Datasets
Twelve scRNA-seq datasets organized into four transfer scenarios and multiple source-to-target tasks.
Task
Cross-dataset single-cell drug response transfer
Cross-dataset response prediction evaluates transfer across batches, drugs, tumors and patients.
Sources
scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captionsOverview
Metrics
Cell-group accuracy averages within-cluster prediction accuracy; a separate malignant-cell-group accuracy is also described.
Allowed inputs
Source and target single-cell expression representations.
Sources
scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captionsEvaluation 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
scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions; Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2Source reviewed · 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
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: scXDR | Task: Cross-dataset single-cell drug response transfer Dataset: scXDR transfer scenario 2 | 0.825 AUC unitless · unknown Uncertainty: ± 0.1573 standard deviation Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcescXDR: Cross-dataset single-cell drug response transfer Single-cell-to-single-cell transfer; source scenario 2. Aggregation: Not reported scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Table 2, scXDR row, Scenario 2 column |
| Configuration: scVI | Task: Cross-dataset single-cell drug response transfer Dataset: scXDR transfer scenario 2 | 0.697 AUC unitless · unknown Uncertainty: ± 0.2463 standard deviation Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourcescVI: Cross-dataset single-cell drug response transfer Single-cell-to-single-cell transfer; source scenario 2. Aggregation: Not reported scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Table 2, scVI row, Scenario 2 column |
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
Twelve scRNA-seq datasets organized into four transfer scenarios and multiple source-to-target tasks. The paper defines 20 cross-dataset transfers across batch, drug, tumor and patient scenarios. Comparator protocols differ: scVI/ComBat combine corrected datasets before a train/test partition, whereas transfer methods retain source/target roles. The exact per-task split must accompany any comparison. Cell-group accuracy averages within-cluster prediction accuracy; a separate malignant-cell-group accuracy is also described. Bulk-transfer scDEAL, SCAD, CaDRReS-Sc and DREEP plus conventional MLP/SVM/correlation references; single-cell transfer comparators form another comparison group. Training and evaluation transfer between datasets, drugs, tumors or patients as listed in Supplementary Table S2. Table S11 excludes drug–cell relations from the representation graph’s listed relation types. These controls do not by themselves establish an audit of target-label use during every adaptation or hyperparameter-selection step.
Sources (2)
scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning; scxdr-2026__42003_2025_9418_MOESM1_ESM.pdf · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions; Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2; Supplementary Tables S2 and S11Recorded 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 supported by sources
No source-reviewed explanatory claims are recorded here yet.
Limitations and conditions
- Cell-level, cluster-averaged and malignant-group metrics are distinct aggregates. Transfer between datasets does not by itself establish that every comparator used identical target-label access or hyperparameter selection.
Sources (11)
scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning; scxdr-2026__42003_2025_9418_MOESM1_ESM.pdf; GSE134839.soft; GSE149214.soft; GSE108394.soft; GSE164614.soft; GSE230538.soft; GSE117872.soft; GSE127298.soft; GSE140440.soft; GSE147326.soft · Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2; Supplementary Tables S2 and S11; Supplementary Table S1; GEO Series_sample_organism and Series_sample_taxid fields
Profile review details
Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.
Stable record: reported-task-167f08013c270eSpecifications
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.
