Task

Gene-regulatory link prediction

Gene-regulatory link prediction uses cell-specific reference networks to evaluate inferred regulatory edges.

Sources

Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph

2 evaluations · 2 results

Overview

Datasets

Seven BEELINE human/mouse scRNA-seq datasets with cell-type-specific ChIP-seq reference networks.

Metrics

AUROC and AUPRC; selected comparisons average results from two gene-panel sizes.

Allowed inputs

Single-cell expression and candidate regulatory gene pairs.

Sources

Evaluation procedure diagram

How it worksComputational evaluation flow

Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol.

Sources

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

Results

Results are available, but no reviewed comparison panel is linked in this release.

All evaluations

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
Pipeline: scRegNet (Geneformer backbone)	Task: Gene-regulatory link prediction Dataset: hESC cell-type-specific GRN	0.89 AUROC unitless · unknown Uncertainty: ± 0.00 as printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scRegNet (Geneformer backbone): Gene-regulatory link prediction TFs plus 500 variable genes; mean from 50 independent evaluations. Aggregation: Not reported Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Table 2, scRegNet (w/ Geneformer) row, hESC AUROC entry
Pipeline: scRegNet (scBERT backbone)	Task: Gene-regulatory link prediction Dataset: hESC cell-type-specific GRN	0.88 AUROC unitless · unknown Uncertainty: ± 0.00 as printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scRegNet (scBERT backbone): Gene-regulatory link prediction TFs plus 500 variable genes; mean from 50 independent evaluations. Aggregation: Not reported Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Table 2, scRegNet (w/ scBERT) row, hESC AUROC entry

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

Methods and evaluation design

Procedure, tasks and evaluated configurations

How it works

Evaluation methodology

Seven BEELINE human/mouse scRNA-seq datasets with cell-type-specific ChIP-seq reference networks. For each TF, target/non-target edges are divided 67:33 into training/test, with part of training reserved for validation; all TFs contribute to both arms. AUROC and AUPRC; selected comparisons average results from two gene-panel sizes. GNNLink, GENELink, GNE, CNNC, DeepDRIM, GRN-transformer, PCC, GRNBoost2 and GENIE3. Target genes are separated between training and testing for the same TF. This is not a leave-TF-out design and does not establish globally disjoint gene identities across all TFs. The cited text-accessible evaluation sections give no confidence-interval, resampling or repeat-run error-bar specification. Image-only tables and uninspected supplements are outside this absence claim.

Sources

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Run instructions

No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths and considerations

No source-reviewed explanatory claims are recorded here yet.

Limitations and conditions

Gene-panel definition and incomplete reference networks affect the candidate edge population. Negative-pair construction and cross-TF gene sharing must be distinguished from the stated per-TF target split.
Sources
Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph

Profile review details

Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced.

Stable record: reported-task-3063ed4da76b4b

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	Seven BEELINE human/mouse scRNA-seq datasets with cell-type-specific ChIP-seq reference networks. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Splits	For each TF, target/non-target edges are divided 67:33 into training/test, with part of training reserved for validation; all TFs contribute to both arms. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Metrics	AUROC and AUPRC; selected comparisons average results from two gene-panel sizes. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Baselines	GNNLink, GENELink, GNE, CNNC, DeepDRIM, GRN-transformer, PCC, GRNBoost2 and GENIE3. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Leakage controls	Target genes are separated between training and testing for the same TF. This is not a leave-TF-out design and does not establish globally disjoint gene identities across all TFs. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Uncertainty	The cited text-accessible evaluation sections give no confidence-interval, resampling or repeat-run error-bar specification. Image-only tables and uninspected supplements are outside this absence claim. · Not reported in inspected sources Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Entity type	Paper-specific computational evaluation protocol. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Organisms	Human and mouse. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Assays	Single-cell RNA-seq with cell-type-specific ChIP-seq regulatory references. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Allowed inputs	Single-cell expression and candidate regulatory gene pairs. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph
Adaptation	Supervised regulatory-link prediction compared with neural and conventional network-inference methods. Sources Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning · Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph

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. Primary-source discovery and table/protocol screening; source checked is not independently reproduced. Raw acquisitions not automatically numerical publication approval.

