Task

GlycanML immunogenicity prediction

This GlycanML task evaluates immunogenicity classification using glycan representations.

Sources (3)

GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)

10 evaluations · 10 metric rows

At a glance

Inputs, training, access and other details

Explanatory profile: source reviewed · 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	Binary glycan immunogenicity annotations. Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)
Splits	Glycans are represented by motif frequencies and clustered; motif groups are allocated to training, validation and test in an 8:1:1 grouping scheme. Table 1 preserves the resulting dataset-specific counts. Sources glycanml primary benchmark evidence · Sections 3.1–3.3; Table 1
Metrics	AUROC and AUPRC in the checked binary-classification configuration. Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)
Baselines	Sequence-model CNN, ResNet, LSTM and BERT configurations; graph-model GCN, RGCN, GAT, GIN, CompGCN and MPNN configurations. Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)
Leakage controls	Motif-based cluster separation tests transfer to structurally different glycans. This is a glycan-structure control, not a claim that all organisms or source studies are held out. Sources glycanml primary benchmark evidence · Sections 3.1–3.3; Table 1
Uncertainty	Every experiment uses seeds 0, 1 and 2; reported summaries are the mean and standard deviation over those three runs. Sources glycanml primary benchmark evidence · Sections 3.1–3.4 and 5.1; Tables 1 and 3
Entity type	Constituent benchmark task: GlycanML immunogenicity prediction Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)
Organisms	Taxonomy tasks explicitly predict organism categories; species scope depends on the constituent dataset. Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)
Assays	Taxonomy, immunogenicity, glycosylation-type and protein–glycan interaction annotations. Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)
Allowed inputs	Glycan sequence or graph representation with an immunogenicity target. Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)
Adaptation	Separate single-task and multi-task training configurations are supplied. Sources (3) GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)

How it works

How it worksEvaluation procedure

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

Sources (4)

GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py; glycanml primary benchmark evidence · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1

Evaluation methodology

Binary glycan immunogenicity annotations. Dataset loader retains the train/validation/test assignments supplied in the downloaded CSV; this inspection does not establish how those original assignments were constructed. AUROC and AUPRC in the checked binary-classification configuration. Sequence-model CNN, ResNet, LSTM and BERT configurations; graph-model GCN, RGCN, GAT, GIN, CompGCN and MPNN configurations.

Sources (3)

Evaluation design

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

Benchmarks

GlycanML

Protocols

SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)

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

Published comparisons

Explore the results reported under one evaluation protocol. Each figure keeps its source, dataset and metric together; it is not a ranking across studies.

Comparison and metric

GlycanML Immuno · Table 3, p. 8

AUPRC (dimensionless) · Higher values are better for this metric.

Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column.

Evaluation protocol · SugarBase immunogenicity · Immuno

Shallow CNN · Configuration · Author-reported evaluation0.776(0.267)
ResNet · Configuration · Author-reported evaluation0.754(0.124)
LSTM · Configuration · Author-reported evaluation0.862(0.016)
Transformer · Configuration · Author-reported evaluation0.729(0.069)
GCN · Configuration · Author-reported evaluation0.688(0.023)
GAT · Configuration · Author-reported evaluation0.685(0.053)
GIN · Configuration · Author-reported evaluation0.716(0.051)
MPNN · Configuration · Author-reported evaluation0.674(0.119)
RGCN · Configuration · Author-reported evaluation0.780(0.006)
CompGCN · Configuration · Author-reported evaluation0.692(0.006)

Source order is preserved. Plotted marks show point estimates; uncertainty, where reported, is retained in the printed values and table. Differences do not establish statistical significance.

GlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, p. 8, Immuno

Values, uncertainty and evidence

AUPRC: original source values
Tested entity	Printed value	Uncertainty	Evidence
Shallow CNN · Configuration	0.776(0.267) dimensionless	type: standard_deviation; value: 0.267; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row Shallow CNN, column Immuno
ResNet · Configuration	0.754(0.124) dimensionless	type: standard_deviation; value: 0.124; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row ResNet, column Immuno
LSTM · Configuration	0.862(0.016) dimensionless	type: standard_deviation; value: 0.016; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row LSTM, column Immuno
Transformer · Configuration	0.729(0.069) dimensionless	type: standard_deviation; value: 0.069; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row Transformer, column Immuno
GCN · Configuration	0.688(0.023) dimensionless	type: standard_deviation; value: 0.023; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row GCN, column Immuno
GAT · Configuration	0.685(0.053) dimensionless	type: standard_deviation; value: 0.053; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row GAT, column Immuno
GIN · Configuration	0.716(0.051) dimensionless	type: standard_deviation; value: 0.051; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row GIN, column Immuno
MPNN · Configuration	0.674(0.119) dimensionless	type: standard_deviation; value: 0.119; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row MPNN, column Immuno
RGCN · Configuration	0.780(0.006) dimensionless	type: standard_deviation; value: 0.006; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row RGCN, column Immuno
CompGCN · Configuration	0.692(0.006) dimensionless	type: standard_deviation; value: 0.006; n: 3; unit: same_as_metric	Author-reported evaluation · source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row CompGCN, column Immuno

Scope and limitations

Taxonomic levels are distinct tasks; do not combine raw accuracy across levels.
Protein–glycan scores evaluate combined glycan encoders plus ESM-1b/MLP, not an ESM-1b standalone checkpoint.
Cluster manifests and trained checkpoint hashes are not enumerated in the paper table.

Source transcription and grouping reviewed by automated source review on 2026-09-17. These experiments were not independently reproduced by rewire.

Tested entities and results

Release 2026-09-17-d277315f7d76 · 10 evaluations · 10 metric rows. Different protocols are not a single leaderboard.

Evaluation setupMetricEvidence origin

Results grouped by the exact reported evaluation
Metric and finding	Coverage and uncertainty	Evidence
LSTM: SugarBase immunogenicity · Immuno Configuration: LSTMProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.862(0.016) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.016; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row LSTM, column Immuno Source checking is not independent reproduction.
ResNet: SugarBase immunogenicity · Immuno Configuration: ResNetProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.754(0.124) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.124; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row ResNet, column Immuno Source checking is not independent reproduction.
RGCN: SugarBase immunogenicity · Immuno Configuration: RGCNProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.780(0.006) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.006; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row RGCN, column Immuno Source checking is not independent reproduction.
MPNN: SugarBase immunogenicity · Immuno Configuration: MPNNProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.674(0.119) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.119; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row MPNN, column Immuno Source checking is not independent reproduction.
CompGCN: SugarBase immunogenicity · Immuno Configuration: CompGCNProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.692(0.006) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.006; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row CompGCN, column Immuno Source checking is not independent reproduction.
Transformer: SugarBase immunogenicity · Immuno Configuration: TransformerProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.729(0.069) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.069; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row Transformer, column Immuno Source checking is not independent reproduction.
GIN: SugarBase immunogenicity · Immuno Configuration: GINProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.716(0.051) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.051; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row GIN, column Immuno Source checking is not independent reproduction.
GCN: SugarBase immunogenicity · Immuno Configuration: GCNProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.688(0.023) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.023; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row GCN, column Immuno Source checking is not independent reproduction.
GAT: SugarBase immunogenicity · Immuno Configuration: GATProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.685(0.053) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.053; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row GAT, column Immuno Source checking is not independent reproduction.
Shallow CNN: SugarBase immunogenicity · Immuno Configuration: Shallow CNNProtocol: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)Dataset: SugarBase immunogenicity · Immuno Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column. Author-reported evaluation · Evaluation metadata: needs review
0.776(0.267) AUPRC Unit: dimensionless · Direction: higher	Uncertainty: type: standard deviation; value: 0.267; n: 3; unit: same as metric Scored: Not reported · Eligible: Not reported	source checkedGlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Table 3, row Shallow CNN, column Immuno Source checking is not independent reproduction.

Papers and result coverage

Last literature check: 2026-09-17. Primary-paper within-study comparison; numerical results not independently reproduced.

Paper or primary resource	Version	Reference
GlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning	arXiv:2405.16206v1, 2024-05-25	Read source

What is still missing

Paper table does not enumerate trained checkpoint hashes; source configuration names retained.

