scTab published annotation evaluation
Source study/challenge grouping of the exact imported protocols. No claim that this intake implements or reproduces an executable benchmark.
Overview
An explanatory profile has not yet been reviewed for this record.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
21 recorded evaluations, 26 metric rows. A comparison chart has not yet been validated for these results. The table retains the individual findings and their sources.
Results
Results are available, but no reviewed comparison panel is linked in this release.
All evaluations
21 evaluations · 26 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: CellTypist — scTab Table 1 | Protocol: scTab Table 1b: seeded fits plus donor bootstrap Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7291 ± 0.0072 Macro F1 fraction · higher Uncertainty: ± 0.0072 (type not stated). source_printed_spread_semantics_conflicting. Heading: confidence intervals; column: standard deviation. Do not interpret as 95% CI. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1b, celltypist row |
| Configuration: Optimized linear logistic regression — scTab Table 1 | Protocol: scTab Table 1b: seeded fits plus donor bootstrap Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7846 ± 0.0072 Macro F1 fraction · higher Uncertainty: ± 0.0072 (type not stated). source_printed_spread_semantics_conflicting. Heading: confidence intervals; column: standard deviation. Do not interpret as 95% CI. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1b, linear row |
| Configuration: MLP — scTab Table 1 | Protocol: scTab Table 1b: seeded fits plus donor bootstrap Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7973 ± 0.0074 Macro F1 fraction · higher Uncertainty: ± 0.0074 (type not stated). source_printed_spread_semantics_conflicting. Heading: confidence intervals; column: standard deviation. Do not interpret as 95% CI. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1b, mlp row |
| Configuration: scTab — scTab Table 1 | Protocol: scTab Table 1b: seeded fits plus donor bootstrap Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.8300 ± 0.0069 Macro F1 fraction · higher Uncertainty: ± 0.0069 (type not stated). source_printed_spread_semantics_conflicting. Heading: confidence intervals; column: standard deviation. Do not interpret as 95% CI. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1b, sctab row |
| Configuration: XGBoost — scTab Table 1 | Protocol: scTab Table 1b: seeded fits plus donor bootstrap Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.8136 ± 0.0060 Macro F1 fraction · higher Uncertainty: ± 0.0060 (type not stated). source_printed_spread_semantics_conflicting. Heading: confidence intervals; column: standard deviation. Do not interpret as 95% CI. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1b, xgboost row |
| Configuration: celltypist default parameters — scTab Table 6 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.6258 ± 0.0036 Macro F1 fraction · higher Uncertainty: SD 0.0036 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 6, celltypist, default parameters |
| Configuration: xgboost default parameters — scTab Table 6 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.5855 ± 0.0112 Macro F1 fraction · higher Uncertainty: SD 0.0112 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 6, xgboost, default parameters |
| Configuration: CellTypist — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~2000 Inference throughput sample-per-second · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, celltypist, inference time |
| Configuration: CellTypist — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~16h Runtime (training) hour · lower Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, celltypist, training time |
| Configuration: Optimized linear logistic regression — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~29500 Inference throughput sample-per-second · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, linear, inference time |
| Configuration: Optimized linear logistic regression — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~20h Runtime (training) hour · lower Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, linear, training time |
| Configuration: MLP — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~21400 Inference throughput sample-per-second · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, mlp, inference time |
| Configuration: MLP — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~29h Runtime (training) hour · lower Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, mlp, training time |
| Configuration: scTab — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~10800 Inference throughput sample-per-second · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, sctab, inference time |
| Configuration: scTab — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~33h Runtime (training) hour · lower Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, sctab, training time |
| Configuration: XGBoost — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~4200 Inference throughput sample-per-second · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, xgboost, inference time |
| Configuration: XGBoost — scTab Table 1 | Protocol: scTab Table 2 hardware-specific resources Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | ~10h Runtime (training) hour · lower Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 2, xgboost, training time |
