Relevance of Combination phenotype prediction from small training sets: test RMSE at fixed average gene frequency to "Select perturbations for a defined cellular response"
- Use case
- Select perturbations for a defined cellular response
- Assessed by
- Combination phenotype prediction from small training sets: test RMSE at fixed average gene frequency
- Relevance
- outside_scope
- Endpoint
- Test RMSE (x10^-2) of predicted two-gene viability on a 30% held-out set after training on subsamples in which each gene appears on average 4, 10 or 20 times, for NAIAD, RECOVER, GEARS, a linear model and an MLP (not shown as evidence)
Outside scope. Prediction error averaged over all held-out pairs measures how well combination effects are predicted, not which pairs to test under a budget; the same paper's active-learning protocol is linked separately for that. Recorded so these errors are not read as selection performance.
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.
15 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
Citation locations1 values
| Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
Constraints1 values
| Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
| Endpoint Test RMSE (x10^-2) of predicted two-gene viability on a 30% held-out set after training on subsamples in which each gene appears on average 4, 10 or 20 times, for NAIAD, RECOVER, GEARS, a linear model and an MLP (not shown as evidence) Context-only references | Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
| Claimed field links:assessed_by:pheno-20261022-qin2025-protocol-downsampling-rmse Context-only references | Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
Limitations2 values
| Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
Pinned records3 values
| Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
| Rationale Outside scope. Prediction error averaged over all held-out pairs measures how well combination effects are predicted, not which pairs to test under a budget; the same paper's active-learning protocol is linked separately for that. Recorded so these errors are not read as selection performance. Context-only references | Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
| Reason Add relevance judgement for a protocol extracted in the 2026-10-22 refresh (issue #184). Drafted by the Claude (Opus 5.5) judgement agent in the evidence-pass workflow for issue #184 (batch data/omics/use-case-refresh-phenotype-20261022); no human review claimed. Context-only references | Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
| Relevance outside_scope Context-only references | Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
| Revision 1 Context-only references | Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6 Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py |
Sources and history
Release 2026-10-23-5347695500ed · Record review: source checked
1 source record and release history
- Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Original source · Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read.
Technical metadata and extraction receipts
Stable ID: use-case-mapping-phenotype-20261022-qin2025-downsampling-rmse
- field
- links:assessed_by:pheno-20261022-qin2025-protocol-downsampling-rmse
- value
- pheno-20261022-qin2025-protocol-downsampling-rmse
- relevance
- outside_scope
- endpoint
- Test RMSE (x10^-2) of predicted two-gene viability on a 30% held-out set after training on subsamples in which each gene appears on average 4, 10 or 20 times, for NAIAD, RECOVER, GEARS, a linear model and an MLP (not shown as evidence)
- rationale
- Outside scope. Prediction error averaged over all held-out pairs measures how well combination effects are predicted, not which pairs to test under a budget; the same paper's active-learning protocol is linked separately for that. Recorded so these errors are not read as selection performance.
- constraints
- Do not cite these RMSE values as evidence of which combinations to test.
- limitations
- Author-reported: the NAIAD authors ran every model, adapting RECOVER and GEARS to this setting.; Training-set sizes per gene frequency are not printed with Table 1.
- revision
- 1
- reason
- Add relevance judgement for a protocol extracted in the 2026-10-22 refresh (issue #184). Drafted by the Claude (Opus 5.5) judgement agent in the evidence-pass workflow for issue #184 (batch data/omics/use-case-refresh-phenotype-20261022); no human review claimed.
- citation locators
- source id: pheno-20261022-qin2025-source-pmlr; locator: Sections 4.2 and 5.2; Table 1; Table 6
- source locator
- pheno-20261022-qin2025-source-pmlr: Sections 4.2 and 5.2; Table 1; Table 6
- review
- method: source-hash-verification; scope-review; reviewer: claude; reviewer note: Claude (Opus 5.5) use-case review agent in the evidence-pass workflow for issue #184, independent of the agent that drafted these judgements and the summary; no human review claimed; reviewed at: 2026-10-22T18:00:00Z; note: Outside scope confirmed: test RMSE over all held-out pairs (Sections 4.2 and 5.2, Tables 1 and 6) measures prediction accuracy, not which pairs to test under a budget. Kept with its assessed_by link, following the store's outside-scope precedent, so the values are not read as selection performance. No reviewed_evaluations, as for other outside-scope judgements; the source is clean. See docs/reviews/use-cases/phenotype-perturbation-selection-2026-10-22.md.
- pins
- record id: pheno-20261022-qin2025-protocol-downsampling-rmse; fields: status; name; description; facets; source_ids; links; attributes; sha256: 4a747d3a1e3664ee19968d2704b6fb14bb4e195664c8c806f674b881a248af04; record id: pheno-20261022-qin2025-source-pmlr; fields: status; attributes.artifact_sha256; sha256: 1f6d6ab61923eb5a0c5b288bee81d4696d780cf5bee04e9eb0c472f1795ef5cd; record id: use-case-phenotype-perturbation-selection; fields: status; name; description; facets; attributes.question; attributes.decision; attributes.inputs; attributes.output; attributes.exclusions; attributes.setting; attributes.clinical_scope; attributes.intended_users; sha256: f00ad84ac33a7fc848a5df63f06252d4515e87648cf34342405b06f91a115fd4