| Property | Description and evidence |
|---|---|
| Datasets | Twelve scRNA-seq datasets organized into four transfer scenarios and multiple source-to-target tasks.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
| Splits | The paper defines 20 cross-dataset transfers across batch, drug, tumor and patient scenarios. Comparator protocols differ: scVI/ComBat combine corrected datasets before a train/test partition, whereas transfer methods retain source/target roles. The exact per-task split must accompany any comparison.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2 |
| Metrics | Cell-group accuracy averages within-cluster prediction accuracy; a separate malignant-cell-group accuracy is also described.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
| Baselines | Bulk-transfer scDEAL, SCAD, CaDRReS-Sc and DREEP plus conventional MLP/SVM/correlation references; single-cell transfer comparators form another comparison group.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
| Leakage controls | Training and evaluation transfer between datasets, drugs, tumors or patients as listed in Supplementary Table S2. Table S11 excludes drug–cell relations from the representation graph’s listed relation types. These controls do not by themselves establish an audit of target-label use during every adaptation or hyperparameter-selection step.Sources (2)scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning; scxdr-2026__42003_2025_9418_MOESM1_ESM.pdf · Supplementary Tables S2 and S11 |
| Uncertainty | Transfer-scenario comparisons report mean and standard deviation; variation across source/target tasks must not be interpreted as an individual-cell confidence interval.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
| Entity type | Paper-specific computational evaluation protocol.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
| Organisms | Human (Homo sapiens). The nine GEO accessions listed in Supplementary Table S1, including the repeated patient/drug subsets of GSE147326, all identify their sample organism as Homo sapiens (taxon 9606).Sources (10)scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning; GSE134839.soft; GSE149214.soft; GSE108394.soft; GSE164614.soft; GSE230538.soft; GSE117872.soft; GSE127298.soft; GSE140440.soft; GSE147326.soft · Supplementary Table S1; GEO Series_sample_organism and Series_sample_taxid fields |
| Assays | Single-cell drug-response measurements.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
| Allowed inputs | Source and target single-cell expression representations.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
| Adaptation | Cross-dataset transfer; exact target-label access must be distinguished by transfer scenario.SourcesscXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions |
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-paper discovery and source inspection. Source-checked results are not independently reproduced experiments.
| Paper or primary resource | Version | Reference |
|---|---|---|
| scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning | PMC archival version PMC12859067.1 | Read source DOI: 10.1038/s42003-025-09418-5 |
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.
- Keep bulk and single-cell settings explicit; do not call this feature-unseen inductive transfer.
- Standard deviations are printed; resolve exact aggregation population from scenario-level source data before attaching seed-count error bars.
Search and extraction details
primary comparison table screened
Searches
- "scXDR" cross dataset drug response
Evidence locations
- Tables 1–2
- Discussion: cross-dataset scenarios and target-label access
- 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.
38 evidence rows matching the loaded filters
| 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 | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions; Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2 Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Diagram steps
| scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions; Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2 Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Computational evaluation flow Individual claims | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions; Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2 Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Datasets Twelve scRNA-seq datasets organized into four transfer scenarios and multiple source-to-target tasks. Individual claims | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Splits The paper defines 20 cross-dataset transfers across batch, drug, tumor and patient scenarios. Comparator protocols differ: scVI/ComBat combine corrected datasets before a train/test partition, whereas transfer methods retain source/target roles. The exact per-task split must accompany any comparison. Individual claims | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; Comparison experiments at the individual cell level; Supplementary Table S2 Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Adaptation Cross-dataset