Configuration

BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)

BioDiscoveryAgent (No-Tools), GPT-4o: No-Tools variant: no literature search, gene search or critic tool.

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

6 evaluations · 12 results. Different protocols are not a single leaderboard.

Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes

Sorted by Recall (hit ratio after round 5, all genes; mean of 10 runs) (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	Recall (hit ratio after round 5, all genes; mean of 10 runs)	Recall (hit ratio after round 5, non-essential genes only; mean of 10 runs)
BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)Evaluation	Scharenberg et al. 2023 screen, lysosomal choline recycling in pancreatic cells	0.311 (best in column)	0.286 (best in column)
BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)Evaluation	CAR-T proliferation screen (unpublished dataset used by Roohani et al. 2025)	0.116	0.12
BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)Evaluation	Schmidt et al. 2022 screen, interleukin-2 production in primary human T cells	0.074	0.09
BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)Evaluation	Schmidt et al. 2022 screen, interferon-gamma production in primary human T cells	0.059	0.068
BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)Evaluation	Carnevale et al. 2022 screen, T-cell resistance to tumour-microenvironment inhibitory signals	0.037	0.04
BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)Evaluation	Sanchez et al. 2021 screen, endogenous tau protein level in neurons	0.03	0.043

All 12 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: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Carnevale et al. 2022 screen, T-cell resistance to tumour-microenvironment inhibitory signals	0.037 Recall (hit ratio after round 5, all genes; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Carnev. (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'All' under 'Carnev.'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Carnevale et al. 2022 screen, T-cell resistance to tumour-microenvironment inhibitory signals	0.04 Recall (hit ratio after round 5, non-essential genes only; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Carnev. (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'N/E' under 'Carnev.'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: CAR-T proliferation screen (unpublished dataset used by Roohani et al. 2025)	0.116 Recall (hit ratio after round 5, all genes; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on CAR-T (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'All' under 'CAR-T'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: CAR-T proliferation screen (unpublished dataset used by Roohani et al. 2025)	0.12 Recall (hit ratio after round 5, non-essential genes only; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on CAR-T (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'N/E' under 'CAR-T'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, endogenous tau protein level in neurons	0.03 Recall (hit ratio after round 5, all genes; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Sanchez (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'All' under 'Sanchez'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, endogenous tau protein level in neurons	0.043 Recall (hit ratio after round 5, non-essential genes only; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Sanchez (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'N/E' under 'Sanchez'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Scharenberg et al. 2023 screen, lysosomal choline recycling in pancreatic cells	0.311 Recall (hit ratio after round 5, all genes; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Scharen. (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'All' under 'Scharen.'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Scharenberg et al. 2023 screen, lysosomal choline recycling in pancreatic cells	0.286 Recall (hit ratio after round 5, non-essential genes only; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Scharen. (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'N/E' under 'Scharen.'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Schmidt et al. 2022 screen, interferon-gamma production in primary human T cells	0.059 Recall (hit ratio after round 5, all genes; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Schmidt1 (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'All' under 'Schmidt1'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Schmidt et al. 2022 screen, interferon-gamma production in primary human T cells	0.068 Recall (hit ratio after round 5, non-essential genes only; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Schmidt1 (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'N/E' under 'Schmidt1'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Schmidt et al. 2022 screen, interleukin-2 production in primary human T cells	0.074 Recall (hit ratio after round 5, all genes; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Schmidt2 (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'All' under 'Schmidt2'
Configuration: BioDiscoveryAgent (No-Tools), GPT-4o (Roohani et al. 2025)	Protocol: BioDiscoveryAgent 1-gene perturbation design: hit ratio after 5 rounds of 128 genes Dataset: Schmidt et al. 2022 screen, interleukin-2 production in primary human T cells	0.09 Recall (hit ratio after round 5, non-essential genes only; mean of 10 runs) fraction · higher Uncertainty: Not yet extracted: Appendix Table 7 prints one standard deviation over 10 runs for these values; not extracted in this pass. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4o on Schmidt2 (Roohani et al. 2025) tgtval-20261009-protocol-roohani2025-hitratio-round5 Aggregation: Not reported BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Table 1, row 'GPT-4o', column 'N/E' under 'Schmidt2'

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

BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments · Original source · arXiv:2405.17631 version 3, updated 2025-03-09; published as a conference paper at ICLR 2025

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: tgtval-20261009-config-roohani2025-gpt-4o

areas: cells-tissues
contexts: research
method types: foundation_model
reported name: GPT-4o
foundation model eligible: true
source locator: Table 1, BioDiscoveryAgent (No-Tools) block
missing metadata: version: reason: unreported; note: No release or commit is printed for this configuration
parameters: No-Tools variant: no literature search, gene search or critic tool
provenance note: Configuration as run in the cited comparison.

Related records

configuration of: BioDiscoveryAgent
uses model: GPT-4o
system: GPT-4o on Carnev. (Roohani et al. 2025)
system: GPT-4o on CAR-T (Roohani et al. 2025)
system: GPT-4o on Sanchez (Roohani et al. 2025)
system: GPT-4o on Scharen. (Roohani et al. 2025)
system: GPT-4o on Schmidt1 (Roohani et al. 2025)
system: GPT-4o on Schmidt2 (Roohani et al. 2025)

Suggest a correction