Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B)

Overview

Genuine INTER-dataset (cross-dataset, cross-species where applicable) evaluation: three brain datasets — VISp (mouse, primary visual cortex), ALM (mouse, anterior lateral motor area), MTG (human, middle temporal gyrus) — all SMART-Seq v4. VISp and ALM share GSE115746. MTG is a separate human single-nucleus dataset; combinations involving MTG include cross-species data. Some two-dataset training combinations still share the mouse study with the test dataset, so independent-study holdout must not be inferred for all nine combinations. Methods, quoted verbatim: "We used the three brain datasets, VISp, ALM, and MTG with two levels of annotations, 3 and 34 cell populations. We tested all possible train-test combinations, by either using one dataset to train and test on another (6 experiments) or using two concatenated datasets to train and test on the third (3 experiments). A total of 9 experiments[...]." This protocol scopes the deeper, 34-cell-population annotation level (Figure S10 Panel B); the separate 3-population major-lineage level (Figure S10 Panel A) is a different protocol and is not ingested here. Percentage of cells left "unlabeled" (rejected) is reported per classifier per train-test combination; this protocol ingests only the SVMrejection row (9 values), a conventional rejection-option baseline, as a bounded, coherent subset — not the other 17 classifiers shown in the same figure.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

9 recorded evaluations, 9 metric rows. A comparison chart has not yet been validated for these results. The table retains the individual findings and their sources.

View coverage and remaining gaps across all benchmarks

Results

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

All evaluations

9 evaluations · 9 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
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: ALM brain single-cell RNA-seq (scRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	84.6% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=ALM train=MTG Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set ALM, training set(s) MTG
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: ALM brain single-cell RNA-seq (scRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	22.2% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=ALM train=VISp & MTG (concatenated) Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set ALM, training set(s) VISp & MTG (concatenated)
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: ALM brain single-cell RNA-seq (scRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	16.4% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=ALM train=VISp Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set ALM, training set(s) VISp
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: MTG brain single-nucleus RNA-seq (snRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	99.6% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=MTG train=ALM Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set MTG, training set(s) ALM
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: MTG brain single-nucleus RNA-seq (snRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	99.5% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=MTG train=VISp & ALM (concatenated) Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set MTG, training set(s) VISp & ALM (concatenated)
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: MTG brain single-nucleus RNA-seq (snRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	99% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=MTG train=VISp Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set MTG, training set(s) VISp
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: VISp brain single-cell RNA-seq (scRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	22.1% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=VISp train=ALM & MTG (concatenated) Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set VISp, training set(s) ALM & MTG (concatenated)
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: VISp brain single-cell RNA-seq (scRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	14.8% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=VISp train=ALM Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set VISp, training set(s) ALM
Configuration: SVMrejection — Abdelaal et al. 2019 Table 1	Protocol: Abdelaal et al. 2019 inter-dataset brain comparison: VISp/ALM/MTG, 34-population annotation (Figure S10 Panel B) Dataset: VISp brain single-cell RNA-seq (scRNA-seq) dataset (Abdelaal et al. 2019, Table 2)	81.1% pct-unlabeled percent · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source abdelaal brain svmrejection test=VISp train=MTG Not reported Aggregation: Not reported A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Figure S10, Panel B (page 15 of 18), row "SVMrejection", test set VISp, training set(s) MTG

Source checking is not independent reproduction. Release 2026-10-07-e11db1d1c586.

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Baseline coverage

Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.

0 of 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.

No execution recipe linked to this protocol. Recipe availability does not establish a completed evaluation.

Author-reported evaluations: 9

Literature evidence is not a Rewire measurement. Executed but unpublished runs and private review status are not included.

Null control

Proposed control: requires review

Training-set class prior where supervised fitting is permitted

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Conventional reference

Proposed control: requires review

Regularised classifier on simple permitted features, or protocol's conventional reference

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Protocol coverage CSV · Model evaluation matrix · Source table · Release and checksums

Coverage is derived from release 2026-10-07-e11db1d1c586. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run instructions

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

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

Claims, original sources and review scope · Release 2026-10-07-e11db1d1c586
Property and statement	Original source and location	Review and provenance

No evidence rows match these filters. Choose another scope or clear the search.