| Configuration: CellTypist — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7304 ± 0.0015 Macro F1 fraction · higher Uncertainty: SD 0.0015 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, CellTypist row, macro F1 and number-of-runs columns |
| Configuration: CIForm — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.766 Macro F1 fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, CIForm row, macro F1 and number-of-runs columns |
| Configuration: Optimized linear logistic regression — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7848 ± 0.0001 Macro F1 fraction · higher Uncertainty: SD 0.0001 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, Optimized linear logistic regression row, macro F1 and number-of-runs columns |
| Configuration: MLP — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7971 ± 0.0012 Macro F1 fraction · higher Uncertainty: SD 0.0012 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, MLP row, macro F1 and number-of-runs columns |
| Configuration: scGPT fine-tuned — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.749 Macro F1 fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, scGPT fine-tuned row, macro F1 and number-of-runs columns |
| Configuration: scGPT frozen embeddings + logistic regression — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7301 ± 0.0035 Macro F1 fraction · higher Uncertainty: SD 0.0035 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, scGPT frozen embeddings + logistic regression row, macro F1 and number-of-runs columns |
| Configuration: scTab — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.8295 ± 0.0007 Macro F1 fraction · higher Uncertainty: SD 0.0007 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, scTab row, macro F1 and number-of-runs columns |
| Configuration: UCE frozen embeddings + classifier — scTab Table 1 | Protocol: scTab Table 1a: seeded model fits Dataset: scTab processed CELLxGENE 2023-05-15 donor holdouts | 0.7611 ± 0.0018 Macro F1 fraction · higher Uncertainty: SD 0.0018 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported scTab supplementary comparison tables · Supplementary Table 1a, UCE frozen embeddings + classifier row, macro F1 and number-of-runs columns |
Source checking is not independent reproduction. Release 2026-11-21-adc15377fab2.
Further results on linked records
2 linked records have results that are not included above, because no reviewed claim yet ties the link to this benchmark. Each record keeps its own settings and evaluations, and its scores are on its own page.
- scTab deep-ensemble uncertainty ROC-AUC: known-type vs. absent-type (Group 1 vs. Group 3) Protocol · part of this benchmark · 1 evaluation · 1 result
- scTab deep-ensemble uncertainty ROC-AUC: known-type correct vs. incorrect (Group 1 vs. Group 2) Protocol · part of this benchmark · 1 evaluation · 1 result
Methods and evaluation design
Procedure, tasks and evaluated configurations
Evaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
Protocols
These source-backed links do not make different protocols or scores interchangeable.
Baseline coverage
Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.
0 of 10 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.
Baseline status by linked protocol
- scTab Table 1b: seeded fits plus donor bootstrap · 0/2 roles measured
- scTab Table 2 hardware-specific resources · 0/2 roles measured
- scTab Table 1a: seeded model fits · 0/2 roles measured
- scTab deep-ensemble uncertainty ROC-AUC: known-type vs. absent-type (Group 1 vs. Group 3) · 0/2 roles measured
- scTab deep-ensemble uncertainty ROC-AUC: known-type correct vs. incorrect (Group 1 vs. Group 2) · 0/2 roles measured
Protocol coverage CSV (gzip) · Model evaluation matrix (gzip) · Source table (gzip) · Release and checksums (gzip)
Coverage is derived from release 2026-11-21-adc15377fab2. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.
Run this benchmark
Choose a concrete protocol before running an evaluation. Its inputs, split and scoring rules determine which results can be compared.
Run instructions
No runnable recipe has been reviewed for this benchmark. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
0 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
Sources and history
Release 2026-11-21-adc15377fab2 · Record review: source checked
3 source records and release history
- scTab: Scaling cross-tissue single-cell annotation models · Original source · 10.1038/s41467-024-51059-5; published article XML retrieved 2026-09-30
- scTab supplementary comparison tables · Original source · Supplementary Information to 10.1038/s41467-024-51059-5
- scTab official reproducibility repository · Original source · 5ede7f2ba1f9618b86924f2ff587931de18f4ada
Technical metadata and record identifiers
Technical metadata for this record is tracked under stable ID ucc-research-benchmark-sctab.
Source locator: Supplementary Tables 1,2,6
Related records
- part of: scTab Table 1b: seeded fits plus donor bootstrap
- part of: scTab Table 2 hardware-specific resources
- part of: scTab Table 1a: seeded model fits
- part of: scTab deep-ensemble uncertainty ROC-AUC: known-type vs. absent-type (Group 1 vs. Group 3)
- part of: scTab deep-ensemble uncertainty ROC-AUC: known-type correct vs. incorrect (Group 1 vs. Group 2)