Search and extraction details

complete comparison extracted

Searches

GlycanML benchmark 2405.16206

Evidence locations

Sections 3–5
Tables 1–4

Strengths and limitations

Strengths supported by sources

Shared configurations allow controlled comparisons between single-task and multi-task learning.
Sources (3)
GlycanML/GlycanML official source; GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml; GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods)

Limitations and conditions

Taxonomy, immunogenicity and glycosylation tasks hold out glycan motif clusters; interaction prediction holds out protein clusters. These boundaries do not imply that both proteins and glycans are unseen in the interaction task.
Sources
glycanml primary benchmark evidence · Sections 3.1–3.4 and 5.1; Tables 1 and 3

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-glycanml-immunogenicity-prediction

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.

50 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-17-d277315f7d76
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	GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T20:43:24.817696+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: `73bb768deff2b22b4cf57199353267dc0778b42ccbbaee8035e91ec67b3a3a1f` 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	GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T20:42:46.930612+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: `5c92f79c530629642450ca4536cd864cccc3a626f8ecdfeaaf6fcd2ece356901` 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	glycanml primary benchmark evidence Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 2405.16206v1 Retrieved: 2026-09-16T21:04:56.017006+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: `9ba3db678b4550898a612b42e8832bf9dee40935090fc4d969ba8f6ac5106979` 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	GlycanML/GlycanML official source Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T10:30:24.094166+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: `5237cd1af3f1d7b9cf07cfe3cf722987bb27e10ee356ec108584943446bbb836` Hash scope: Hash scope not separately documented; inspect source record
Diagram steps ["Allowed inputs: Glycan sequence or graph representation with an immunogenicity target.","Splits: Glycans are represented by motif frequencies and clustered; motif groups are allocated to training, validation and test in an 8:1:1 grouping scheme. Table 1 preserves the resulting dataset-specific counts.","Metrics: AUROC and AUPRC in the checked binary-classification configuration."] Individual claims	GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T20:43:24.817696+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: `73bb768deff2b22b4cf57199353267dc0778b42ccbbaee8035e91ec67b3a3a1f` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps ["Allowed inputs: Glycan sequence or graph representation with an immunogenicity target.","Splits: Glycans are represented by motif frequencies and clustered; motif groups are allocated to training, validation and test in an 8:1:1 grouping scheme. Table 1 preserves the resulting dataset-specific counts.","Metrics: AUROC and AUPRC in the checked binary-classification configuration."] Individual claims	GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T20:42:46.930612+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: `5c92f79c530629642450ca4536cd864cccc3a626f8ecdfeaaf6fcd2ece356901` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps ["Allowed inputs: Glycan sequence or graph representation with an immunogenicity target.","Splits: Glycans are represented by motif frequencies and clustered; motif groups are allocated to training, validation and test in an 8:1:1 grouping scheme. Table 1 preserves the resulting dataset-specific counts.","Metrics: AUROC and AUPRC in the checked binary-classification configuration."] Individual claims	glycanml primary benchmark evidence Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 2405.16206v1 Retrieved: 2026-09-16T21:04:56.017006+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: `9ba3db678b4550898a612b42e8832bf9dee40935090fc4d969ba8f6ac5106979` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps ["Allowed inputs: Glycan sequence or graph representation with an immunogenicity target.","Splits: Glycans are represented by motif frequencies and clustered; motif groups are allocated to training, validation and test in an 8:1:1 grouping scheme. Table 1 preserves the resulting dataset-specific counts.","Metrics: AUROC and AUPRC in the checked binary-classification configuration."] Individual claims	GlycanML/GlycanML official source Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T10:30:24.094166+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: `5237cd1af3f1d7b9cf07cfe3cf722987bb27e10ee356ec108584943446bbb836` Hash scope: Hash scope not separately documented; inspect source record
Diagram title Evaluation procedure Individual claims	GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T20:43:24.817696+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: `73bb768deff2b22b4cf57199353267dc0778b42ccbbaee8035e91ec67b3a3a1f` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Evaluation procedure Individual claims	GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml Original source ↗ Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); Sections 3.1–3.3; Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 9f392aa6f9c6d74a296a250199beb347923d04e0 Retrieved: 2026-09-16T20:42:46.930612+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: `5c92f79c530629642450ca4536cd864cccc3a626f8ecdfeaaf6fcd2ece356901` 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

