Accuracy (3-class (non-DE, up, down) accuracy within true |log2 FC| quantile Q3 (0.037–0.065); quantile bins per cell line, averaged over four cell lines) of STATE (few-shot cross-cell-line) (Sun et al. 2026) on K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)
99% Accuracy (3-class (non-DE, up, down) accuracy within true |log2 FC| quantile Q3 (0.037–0.065); quantile bins per cell line, averaged over four cell lines)
Methods
- Tested configuration
- STATE (few-shot cross-cell-line) (Sun et al. 2026)
- Protocol
- Sun et al. 2026 cellular-context DE classification (cross cell line)
- Dataset subset
- K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)
- Procedure
- Sun et al. 2026 cellular-context DE classification (cross cell line)
- Evaluation
- STATE on Sun et al. 2026 cross-cell-line DE classification, four cell lines
- Coverage
- Not reported scored / Not reported eligible
- Uncertainty
- Not reported by the source
- Evidence
- Independent external evaluation · source checkedBeyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 4, row 'STATE', column 'Q3 (0.037–0.065)'
A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: source checked. Source checked does not mean independently reproduced.
Reproduction
- Split
- 60/40 split by perturbation; held-out cell line
- Adaptation
- Trained per split
- Scoring implementation
- Wilcoxon rank-sum DE calls, BH-adjusted p < 0.01, on predicted and observed expression
No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.
Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.
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.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Reported result 98.99% Individual claims | Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models Table 4, row 'STATE', column 'Q3 (0.037–0.065)' Version: bioRxiv preprint version 1, posted 2026-07-24 | source checked ["source-hash-verification","deterministic-table-parse","independent-cell-check"] · 2026-10-12 independent paper Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsExtraction: Parsed the pdftotext -layout text of the pinned PDF with extract_sun2026.py: method rows matched by printed name with exactly five percentages, row order and column headers asserted. Table 4, row 'STATE', column 'Q3 (0.037–0.065)'. Independent review 2026-10-12: value, locator, metric, unit, direction and linked configuration, protocol and dataset checked by the reviewer's own parse of all Tables 1-4 cells and a hand check of 48 values against the rendered pages. Corrections from the review applied by the extractor as listed in review.md. Field: Source artifact SHA-256: Hash scope: Exact bytes of one download; bioRxiv serves slightly different PDF bytes per download, so the gzip copy in artifacts/ is the reference. Extraction artifact SHA-256: |
Sources and history
Release 2026-10-23-5347695500ed · Record review: source checked
1 source record and release history
- Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Original source · bioRxiv preprint version 1, posted 2026-07-24
Technical metadata and extraction receipts
Stable ID: lit-cells-networks-20261011-sun2026de-result-t4-cross-q3-state
- metric
- accuracy
- metric direction
- higher
- unit
- percent
- printed value
- 98.99%
- source locator
- Table 4, row 'STATE', column 'Q3 (0.037–0.065)'
- review
- method: source-hash-verification; deterministic-table-parse; independent-cell-check; method note: Parsed the pdftotext -layout text of the pinned PDF with extract_sun2026.py: method rows matched by printed name with exactly five percentages, row order and column headers asserted. Table 4, row 'STATE', column 'Q3 (0.037–0.065)'; reviewer: claude; reviewer note: Claude review agent reviewer-cn-1, independent of the extractor lit-cells-networks; no human review claimed; date: 2026-10-12; artifact sha256: fa1c6a6345c45e14b9c373dfa8aff12633ae646feece1cfa3899c3ed61f6cfdf; retrieval url: https://www.biorxiv.org/content/10.64898/2026.07.20.739620v1.full.pdf; note: Extraction: Parsed the pdftotext -layout text of the pinned PDF with extract_sun2026.py: method rows matched by printed name with exactly five percentages, row order and column headers asserted. Table 4, row 'STATE', column 'Q3 (0.037–0.065)'. Independent review 2026-10-12: value, locator, metric, unit, direction and linked configuration, protocol and dataset checked by the reviewer's own parse of all Tables 1-4 cells and a hand check of 48 values against the rendered pages. Corrections from the review applied by the extractor as listed in review.md.
- numeric value
- 98.99
- metric qualifier
- 3-class (non-DE, up, down) accuracy within true |log2 FC| quantile Q3 (0.037–0.065); quantile bins per cell line, averaged over four cell lines
- missing metadata
- uncertainty: reason: unreported