Configuration

NAIAD with MPE acquisition

NAIAD ensemble retrained each round, with the next pairs chosen by MPE acquisition.

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 · 17 results. Different protocols are not a single leaderboard.

Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4

Sorted by True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) (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	True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning)	Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4)	True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 0 of active learning)	True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 1 of active learning)	True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 2 of active learning)	True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 3 of active learning)
NAIAD with MPE acquisitionEvaluation	Horlbeck et al. 2018 combinatorial CRISPRi in Jurkat T cells, gene-level viability (as processed by Qin et al. 2025)	161 (best in column)	23	Not reported	Not reported	Not reported	Not reported
NAIAD with MPE acquisitionEvaluation	Horlbeck et al. 2018 combinatorial CRISPRi in Jurkat T cells, gene-level viability (as processed by Qin et al. 2025)	150.0 (4.0)	Not reported	Not reported	Not reported	Not reported	Not reported
NAIAD with MPE acquisitionEvaluation	Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	144	14	Not reported	Not reported	Not reported	Not reported
NAIAD with MPE acquisitionEvaluation	Simpson et al. 2023 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	144	27 (best in column)	Not reported	Not reported	Not reported	Not reported
NAIAD with MPE acquisitionEvaluation	Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	143.0 (1.5)	Not reported	Not reported	Not reported	Not reported	Not reported
NAIAD with MPE acquisitionEvaluation	Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	143.0 (1.5)	Not reported	86.0 (1.0)	100.3 (0.9)	112.3 (1.7)	128.0 (2.5)
NAIAD with MPE acquisitionEvaluation	Simpson et al. 2023 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	141.7 (3.7)	Not reported	Not reported	Not reported	Not reported	Not reported
NAIAD with MPE acquisitionEvaluation	Horlbeck et al. 2018 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	99.7 (4.2)	Not reported	Not reported	Not reported	Not reported	Not reported
NAIAD with MPE acquisitionEvaluation	Horlbeck et al. 2018 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	94	14.3	Not reported	Not reported	Not reported	Not reported

All 17 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: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Horlbeck et al. 2018 combinatorial CRISPRi in Jurkat T cells, gene-level viability (as processed by Qin et al. 2025)	23 Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Horlbeck Jurkat (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 3, row "MPE", column "Horlbeck Jurkat"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Horlbeck et al. 2018 combinatorial CRISPRi in Jurkat T cells, gene-level viability (as processed by Qin et al. 2025)	161 True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Horlbeck Jurkat (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 2, row "MPE", column "Horlbeck Jurkat"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Horlbeck et al. 2018 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	14.3 Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Horlbeck K562 (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 3, row "MPE", column "Horlbeck K562"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Horlbeck et al. 2018 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	94 True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Horlbeck K562 (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 2, row "MPE", column "Horlbeck K562"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	14 Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 3, row "MPE", column "Norman"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	144 True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 2, row "MPE", column "Norman"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Simpson et al. 2023 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	27 Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Simpson (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 3, row "MPE", column "Simpson"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Simpson et al. 2023 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	144 True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Simpson (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 2, row "MPE", column "Simpson"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Horlbeck et al. 2018 combinatorial CRISPRi in Jurkat T cells, gene-level viability (as processed by Qin et al. 2025)	150.0 (4.0) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: SE 4.0. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Horlbeck Jurkat (Table 4, NAIAD against RECOVER by acquisition function) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 4, row "NAIAD MPE", column "Horlbeck Jurkat"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Horlbeck et al. 2018 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	99.7 (4.2) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: SE 4.2. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Horlbeck K562 (Table 4, NAIAD against RECOVER by acquisition function) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 4, row "NAIAD MPE", column "Horlbeck K562"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	143.0 (1.5) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: SE 1.5. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Table 4, NAIAD against RECOVER by acquisition function) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 4, row "NAIAD MPE", column "Norman"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Simpson et al. 2023 combinatorial CRISPRi in K562, gene-level viability (as processed by Qin et al. 2025)	141.7 (3.7) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: SE 3.7. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Simpson (Table 4, NAIAD against RECOVER by acquisition function) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 4, row "NAIAD MPE", column "Simpson"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	86.0 (1.0) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 0 of active learning) count · higher Uncertainty: SE 1.0. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Table 7, NAIAD, RECOVER and GEARS over rounds on Norman) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 7, row "NAIAD MPE", column "Round 0"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	100.3 (0.9) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 1 of active learning) count · higher Uncertainty: SE 0.9. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Table 7, NAIAD, RECOVER and GEARS over rounds on Norman) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 7, row "NAIAD MPE", column "Round 1"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	112.3 (1.7) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 2 of active learning) count · higher Uncertainty: SE 1.7. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Table 7, NAIAD, RECOVER and GEARS over rounds on Norman) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 7, row "NAIAD MPE", column "Round 2"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	128.0 (2.5) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 3 of active learning) count · higher Uncertainty: SE 2.5. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Table 7, NAIAD, RECOVER and GEARS over rounds on Norman) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 7, row "NAIAD MPE", column "Round 3"
Configuration: NAIAD with MPE acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	143.0 (1.5) True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: SE 1.5. Caption: error is SE across three cross-fold replicates Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with MPE acquisition on Norman (Table 7, NAIAD, RECOVER and GEARS over rounds on Norman) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 7, row "NAIAD MPE", column "Round 4"

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

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.

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: pheno-20261022-qin2025-config-naiad-mpe

areas: cells-tissues
contexts: research
method types: supervised_machine_learning
reported name: NAIAD, MPE
configuration type: paper_evaluation
foundation model eligible: false
version: As run by Qin et al. 2025 (PMLR 267); no release or commit is printed
access reuse: NAIAD code at https://github.com/NeptuneBio/NAIAD (Apache-2.0 per GitHub metadata); the commit used for the paper is not stated
parameters: NAIAD ensemble with adaptive gene-embedding size (Table 9 schedule), retrained each round; seed set of uniformly sampled pairs shared by all acquisition functions. Ensemble size not stated.
source locator: Tables 2 to 4 row labels; Section 5.3; Appendix C
provenance note: The paper's own method.
review: method: source-hash-verification; scope-review; reviewer: claude; reviewer note: Claude (Opus 5.5) review agent, independent of the extraction agent; no human review claimed; date: 2026-10-22; artifact sha256: 983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb; retrieval url: https://raw.githubusercontent.com/mlresearch/v267/main/assets/qin25g/qin25g.pdf; method note: Re-hashed the PDF in the scratchpad (SHA-256 matches the source record). Read the cited sections in the pdftotext layer of the pinned PDF.; note: Fields checked against the cited locator; nothing found that the source does not state.
selection: Maximum Predicted Effect acquisition: pairs with the highest absolute ensemble-mean prediction (Appendix C.2)

Related records

Suggest a correction