Protocol

Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes

Evidence concern: excluded from comparisons

Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A"

A system predicts whether a CRISPR knockout of a gene produces a complex phenotype, on balanced queries curated from two late-2024 screen publications.

3 evaluations · 24 results

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.

3 recorded evaluations, 24 metric rows. A comparison chart has not yet been validated for these results. The table retains the individual findings and their sources.

View coverage and remaining gaps across all benchmarks

Results

Results are available, but no reviewed comparison panel is linked in this release.

All evaluations

3 evaluations · 24 results. Different protocols are not a single leaderboard.

Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes · Dataset: Virtual CRISPR Difficult Benchmark

Sorted by AUPRC (Difficult Benchmark) (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	AUPRC (Difficult Benchmark)	AUROC (Difficult Benchmark)	F1 (Difficult Benchmark)	False positive rate (Difficult Benchmark)	Negative predictive value (Difficult Benchmark)	Precision (Difficult Benchmark)	Recall (Difficult Benchmark)	Specificity (Difficult Benchmark)
Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)Evaluation	0.86 (best in column)	0.89 (best in column)	0.84 (best in column)	0.15 (best in column)	0.83 (best in column)	0.85 (best in column)	0.83 (best in column)	0.85 (best in column)
Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)Evaluation	0.69	0.72	0.67	0.26	0.67	0.71	0.64	0.74
GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)Evaluation	N/A	N/A	0.35	0.79	0.23	0.32	0.41	0.21

All 24 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: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.86 AUPRC (Difficult Benchmark) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "AUPRC"
Configuration: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.89 AUROC (Difficult Benchmark) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "AUROC"
Configuration: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.84 F1 (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "F1"
Configuration: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.15 False positive rate (Difficult Benchmark) fraction · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "FPR"
Configuration: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.83 Negative predictive value (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "NPV"
Configuration: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.85 Precision (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "PPV"
Configuration: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.83 Recall (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "Sens."
Configuration: Virtual CRISPR embedding classifier, raw-text embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.85 Specificity (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Raw Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Raw Emb.", column "Spec."
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.69 AUPRC (Difficult Benchmark) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "AUPRC"
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.72 AUROC (Difficult Benchmark) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "AUROC"
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.67 F1 (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "F1"
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.26 False positive rate (Difficult Benchmark) fraction · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "FPR"
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.67 Negative predictive value (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "NPV"
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.71 Precision (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "PPV"
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.64 Recall (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "Sens."
Configuration: Virtual CRISPR embedding classifier, GPT-4o summary embeddings (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.74 Specificity (Difficult Benchmark) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source Summ. Emb. on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "Summ. Emb.", column "Spec."
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	N/A AUPRC (Difficult Benchmark) unitless · higher Uncertainty: Not applicable Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "AUPRC"
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	N/A AUROC (Difficult Benchmark) unitless · higher Uncertainty: Not applicable Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "AUROC"
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.35 F1 (Difficult Benchmark) fraction · higher Uncertainty: ± 0.17 (type not stated). Printed on the line below the row; the caption says results are aggregated across prompting strategies but does not name the statistic. Table 1 uses mean and standard deviation for the same kind of aggregation. Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "F1"
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.79 False positive rate (Difficult Benchmark) fraction · lower Uncertainty: ± 0.19 (type not stated). Printed on the line below the row; the caption says results are aggregated across prompting strategies but does not name the statistic. Table 1 uses mean and standard deviation for the same kind of aggregation. Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "FPR"
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.23 Negative predictive value (Difficult Benchmark) fraction · higher Uncertainty: ± 0.13 (type not stated). Printed on the line below the row; the caption says results are aggregated across prompting strategies but does not name the statistic. Table 1 uses mean and standard deviation for the same kind of aggregation. Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "NPV"
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.32 Precision (Difficult Benchmark) fraction · higher Uncertainty: ± 0.14 (type not stated). Printed on the line below the row; the caption says results are aggregated across prompting strategies but does not name the statistic. Table 1 uses mean and standard deviation for the same kind of aggregation. Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "PPV"
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.41 Recall (Difficult Benchmark) fraction · higher Uncertainty: ± 0.22 (type not stated). Printed on the line below the row; the caption says results are aggregated across prompting strategies but does not name the statistic. Table 1 uses mean and standard deviation for the same kind of aggregation. Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "Sens."
Configuration: GPT-4o, direct prompting, mean over prompting strategies (Song et al. 2025)	Protocol: Virtual CRISPR Difficult Benchmark: hit prediction for complex phenotypes Dataset: Virtual CRISPR Difficult Benchmark	0.21 Specificity (Difficult Benchmark) fraction · higher Uncertainty: ± 0.19 (type not stated). Printed on the line below the row; the caption says results are aggregated across prompting strategies but does not name the statistic. Table 1 uses mean and standard deviation for the same kind of aggregation. Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Evidence concern: excluded from comparisons Appendix A.2 describes the random baseline as predicting hit with probability equal to 'the overall proportion of actual hits in the specific benchmark dataset'. With one pooled rate (485 of 1,175 = 41.3%, Table 3) the expected PPV, sensitivity, F1 and FPR are all about 0.41 and specificity and NPV about 0.59. Table 8 prints PPV 0.47, sensitivity 0.46, F1 0.47, FPR 0.36, specificity 0.64 and NPV 0.63. With each screen's own hit rate (Table 3) the expected values are 0.47, 0.47, 0.47, 0.37, 0.63 and 0.63, which match. One draw at the pooled rate would need sensitivity about 2 standard errors above and FPR about 3 below expectation at once. The baseline therefore probably uses per-screen hit rates, which is information a screen designer would not have before the screen. The expected values were computed by the reviewer from Table 3 counts and are not stored as results. Treat the random row as a per-screen-prevalence classifier, not a matched random-selection comparator.Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Appendix A.2; Table 3; Table 1, row "Random Baseline"; Table 8, row "Random N/A N/A" Methods, coverage and source GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark pheno-20261022-protocol-song2025-difficult-hit-classification Aggregation: Not reported Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Table 2, row "GPT-4o (prompting)", column "Spec."

