 4–1,737 classes per level. Official fixed GlycanML motif-based train/validation/test splits (8:1:1). Section 2.5 reports class-balanced classifiers, train ∪ validation hyperparameter selection by randomized search with 3-fold cross-validation, followed by one evaluation on the held-out test set; complete procedure repeated with three random seeds, reporting mean and standard deviation. GlycanGT large model pretrained with 35% masking provides [Graph] embeddings to SVM/LightGBM; the selected classifier for each S4 row is not identified. Section 2.6 states that all four graph baselines were trained and evaluated on the same datasets/splits; it does not establish that each baseline used the GlycanGT downstream classifier search. SVM search: 10 iterations, RBF/linear, C logU(1e-3,1e2), gamma logU(1e-4,1e-1); LightGBM: 15 randomized iterations. Do not combine with separate original GlycanML-paper protocols.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Results
Each comparison retains its reviewed evaluation scope, dataset and metric. Results are shown without a pooled ranking.
GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
accuracy (fraction) · Higher values are better.
Every method GlycanML reports on GlycanGT study: kingdom Accuracy, scored with Accuracy on SugarBase taxonomy kingdom; GlycanML official motif split.
GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy · SugarBase taxonomy kingdom; GlycanML official motif split (GlycanML split)
Evidence origin: Author-reported evaluation. Numerical source review does not establish independent reproduction.
GlycanGT published supplementary archive, Table S4; glycangt: Journal full-text XML · GlycanGT primary article Sections 2.1, 2.5, 2.6 and 3.1–3.2 (PMC13105845), Supplementary Table S4; columns A (task), C (metric), D (mean), E (SD)Source-checked values from the GlycanGT authors, not independent reproduction. All 103 reported mean/SD pairs and all five methods retained.
All comparison limitations (9)
- Source-checked values from the GlycanGT authors, not independent reproduction. All 103 reported mean/SD pairs and all five methods retained.
- Accuracy remains on its original 0–1 scale. No cross-task or cross-metric aggregate is inferred.
- Immunogenicity Macro-F1 is reported only for RGCN, SweetNet and GlycanGT; GlycanAA and Graphormer missing rows remain absent.
- Source-reported zeros, including RGCN order accuracy and Macro-F1, are preserved as reported; they were not independently rerun.
- GlycanGT per-task downstream classifier choice is not stated; SVM/LightGBM procedures are described collectively.
- Article states a motif 8:1:1 split but does not give exact held-out counts/manifests in Table S4. Total dataset sizes are not test counts.
- The published methods select masking ratio using downstream scores; the exact separation used for that selection is unreported here. Do not claim a fully independently untouched model-selection test set.
- Separate study/protocol from the original GlycanML paper: shared benchmark family does not establish interchangeable scores.
- The current outer archive hash differs from the previous September 16 snapshot. A new source ID pins the retrieved bytes; no prior source is overwritten.
Automated source review: 2026-09-23.
No unavailable values; missing scores remain labelled and are never plotted as zero.
Showing 5 of 5 matching rows.
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.
Methods and evaluation design
Procedure, tasks and evaluated configurations
Evaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
Benchmarks
These source-backed links do not make different protocols or scores interchangeable.
Recorded evaluations
Each evaluation records what was tested and under which conditions.
- GlycanAA on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- GlycanGT on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- Graphormer on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- RGCN on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- SweetNet on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
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 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.
No execution recipe linked to this protocol. Recipe availability does not establish a completed evaluation.
- Author-reported evaluations
- 5
Literature evidence is not a Rewire measurement. Executed but unpublished runs and private review status are not included.
Null control
Proposed control: requires review
Training-set class prior where supervised fitting is permitted
Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.
This is a suggested selection rule, not a validated method or a measured score.
Conventional reference
Proposed control: requires review
Regularised classifier on simple permitted features, or protocol's conventional reference
Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.
This is a suggested selection rule, not a validated method or a measured score.
Protocol coverage CSV · Model evaluation matrix · Source table · Release and checksums
Coverage is derived from release 2026-09-23-2b89723c6dd9. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.
Run instructions
No runnable recipe has been reviewed for this protocol. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.
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.
