Gene-level logistic regression plus attention edge (Kendiukhov 2026)

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

3 evaluations · 3 results. Different protocols are not a single leaderboard.

Protocol: Gene-level logistic regression under 5-fold GroupKFold by perturbation (Kendiukhov 2026)

Sorted by AUROC difference (Gene-level features plus attention edge minus gene-level features only (gene-level logistic regression); cross-validated AUROC over perturbation-gene pairs, 5-fold GroupKFold by perturbation) (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	AUROC difference (Gene-level features plus attention edge minus gene-level features only (gene-level logistic regression); cross-validated AUROC over perturbation-gene pairs, 5-fold GroupKFold by perturbation)	AUROC difference (Gene-level features plus attention edge minus gene-level features only (logistic regression); AUROC over propensity-matched perturbation-gene pairs (k = 5 negatives per positive))
Gene-level logistic regression plus attention edge (Kendiukhov 2026)Evaluation	Replogle genome-scale CRISPRi, RPE1, 1,167 perturbations, confound battery (Kendiukhov 2026)	0.0001 (best in column)	Not reported
Gene-level logistic regression plus attention edge (Kendiukhov 2026)Evaluation	Replogle genome-scale CRISPRi, K562, 280 evaluable perturbations (Kendiukhov 2026)	−0.0004	Not reported
Gene-level logistic regression plus attention edge (Kendiukhov 2026)Evaluation	Replogle CRISPRi K562, propensity-matched pairs (Kendiukhov 2026)	Not reported	−0.000

All 3 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: Gene-level logistic regression plus attention edge (Kendiukhov 2026)	Protocol: Gene-level logistic regression under 5-fold GroupKFold by perturbation (Kendiukhov 2026) Dataset subset: Replogle genome-scale CRISPRi, K562, 280 evaluable perturbations (Kendiukhov 2026)	−0.0004 AUROC difference (Gene-level features plus attention edge minus gene-level features only (gene-level logistic regression); cross-validated AUROC over perturbation-gene pairs, 5-fold GroupKFold by perturbation) unitless · higher Uncertainty: 95% CI -0.001 to 0.000. Methods 'Statistical framework': bootstrap CIs use 200 iterations with perturbation-level resampling unless otherwise noted; Fig. 4B caption: 'Bootstrap ∆AUROC distribution (n = 100)' Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Gene-level logistic regression plus attention edge on Replogle genome-scale CRISPRi, K562, 280 evaluable perturbations gfm-guides-20261014-kendiukhov2026-protocol-incremental-value-lr Aggregation: Not reported Systematic evaluation of single-cell foundation model interpretability: attention-derived edge scores add no incremental value over gene-level features for perturbation-target prediction (Kendiukhov 2026, BMC Genomics 27:634) · Page 8, 'No incremental pairwise value on the perturbation-target prediction task', paragraph 1, sentence 2: 'adding attention yields ∆AUROC = −0.0004 [−0.001, 0.000]'
Configuration: Gene-level logistic regression plus attention edge (Kendiukhov 2026)	Protocol: Gene-level logistic regression under 5-fold GroupKFold by perturbation (Kendiukhov 2026) Dataset subset: Replogle CRISPRi K562, propensity-matched pairs (Kendiukhov 2026)	−0.000 AUROC difference (Gene-level features plus attention edge minus gene-level features only (logistic regression); AUROC over propensity-matched perturbation-gene pairs (k = 5 negatives per positive)) unitless · higher Uncertainty: 95% CI -0.000 to 0.000. Interval type printed as '95% CI' without naming bootstrap; Methods 'Statistical framework': bootstrap CIs use 200 iterations with perturbation-level resampling unless otherwise noted Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Gene-level logistic regression plus attention edge on Replogle CRISPRi K562, propensity-matched pairs gfm-guides-20261014-kendiukhov2026-protocol-incremental-value-lr Aggregation: Not reported Systematic evaluation of single-cell foundation model interpretability: attention-derived edge scores add no incremental value over gene-level features for perturbation-target prediction (Kendiukhov 2026, BMC Genomics 27:634) · Page 9, paragraph 1 (continuing 'No incremental pairwise value'), sentence 3: 'add zero incremental value over gene-level features (∆AUROC = −0.000; 95% CI: [−0.000, +0.000]'
Configuration: Gene-level logistic regression plus attention edge (Kendiukhov 2026)	Protocol: Gene-level logistic regression under 5-fold GroupKFold by perturbation (Kendiukhov 2026) Dataset subset: Replogle genome-scale CRISPRi, RPE1, 1,167 perturbations, confound battery (Kendiukhov 2026)	0.0001 AUROC difference (Gene-level features plus attention edge minus gene-level features only (gene-level logistic regression); cross-validated AUROC over perturbation-gene pairs, 5-fold GroupKFold by perturbation) unitless · higher Uncertainty: 95% CI 0.0001 to 0.0001. Interval type printed as '95% CI' without naming bootstrap; Methods 'Statistical framework': bootstrap CIs use 200 iterations with perturbation-level resampling unless otherwise noted Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Gene-level logistic regression plus attention edge on Replogle genome-scale CRISPRi, RPE1, 1,167 perturbations, confound battery gfm-guides-20261014-kendiukhov2026-protocol-incremental-value-lr Aggregation: Not reported Systematic evaluation of single-cell foundation model interpretability: attention-derived edge scores add no incremental value over gene-level features for perturbation-target prediction (Kendiukhov 2026, BMC Genomics 27:634) · Page 10, 'Cross-cell-type generalisation of confound pattern', item (ii): 'adding attention yields ∆AUROC = +0.0001 [95% CI: +0.0001, +0.0001]'

Source checking is not independent reproduction. Release 2026-11-02-a7f03c9e55a5.

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 2026-11-02-a7f03c9e55a5
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 2026-11-02-a7f03c9e55a5 · Record review: source checked

1 source record and release history

Systematic evaluation of single-cell foundation model interpretability: attention-derived edge scores add no incremental value over gene-level features for perturbation-target prediction (Kendiukhov 2026, BMC Genomics 27:634) · Original source · Version of record, BMC Genomics 27:634, published 2026-07-22 (received 2026-03-10, accepted 2026-05-14)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: gfm-guides-20261014-kendiukhov2026-config-gene-level-lr-plus-attention

areas: single-cell; biological-networks
contexts: research
method types: supervised_machine_learning
reported name: gene-level plus attention
foundation model eligible: false
representation: Gene-level features of the candidate target plus one Geneformer V2-316M attention edge score
training setting: Fitted under 5-fold GroupKFold by perturbation on each dataset
parameters: Attention layer used for the edge feature not stated in the article
access reuse: Analysis code github.com/Biodyn-AI/biomechinterp-framework and Zenodo 10.5281/zenodo.18701417; not pinned to a commit in the article
source locator: Page 8 'No incremental pairwise value on the perturbation-target prediction task', paragraph 1; page 16 'Confound decomposition'
missing metadata: version: reason: inapplicable; note: Defined in the article

Related records