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

Release 2026-10-07-e11db1d1c586 · Record review: source checked

2 source records and release history

A comparison of automatic cell identification methods for single-cell RNA sequencing data · Original source · 10.1186/s13059-019-1795-z; published article XML
A comparison of automatic cell identification methods for single-cell RNA sequencing data — Additional File 1 (Supplementary Data) · Original source · 10.1186/s13059-019-1795-z; Additional File 1 (Supplementary Data PDF)

Download this release

Technical metadata and extraction receipts

Stable ID: ucc-research-protocol-abdelaal-brain-inter-34pop

review: method: automated_source_review; actor: Claude Sonnet cell-type-annotation-transfer evidence-research worker; reviewed at: 2026-10-07T07:44:08Z; note: Source-backed transcription of Abdelaal et al. 2019 Additional File 1 (Supplementary Data PDF), a separate byte artifact from the already-catalogued main-text XML. Originally retrieved 2026-10-06T21:17:27Z via the Springer/BMC static-content route (https://static-content.springer.com/esm/art%3A10.1186%2Fs13059-019-1795-z/MediaObjects/13059_2019_1795_MOESM1_ESM.pdf), HTTP 200, 13,495,279 bytes. Re-verified byte-identical (SHA-256 unchanged) in this pass against the cached copy in the neighboring cell-type-evidence-20261006 worktree before archiving. Figure S10 (page 15 of 18) read directly from a pdftoppm image render at 2400 dpi (not OCR, not color inference); Panel B's printed numeric cell values independently re-confirmed against the original PDF by a separate reviewer prior to this ingestion. No new model execution, independent experimental replication, or qualified human scientific review.
source locator: Methods, "Brain" (train-test design, main-text XML); Figure S10, Panel B caption and heatmap (Additional File 1, page 15 of 18)
limitations: This is a conventional rejection-option BASELINE result only: the percentage of test-set cells SVMrejection declined to label. It is NOT a known-type accuracy claim, NOT an unknown/absent-type-detection accuracy claim, and NOT a claim that SVMrejection outperforms or underperforms any other classifier — no comparison across methods is asserted by this protocol or its results.; The scored denominator (exact number of cells actually scored in each individual train-test experiment) is UNREPORTED in the retrieved main text and supplement. Table 2's raw per-dataset cell counts (VISp 12,832; ALM 8,758; MTG 14,636) are recorded on the linked dataset records as context only, not as a confirmed denominator, and must not be used to back-calculate an implied rejected-cell count.; metric_direction is recorded as "unknown" for every result under this protocol: percentage unlabeled is a coverage/rejection-rate figure, not a standalone performance metric with one universal better/worse direction; it depends on context (e.g., the correctness of the calls a classifier would otherwise have made) not captured by this figure alone.; Only the SVMrejection row (9 values: 6 single-dataset-train + 3 two-dataset-train combinations) is ingested; the other 17 classifiers shown in the same Figure S10 Panel B heatmap are not ingested in this pass.; Figure S10 Panel A (3-population major-lineage level) is a separate figure/protocol and is not ingested here.; Figure S9 Panel A (PBMC, train-test median-F1/implied-unlabeled heatmap) carries no printed digit labels in the retrieved PDF at any resolution checked (up to 4800 dpi), and separately shows a source-internal conflict between its own caption text ("median F1-score"), the page's only legend ("Unlabeled (%)"), and the figure's overall title ("Percentage of unlabeled cells"); neither is resolved, and nothing from Figure S9 Panel A is ingested.; Reference labels are the original brain-atlas studies' author-assigned cell-population annotations, not independently adjudicated ground truth.; VISp and ALM share GSE115746. MTG is a separate human single-nucleus dataset; combinations involving MTG include cross-species data. Some two-dataset training combinations still share the mouse study with the test dataset, so independent-study holdout must not be inferred for all nine combinations.; Automated source review only; independent human scientific review remains outstanding.

Related records