Dataset

Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons

The Sanchez et al. 2021 screen restricted to decreased tau expression (Gupta et al. 2025 appendix B.1.1).

Research readiness

These checks assess whether the evidence supports a reproducible investigation. A source-checked score alone does not meet these requirements.

Release 2026-10-10-6e93f504adfc · Evidence verified: Not verified

Evidence incomplete

Replay metrics

Exact outcomes, predictions, identifiers and evaluator are connected.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing

Verified: Not verified

Evidence incomplete

Investigate discrepancies

Replay evidence includes annotations and an assessment of dependence. Unknown independence permits descriptive analysis only.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing
annotations: verification is missing
dependence: verification is missing

Verified: Not verified

Evidence incomplete

Run locally

A pinned recipe describes the inputs, environment and resource requirements.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
recipe pinned: verification is missing
resource estimate: verification is missing

Verified: Not verified

Evidence incomplete

Validate independently

Separate data and exposure records support an independent test.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
independent validation: verification is missing
overlap checked: verification is missing

Verified: Not verified

Readiness describes the evidence in this release. Availability on your computer is checked separately when an investigation runs. Existing data exposure can prevent independent validation even when files are available.

Artifacts and reproduction

No verified artifact manifest is connected to this record yet. The gaps above identify what is needed before analysis can begin.

Read reviewed discrepancy investigations

Evaluation results

11 evaluations · 11 results. Different protocols are not a single leaderboard.

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: BDA, llama-3-1-8b backbone (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	43.6 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source BDA (Llama-3.1-8B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 1, 'Llama-3.1-8B backbone' block, row 'BDA', column 'Sanchez Down'
Configuration: BDA, qwen-2-7b backbone (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	26 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source BDA (Qwen-2-7B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 1, 'Qwen-2-7B backbone' block, row 'BDA', column 'Sanchez Down'
Configuration: BDA-Rand, claude-3-5-sonnet backbone (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	57.6 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source BDA-Rand (Claude 3.5 Sonnet backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 1, 'Claude 3.5 Sonnet backbone' block, row 'BDA-Rand', column 'Sanchez Down'
Configuration: BDA-Rand, llama-3-1-8b backbone (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	45 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source BDA-Rand (Llama-3.1-8B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 1, 'Llama-3.1-8B backbone' block, row 'BDA-Rand', column 'Sanchez Down'
Configuration: BDA-Rand, qwen-2-7b backbone (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	35.4 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source BDA-Rand (Qwen-2-7B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 1, 'Qwen-2-7B backbone' block, row 'BDA-Rand', column 'Sanchez Down'
Configuration: BDA (Replicated), claude-3-5-sonnet backbone (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	51.8 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source BDA (Replicated) (Claude 3.5 Sonnet backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 1, 'Claude 3.5 Sonnet backbone' block, row 'BDA (Replicated)', column 'Sanchez Down'
Configuration: BDA (Reported Numbers), claude-3-5-sonnet backbone (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	N/A true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Result quoted from another source · Source checked Methods, coverage and source BDA (Reported Numbers) (Claude 3.5 Sonnet backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 1, 'Claude 3.5 Sonnet backbone' block, row 'BDA (Reported Numbers)', column 'Sanchez Down'
Configuration: GP over llama-3-1-8b embeddings (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	30 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source GP (Llama-3.1-8B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 2, 'Llama-3.1-8B backbone' block, row 'GP', column 'Sanchez Down'
Configuration: GP over qwen-2-7b embeddings (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	30 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source GP (Qwen-2-7B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 2, 'Qwen-2-7B backbone' block, row 'GP', column 'Sanchez Down'
Configuration: Linear UCB over llama-3-1-8b embeddings (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	42 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Linear UCB (Llama-3.1-8B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 2, 'Llama-3.1-8B backbone' block, row 'Linear UCB', column 'Sanchez Down'
Configuration: Linear UCB over qwen-2-7b embeddings (Gupta et al. 2025)	Protocol: Independent replication of 1-gene perturbation design: cumulative hits after 5 rounds of 128 genes Dataset: Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons	41 true-positive-count count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Linear UCB (Qwen-2-7B backbone) on Sanchez Down (Gupta et al. 2025) tgtval-20261009-protocol-gupta2025-cumulative-hits-round5 Aggregation: Not reported LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Table 2, 'Qwen-2-7B backbone' block, row 'Linear UCB', column 'Sanchez Down'

Source checking is not independent reproduction. Release 2026-10-10-6e93f504adfc.