Paper or primary resource	Version	Reference
Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning	PMC11838224.2	Read source DOI: 10.1101/2024.12.16.628715

Historical gaps recorded on 2026-09-17

The catalogue now holds 2 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.

Raw XML stacked AUROC/AUPRC cells preserved with explicit linebreak delimiters. Two GENELink settings must not be conflated. Structured table extraction and exact uncertainty scope remain pending.

Search and extraction details

source found structured extraction pending

Searches

Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning primary paper benchmark results

Evidence locations

Tables2–3,500/1000genes and seven cell types; Methods

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.

17 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 summary of the cited evaluation; exact task configuration and source version remain part of the protocol. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps Input: Single-cell expression and candidate regulatory gene pairs. Evaluation: Supervised regulatory-link prediction compared with neural and conventional network-inference methods. Readout: AUROC and AUPRC; selected comparisons average results from two gene-panel sizes. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Computational evaluation flow Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Datasets Seven BEELINE human/mouse scRNA-seq datasets with cell-type-specific ChIP-seq reference networks. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.facts.0.value` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Splits For each TF, target/non-target edges are divided 67:33 into training/test, with part of training reserved for validation; all TFs contribute to both arms. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.facts.1.value` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Adaptation Supervised regulatory-link prediction compared with neural and conventional network-inference methods. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.facts.10.value` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Metrics AUROC and AUPRC; selected comparisons average results from two gene-panel sizes. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.facts.2.value` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Baselines GNNLink, GENELink, GNE, CNNC, DeepDRIM, GRN-transformer, PCC, GRNBoost2 and GENIE3. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.facts.3.value` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Leakage controls Target genes are separated between training and testing for the same TF. This is not a leave-TF-out design and does not establish globally disjoint gene identities across all TFs. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	source checked automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.facts.4.value` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Uncertainty The cited text-accessible evaluation sections give no confidence-interval, resampling or repeat-run error-bar specification. Image-only tables and uninspected supplements are outside this absence claim. Individual claims	Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning Original source ↗ Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph Version: PMC11838224.2 Retrieved: 2026-09-16T10:41:16.541287+00:00	unreported automated source review · 2026-09-16 Audit details Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: `attributes.profile.facts.5.value` Source artifact SHA-256: `65b3272d47bb9c4ee1e7a965169bef63add9dbeb31508d4076e5145b761af4ec` 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: needs review

2 source records and release history

Download this release

Technical metadata and extraction receipts

Stable ID: reported-task-3063ed4da76b4b

areas: cells-tissues
tasks: Gene-regulatory link prediction
entity level: task
version: Not reported
task: Gene-regulatory link prediction
scope note: Paper-specific evaluation task; protocol completeness requires further extraction.
benchmark research: review date: 2026-09-17; status: source_found_structured_extraction_pending; primary sources: evidence-expansion-p2-scregnet-2025-65b3272d47bb; inspected locators: Tables2–3,500/1000genes and seven cell types; Methods; searched queries: Prediction of Gene Regulatory Connections with Joint Single-Cell Foundation Models and Graph-Based Learning primary paper benchmark results; gaps: Raw XML stacked AUROC/AUPRC cells preserved with explicit linebreak delimiters. Two GENELink settings must not be conflated. Structured table extraction and exact uncertainty scope remain pending.; claim scope: Primary-source discovery and table/protocol screening; source checked is not independently reproduced. Raw acquisitions not automatically numerical publication approval.
historical missing metadata: protocol version: not_reported_in_legacy_extract; split: not_reported_in_legacy_extract
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 source-scoped record identifies the biological prediction task and holds its paper context. Preserve the existing task identity; exact split, model adaptation and scoring remain in linked evaluations or separate protocol records.; source ids: scregnet-2025; source locator: Methods: Datasets and data pre-processing; Baseline models and evaluation metrics; Results: ablations; cached text lines 46–48, 55–56, 65; benchmark datasets and per-TF partition paragraph; 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.; Some legacy profile Entity type facts use the generic phrase computational evaluation protocol. That boilerplate is not sufficient to establish a single fixed protocol identity or to merge this task with another protocol record.

Related records

dataset: hESC cell-type-specific GRN
benchmark: scRegNet (Geneformer backbone): Gene-regulatory link prediction
benchmark: scRegNet (scBERT backbone): Gene-regulatory link prediction

Suggest a correction