2 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: part of discovery-benchmark-glycanml Individual claims | glycangt: Journal full-text XML GlycanGT primary article Sections 2.1, 2.5, 2.6 and 3.1–3.2 (PMC13105845), Supplementary Table S4; columns A (task), C (metric), D (mean), E (SD) Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Primary article XML snapshot | source checked automated source review · 2026-09-23 Audit detailsPrimary-source transcription with no human sign-off and no independent reproduction. Field: Claim: glycangt-2026-table-s4-association-kingdom-accuracy Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Relationship: part of discovery-benchmark-glycanml Individual claims | GlycanGT published supplementary archive, Table S4 GlycanGT primary article Sections 2.1, 2.5, 2.6 and 3.1–3.2 (PMC13105845), Supplementary Table S4; columns A (task), C (metric), D (mean), E (SD) Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Published Bioinformatics btag147 supplementary archive, retrieved 2026-09-23; Table_S4.xlsx SHA-256 d7c35909bac6bcb78ca8fdb32c0463f05e692f15861a8184e65e415e7216f493 | source checked automated source review · 2026-09-23 Audit detailsPrimary-source transcription with no human sign-off and no independent reproduction. Field: Claim: glycangt-2026-table-s4-association-kingdom-accuracy Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
View linked audit checks and correction history
Release 2026-09-23-2b89723c6dd9 · Record review: source checked
2 source records and release history
- GlycanGT published supplementary archive, Table S4 · Original source · Published Bioinformatics btag147 supplementary archive, retrieved 2026-09-23; Table_S4.xlsx SHA-256 d7c35909bac6bcb78ca8fdb32c0463f05e692f15861a8184e65e415e7216f493
- glycangt: Journal full-text XML · Original source · Primary article XML snapshot
Technical metadata and extraction receipts
Stable ID: glycangt-2026-table-s4-task-kingdom-accuracy
- areas
- glycans
- tasks
- GlycanGT study: kingdom Accuracy
- metric
- Accuracy
- metric direction
- higher
- dataset
- SugarBase taxonomy kingdom; GlycanML official motif split
- protocol
- Taxonomy: 13,209 glycans total across eight levels, 4–1,737 classes per level. Official fixed GlycanML motif-based train/validation/test splits (8:1:1). Section 2.5 reports class-balanced classifiers, train ∪ validation hyperparameter selection by randomized search with 3-fold cross-validation, followed by one evaluation on the held-out test set; complete procedure repeated with three random seeds, reporting mean and standard deviation. GlycanGT large model pretrained with 35% masking provides [Graph] embeddings to SVM/LightGBM; the selected classifier for each S4 row is not identified. Section 2.6 states that all four graph baselines were trained and evaluated on the same datasets/splits; it does not establish that each baseline used the GlycanGT downstream classifier search. SVM search: 10 iterations, RBF/linear, C logU(1e-3,1e2), gamma logU(1e-4,1e-1); LightGBM: 15 randomized iterations. Do not combine with separate original GlycanML-paper protocols.
- source locator
- GlycanGT primary article Sections 2.1, 2.5, 2.6 and 3.1–3.2 (PMC13105845), Supplementary Table S4; columns A (task), C (metric), D (mean), E (SD)
- comparison panels
- id: glycangt-2026-table-s4-panel-kingdom-accuracy; title: GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy; protocol id: glycangt-2026-table-s4-task-kingdom-accuracy; dataset id: glycangt-2026-table-s4-dataset-sugarbase-taxonomy-kingdom-glycanml-official-motif-split; metric: accuracy; unit: fraction; direction: higher; result ids: glycangt-2026-table-s4-result-rgcn-kingdom-accuracy-accuracy; glycangt-2026-table-s4-result-glycanaa-kingdom-accuracy-accuracy; glycangt-2026-table-s4-result-sweetnet-kingdom-accuracy-accuracy; glycangt-2026-table-s4-result-glycangt-kingdom-accuracy-accuracy; glycangt-2026-table-s4-result-graphormer-kingdom-accuracy-accuracy; source ids: model-coverage-glycangt-table-s4-20260923-source; evidence-official-87c70bddba97a72a5e5d; source locator: GlycanGT primary article Sections 2.1, 2.5, 2.6 and 3.1–3.2 (PMC13105845), Supplementary Table S4; columns A (task), C (metric), D (mean), E (SD); context: Every method GlycanML reports on GlycanGT study: kingdom Accuracy, scored with Accuracy on SugarBase taxonomy kingdom; GlycanML official motif split.; caveats: Source-checked values from the GlycanGT authors, not independent reproduction. All 103 reported mean/SD pairs and all five methods retained.; Accuracy remains on its original 0–1 scale. No cross-task or cross-metric aggregate is inferred.; Immunogenicity Macro-F1 is reported only for RGCN, SweetNet and GlycanGT; GlycanAA and Graphormer missing rows remain absent.; Source-reported zeros, including RGCN order accuracy and Macro-F1, are preserved as reported; they were not independently rerun.; GlycanGT per-task downstream classifier choice is not stated; SVM/LightGBM procedures are described collectively.; Article states a motif 8:1:1 split but does not give exact held-out counts/manifests in Table S4. Total dataset sizes are not test counts.; The published methods select masking ratio using downstream scores; the exact separation used for that selection is unreported here. Do not claim a fully independently untouched model-selection test set.; Separate study/protocol from the original GlycanML paper: shared benchmark family does not establish interchangeable scores.; The current outer archive hash differs from the previous September 16 snapshot. A new source ID pins the retrieved bytes; no prior source is overwritten.; review: method: automated_source_review; date: 2026-09-23
- entity level
- protocol
Related records
- part of: GlycanML
- subject: GlycanML kingdom Accuracy: part of discovery-benchmark-glycanml
- benchmark: GlycanAA on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- benchmark: GlycanGT on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- benchmark: Graphormer on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- benchmark: RGCN on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy
- benchmark: SweetNet on GlycanML kingdom Accuracy: GlycanGT study: kingdom Accuracy