Dataset and evaluation context

A dataset supplies biological observations. The evaluation protocol defines how those observations are split, used and scored.

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.

7 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-10-6e93f504adfc
Property and statement	Original source and location	Review and provenance
attributes.assay Pooled CRISPR perturbation screen with a phenotypic readout Context-only references	LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? Original source ↗ Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Version: Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510 Retrieved: 2026-10-09T20:55:36Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.assay` Source artifact SHA-256: `7dda0b590f2736b7d30e48b97167a0868a2b4bde253589d8aff7929d01506ff9` Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py Inspected artifact
attributes.population Over 18,000 genes; same measurements as the Sanchez screen, hits restricted to decreases Context-only references	LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? Original source ↗ Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Version: Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510 Retrieved: 2026-10-09T20:55:36Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.population` Source artifact SHA-256: `7dda0b590f2736b7d30e48b97167a0868a2b4bde253589d8aff7929d01506ff9` Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py Inspected artifact
attributes.scope_note Replayed offline: the benchmark looks up the measured phenotypic response of each selected gene instead of running a new experiment. Genes the screen did not assay are absent from the candidate pool, so they are neither hits nor non-hits. Context-only references	LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? Original source ↗ Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Version: Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510 Retrieved: 2026-10-09T20:55:36Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.scope_note` Source artifact SHA-256: `7dda0b590f2736b7d30e48b97167a0868a2b4bde253589d8aff7929d01506ff9` Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py Inspected artifact
attributes.source_locator Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Context-only references	LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? Original source ↗ Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Version: Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510 Retrieved: 2026-10-09T20:55:36Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.source_locator` Source artifact SHA-256: `7dda0b590f2736b7d30e48b97167a0868a2b4bde253589d8aff7929d01506ff9` Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py Inspected artifact
attributes.version As replayed by the cited benchmark from the original screen Context-only references	LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? Original source ↗ Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Version: Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510 Retrieved: 2026-10-09T20:55:36Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.version` Source artifact SHA-256: `7dda0b590f2736b7d30e48b97167a0868a2b4bde253589d8aff7929d01506ff9` Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py Inspected artifact
description The Sanchez et al. 2021 screen restricted to decreased tau expression (Gupta et al. 2025 appendix B.1.1). Context-only references	LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? Original source ↗ Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Version: Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510 Retrieved: 2026-10-09T20:55:36Z	not individually reviewed No individual claim review recorded Audit details Field: `description` Source artifact SHA-256: `7dda0b590f2736b7d30e48b97167a0868a2b4bde253589d8aff7929d01506ff9` Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py Inspected artifact
name Sanchez et al. 2021 screen, decreased endogenous tau protein level in neurons Context-only references	LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? Original source ↗ Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1 Version: Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510 Retrieved: 2026-10-09T20:55:36Z	not individually reviewed No individual claim review recorded Audit details Field: `name` Source artifact SHA-256: `7dda0b590f2736b7d30e48b97167a0868a2b4bde253589d8aff7929d01506ff9` Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py Inspected artifact

All evidence (JSONL) (gzip)How to read the evidence

Sources and history

Release 2026-10-10-6e93f504adfc · Record review: source checked

1 source records and release history

LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet? · Original source · Findings of the Association for Computational Linguistics: EMNLP 2025, pages 15482-15510

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: tgtval-20261009-data-sanchez2021-tau-down

areas: cells-tissues
contexts: research
version: As replayed by the cited benchmark from the original screen
population: Over 18,000 genes; same measurements as the Sanchez screen, hits restricted to decreases
assay: Pooled CRISPR perturbation screen with a phenotypic readout
scope note: Replayed offline: the benchmark looks up the measured phenotypic response of each selected gene instead of running a new experiment. Genes the screen did not assay are absent from the candidate pool, so they are neither hits nor non-hits.
source locator: Roohani et al. 2025 section 4.1; Gupta et al. 2025 appendix B.1.1
missing metadata: positives: reason: unreported; note: Roohani et al. do not print the hit count or the threshold tau per screen; Gupta et al. print their own ground-truth hit counts, recorded as a claim on their protocol; total: reason: unreported; note: Stated as 'over 18,000 genes' rather than an exact count

Related records

Suggest a correction