Release 2026-09-17-d277315f7d76 · Record review: discovered

5 source records and release history

GlycanML/GlycanML official source · Original source · 9f392aa6f9c6d74a296a250199beb347923d04e0
GlycanML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning · Original source · arXiv:2405.16206v1, 2024-05-25
GlycanML/GlycanML configs/single_task/BERT/immunogenicity_BERT.yaml · Original source · 9f392aa6f9c6d74a296a250199beb347923d04e0
GlycanML/GlycanML module/custom_datasets/glycan_immunogenicity.py · Original source · 9f392aa6f9c6d74a296a250199beb347923d04e0
glycanml primary benchmark evidence · Original source · 2405.16206v1

Download this release

Technical metadata and extraction receipts

Stable ID: discovery-benchmark-glycanml-immunogenicity-prediction

areas: glycomics
entity level: task
scope note: Specialist molecular or omics evaluation; protocol details require review before numerical comparison.
task: immunogenicity prediction
version: Not reported
comparison panels: id: glycanml-v1-single-immuno; title: GlycanML Immuno · Table 3, p. 8; protocol id: paper-protocol-705c3995af4e74e425; dataset id: paper-dataset-dfa9efa0038fe35cf1; metric: AUPRC; unit: dimensionless; direction: higher; result ids: paper-result-daed94f39b26d6bf9f; paper-result-14cf9ecb36e308a19c; paper-result-0f0c6ac7d2490faded; paper-result-a6f2c25189be55752b; paper-result-c3c413b66dd284a72c; paper-result-cfb7d3096516691659; paper-result-c17bcd75195fbd91e3; paper-result-76c0f9b5fb05a8b66f; paper-result-316c3afd5495e9cc25; paper-result-8afa28bb629f98ba97; source ids: expansion-p3-glycanml-2405-16206v1; source locator: Table 3, p. 8, Immuno; context: Author-reported single-task models on the same held-out task; not a cross-paper ranking. Taxonomy, immunogenicity and glycosylation: motif-frequency K-means cluster allocation 8:1:1. Interaction: MMseqs2 protein clusters at minimum identity 0.5 allocated 8:1:1. Exact counts are attached to each column.; caveats: Taxonomic levels are distinct tasks; do not combine raw accuracy across levels.; Protein–glycan scores evaluate combined glycan encoders plus ESM-1b/MLP, not an ESM-1b standalone checkpoint.; Cluster manifests and trained checkpoint hashes are not enumerated in the paper table.; review: method: automated_source_review; date: 2026-09-17
benchmark research: review date: 2026-09-17; status: complete_comparison_extracted; primary sources: expansion-p3-glycanml-2405-16206v1; inspected locators: Sections 3–5; Tables 1–4; searched queries: GlycanML benchmark 2405.16206; gaps: Paper table does not enumerate trained checkpoint hashes; source configuration names retained.; claim scope: Primary-paper within-study comparison; numerical results not independently reproduced.
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.
legacy kinds: benchmark
entity classification: review date: 2026-09-17; rationale: This record identifies the biological prediction question or a suite-specific task, rather than a uniquely fixed evaluated procedure. Preserve its task identity and leave split, model adaptation and scoring details on linked protocols/evaluations.; source ids: src-discovery-glycanml-glycanml; evidence-benchmark-glycanml-bert-immunogenicity-bert-yaml; evidence-benchmark-glycan-immunogenicity-dataset; source locator: Pinned README: Introduction; Experiment configurations; leaderboard; pinned configs/single_task/BERT/immunogenicity_BERT.yaml, module/custom_datasets/glycan_immunogenicity.py (task metric, dataset class and split methods); ambiguities: A paper- or suite-specific task may constrain some inputs or metrics; that alone does not make it interchangeable with a complete versioned protocol. No protocol equivalence is inferred.

Related records

parent: GlycanML
part of: GlycanML
evaluates task: SugarBase immunogenicity · Immuno (GlycanML immunogenicity prediction)
subject: GlycanML immunogenicity prediction: part of GlycanML

Suggest a correction