Configuration

Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)

Per-gene L2-regularised multinomial logistic regression (strength 0.01) on GenePT perturbation embeddings, adapted from GenePert.

10 evaluations · 20 results

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.

Evaluations and results

10 evaluations · 20 results. Different protocols are not a single leaderboard.

Sorted by Accuracy (3-class (non-DE, up, down) accuracy within true |log2 FC| quantile Q1 (0–0.016); quantile bins per cell line, averaged over four cell lines) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.

Tested configuration	Protocol and dataset	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q1 (0–0.016); quantile bins per cell line, averaged over four cell lines)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q1 (0–0.017); quantile bins per cell line, averaged over four cell lines)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q2 (0.016–0.034); quantile bins per cell line, averaged over four cell lines)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q2 (0.017–0.037); quantile bins per cell line, averaged over four cell lines)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q3 (0.034–0.060); quantile bins per cell line, averaged over four cell lines)	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)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q4 (0.060–0.108); quantile bins per cell line, averaged over four cell lines)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q4 (0.065–0.116); quantile bins per cell line, averaged over four cell lines)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q5 (>0.108); quantile bins per cell line, averaged over four cell lines)	Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q5 (>0.116); quantile bins per cell line, averaged over four cell lines)	Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence))
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 unseen-perturbation DE classification (within cell line), K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	92.4%	Not reported	91.5%	Not reported	89.6%	Not reported	85%	Not reported	70.6%	Not reported	14.7%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 unseen-perturbation DE classification (within cell line), HepG2 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	12.9%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 unseen-perturbation DE classification (within cell line), Jurkat essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	10.4%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 unseen-perturbation DE classification (within cell line), K562 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	18.5%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 unseen-perturbation DE classification (within cell line), RPE1 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	17.1%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 cellular-context DE classification (cross cell line), K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	Not reported	89%	Not reported	88.3%	Not reported	86.5%	Not reported	82.7%	Not reported	68.4%	16.6%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 cellular-context DE classification (cross cell line), HepG2 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	18.9%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 cellular-context DE classification (cross cell line), Jurkat essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	18%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 cellular-context DE classification (cross cell line), K562 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	12.9%
Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)Evaluation	Sun et al. 2026 cellular-context DE classification (cross cell line), RPE1 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	16.6%

All 20 result rows with coverage, uncertainty and sources

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: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	14.7% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 1, row 'Logistic regression', column 'Average'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: HepG2 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	12.9% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, HepG2 lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 1, row 'Logistic regression', column 'HepG2'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: Jurkat essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	10.4% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, Jurkat lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 1, row 'Logistic regression', column 'Jurkat'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: K562 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	18.5% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, K562 lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 1, row 'Logistic regression', column 'K562'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: RPE1 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	17.1% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, RPE1 lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 1, row 'Logistic regression', column 'RPE1'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	92.4% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q1 (0–0.016); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 2, row 'Logistic regression', column 'Q1 (0–0.016)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	91.5% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q2 (0.016–0.034); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 2, row 'Logistic regression', column 'Q2 (0.016–0.034)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	89.6% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q3 (0.034–0.060); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 2, row 'Logistic regression', column 'Q3 (0.034–0.060)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	85% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q4 (0.060–0.108); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 2, row 'Logistic regression', column 'Q4 (0.060–0.108)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 unseen-perturbation DE classification (within cell line) Dataset subset: K562, RPE1, HepG2 and Jurkat test sets (Sun et al. 2026)	70.6% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q5 (>0.108); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 within-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-within-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 2, row 'Logistic regression', column 'Q5 (>0.108)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (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)	16.6% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 3, row 'Logistic regression', column 'Average'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 cellular-context DE classification (cross cell line) Dataset subset: HepG2 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	18.9% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, HepG2 lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 3, row 'Logistic regression', column 'HepG2'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 cellular-context DE classification (cross cell line) Dataset subset: Jurkat essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	18% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, Jurkat lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 3, row 'Logistic regression', column 'Jurkat'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 cellular-context DE classification (cross cell line) Dataset subset: K562 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	12.9% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, K562 lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 3, row 'Logistic regression', column 'K562'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (Sun et al. 2026)	Protocol: Sun et al. 2026 cellular-context DE classification (cross cell line) Dataset subset: RPE1 essential-gene Perturb-seq, 40% test perturbations (Sun et al. 2026)	16.6% Proportion (share of perturbation-gene pairs predicted as differentially expressed (compare with the printed true DE prevalence)) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, RPE1 lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 3, row 'Logistic regression', column 'RPE1'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (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)	89% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q1 (0–0.017); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 4, row 'Logistic regression', column 'Q1 (0–0.017)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (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)	88.3% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q2 (0.017–0.037); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 4, row 'Logistic regression', column 'Q2 (0.017–0.037)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (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)	86.5% 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) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 4, row 'Logistic regression', column 'Q3 (0.037–0.065)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (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)	82.7% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q4 (0.065–0.116); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 4, row 'Logistic regression', column 'Q4 (0.065–0.116)'
Configuration: Per-gene L2 logistic regression on GenePT embeddings (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)	68.4% Accuracy (3-class (non-DE, up, down) accuracy within true \|log2 FC\| quantile Q5 (>0.116); quantile bins per cell line, averaged over four cell lines) percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Methods describe results on both the full data and a PerturbQA-downsampled subset ('We annotate our results to distinguish between these two settings'), but the captions of Tables 1-4 do not say which setting they report. The printed true DE prevalence (1.75-4.41%) matches the full data's 95-98% non-DE share (Table 5), not the downsampled set; this is an inference, not a printed statement.Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Section 3.2 'Full vs. Curated Dataset'; Tables 1-5 Methods, coverage and source Logistic regression on Sun et al. 2026 cross-cell-line DE classification, four cell lines lit-cells-networks-20261011-sun2026de-protocol-cross-cell-line Aggregation: Not reported Beyond Expression Prediction: Benchmarking Differential Expression Classification in Single-Cell Perturbation Models · Table 4, row 'Logistic regression', column 'Q5 (>0.116)'

Source checking is not independent reproduction. Release 2026-10-23-5347695500ed.

Use this configuration

How it works, versions and access

Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.

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 2026-10-23-5347695500ed
Property and statement	Original source and location	Review and provenance

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

How to read the evidence

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

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: lit-cells-networks-20261011-sun2026de-config-logistic-regression

areas: single-cell
contexts: research
method types: supervised_machine_learning
reported name: Logistic regression
foundation model eligible: false
source locator: Section 3.3
missing metadata: version: reason: unreported

Related records

Suggest a correction