transfer; exact target-label access must be distinguished by transfer scenario. Individual claims | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Metrics Cell-group accuracy averages within-cluster prediction accuracy; a separate malignant-cell-group accuracy is also described. Individual claims | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Baselines Bulk-transfer scDEAL, SCAD, CaDRReS-Sc and DREEP plus conventional MLP/SVM/correlation references; single-cell transfer comparators form another comparison group. Individual claims | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Leakage controls Training and evaluation transfer between datasets, drugs, tumors or patients as listed in Supplementary Table S2. Table S11 excludes drug–cell relations from the representation graph’s listed relation types. These controls do not by themselves establish an audit of target-label use during every adaptation or hyperparameter-selection step. Individual claims | scxdr-2026__42003_2025_9418_MOESM1_ESM.pdf Supplementary Tables S2 and S11 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved 2026-09-16; sha256:ad53da81235ba47f762c93ac5140108241b857160d978c63ff934e8e57283758 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record Archive member: 42003_2025_9418_MOESM1_ESM.pdf |
| Leakage controls Training and evaluation transfer between datasets, drugs, tumors or patients as listed in Supplementary Table S2. Table S11 excludes drug–cell relations from the representation graph’s listed relation types. These controls do not by themselves establish an audit of target-label use during every adaptation or hyperparameter-selection step. Individual claims | scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning Supplementary Tables S2 and S11 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: PMC archival version PMC12859067.1 | source checked automated source review · 2026-09-16 Audit detailsTargeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: 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 2026-09-29-06401fd5b220 · Record review: needs review
12 source records and release history
- scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Original source · PMC archival version PMC12859067.1
- scXDR: drug response prediction across single-cell datasets via heterogeneous network transfer learning · Original source · PMC archival version PMC12859067.1
- scxdr-2026__42003_2025_9418_MOESM1_ESM.pdf · Original source · Retrieved 2026-09-16; sha256:ad53da81235ba47f762c93ac5140108241b857160d978c63ff934e8e57283758
- GSE134839.soft · Original source · Retrieved 2026-09-16; sha256:333ba9b2538c6ba1f41ae08d60b41099f26a443d2b473cbd10104d887ef20064
- GSE149214.soft · Original source · Retrieved 2026-09-16; sha256:801532dfc1983e4de5b4183d39e81f74825fa0b22632bb79322524d3f6139d1e
- GSE108394.soft · Original source · Retrieved 2026-09-16; sha256:e4b44c1b50d2a0e7a764630ea901d7848c72dc55cc66e9faf8b706154719fc17
- GSE164614.soft · Original source · Retrieved 2026-09-16; sha256:132e5ca2603b5e5d91c16e708f8928245086d43d13f8f56ec5512373c4248e2c
- GSE230538.soft · Original source · Retrieved 2026-09-16; sha256:63d3b23a3c1266d5cdbd04ca1f4766774ab8427e70b41dbc80e230275286cf0d
- GSE117872.soft · Original source · Retrieved 2026-09-16; sha256:f615587bbf6e89a29811400538966cdf008dc4a7719091fc44448ba4a4aa8f5c
- GSE127298.soft · Original source · Retrieved 2026-09-16; sha256:40a50eb13981d4d9ca3dfcd987c6b4b90289385e6aaeea645c5672a2dc6a9f19
- GSE140440.soft · Original source · Retrieved 2026-09-16; sha256:7b6ca08ec013e2e79b367727f6ce4ff46dfb7ef9aa166bc4907dd02fbbb472e5
- GSE147326.soft · Original source · Retrieved 2026-09-16; sha256:6882efe837a56064645705857c81f21e6f82b773a8a7fd9679180f71bba9bb74
Technical metadata and extraction receipts
Stable ID: reported-task-167f08013c270e
- areas
- cells-tissues
- tasks
- Cross-dataset single-cell drug response transfer
- entity level
- task
- version
- Not reported
- task
- Cross-dataset single-cell drug response transfer
- scope note
- Paper-specific evaluation task; protocol completeness requires further extraction.
- benchmark research
- review date: 2026-09-17; status: primary_comparison_table_screened; primary sources: expansion-p3-scxdr-2026; inspected locators: Tables 1–2; Discussion: cross-dataset scenarios and target-label access; Methods; searched queries: "scXDR" cross dataset drug response; gaps: Keep bulk and single-cell settings explicit; do not call this feature-unseen inductive transfer.; Standard deviations are printed; resolve exact aggregation population from scenario-level source data before attaching seed-count error bars.; claim scope: Primary-paper discovery and source inspection. Source-checked results are not independently reproduced experiments.
- 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: scxdr-2026; source locator: Methods: Data collection and processing; group-level evaluation; cached text lines 55–57, 84–86; uncertainty/repeat-run/statistical-comparison passages; matching task comparison table/ablation captions; 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.