Source checking is not independent reproduction. Release 2026-10-23-5347695500ed.

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Baseline coverage

Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.

0 of 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.

No execution recipe linked to this protocol. Recipe availability does not establish a completed evaluation.

Author-reported evaluations: 2
External evaluations: 1

Literature evidence is not a Rewire measurement. Executed but unpublished runs and private review status are not included.

Null control

Proposed control: requires review

Select a task-valid null control after reviewing inputs and metric

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Conventional reference

Proposed control: requires review

Select an upstream conventional reference after reviewing the full protocol

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Protocol coverage CSV (gzip) · Model evaluation matrix (gzip) · Source table (gzip) · Release and checksums (gzip)

Coverage is derived from release 2026-10-23-5347695500ed. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run instructions

No runnable recipe has been reviewed for this protocol. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

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

Virtual CRISPR: Can LLMs Predict CRISPR Screen Results? · Original source · Proceedings of the 24th Workshop on Biomedical Language Processing (BioNLP 2025), pages 354-364, 1 August 2025; ACL Anthology 2025.bionlp-1.30

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: pheno-20261022-protocol-song2025-difficult-hit-classification

areas: cells-tissues
contexts: research
protocol: Same query form as the Simple Benchmark. The embedding classifier outputs a probability, binarised with Youden's J statistic, and is scored with AUROC, AUPRC, F1, FPR, PPV, NPV, sensitivity and specificity. GPT-4o is scored from its binary answers, averaged across prompting strategies; AUROC and AUPRC are undefined for it.
task: Binary hit/no-hit prediction for gene-phenotype queries from CRISPR screens
version: Song et al., BioNLP 2025, Sections 2.3, 3 and 4.2
metric: f1-score
metric direction: higher
unit: fraction
metric definition: As for the Simple Benchmark; AUROC and AUPRC are computed from the classifier's predicted probabilities.
limitations: Binary classification of pre-selected gene-phenotype queries; there is no candidate pool, ranking step or test budget, so it measures hit prediction rather than which perturbations to test.; Only genes with a statistically significant effect in either direction are queried; genes with no effect, which make up most of a screen, are not negatives.; Retrospective: labels come from published screens and no new experiment was run.; Labels are balanced by construction: every gene appears once per phenotype direction with opposite labels, so the benchmark tests whether the direction of effect is predicted.; Two screens from two publications, both in mouse cells; the authors call the test set constrained by the availability of suitable post-cutoff screens.; No random-baseline row is printed for this benchmark.; The Youden's J threshold is chosen on predicted probabilities; the paper does not say whether it was set on the validation set or on the test set. Threshold-dependent metrics (F1, FPR, PPV, NPV, sensitivity, specificity) may be optimistic for the embedding classifier; AUROC and AUPRC are not affected.
source locator: Sections 2.3, 3 and 4.2; Table 2; Appendix A.3 and Table 4; Limitations

Related records

uses data: Virtual CRISPR Difficult Benchmark
assessment: Raw Emb. on the Virtual CRISPR Difficult Benchmark
assessment: Summ. Emb. on the Virtual CRISPR Difficult Benchmark
assessment: GPT-4o (prompting) on the Virtual CRISPR Difficult Benchmark

Suggest a correction