{"id":"W4399151368","doi":"10.48550/arxiv.2405.17631","title":"BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Evolutionary Algorithms and Applications","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Wu Tsai Neurosciences Institute, Stanford University; Genentech; Defense Advanced Research Projects Agency; Institute for Catastrophic Loss Reduction; National Science Foundation","keywords":"Computer science; Perturbation (astronomy); Artificial intelligence; Physics; Quantum mechanics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002559232,0.001170808,0.0006804791,0.0006818959,0.0005498203,0.001104592,0.002345008,0.001377149,0.007504977],"category_scores_gemma":[0.007047436,0.0006401574,0.0009549782,0.0002638806,0.000943898,0.001657771,0.001583859,0.001966139,0.001856547],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001004445,"about_ca_system_score_gemma":0.002095273,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002824895,"about_ca_topic_score_gemma":0.005565598,"domain_scores_codex":[0.9993373,0.0002326419,0.00005307505,0.0001703113,0.0001629015,0.00004383327],"domain_scores_gemma":[0.9976152,0.001483151,0.0001856106,0.0003233429,0.0002618183,0.0001308408],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001307633,0.0009568611,0.009714692,0.001349327,0.000470649,0.000877444,0.0007052047,0.4814951,0.05100166,0.06214966,0.05969984,0.3302719],"study_design_scores_gemma":[0.00009642148,0.0001098474,0.0002504525,0.00002971205,0.00004127104,0.00007565971,0.0000391249,0.9573887,0.009699307,0.01043215,0.02181024,0.00002712013],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01934894,0.0004188478,0.9253454,0.0009573789,0.0001996035,0.0006399507,0.0009043782,0.04043478,0.01175068],"genre_scores_gemma":[0.1687586,0.0003292696,0.8183392,0.001066227,0.00004726211,0.001245125,0.001582221,0.001484766,0.007147466],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007504977,"threshold_uncertainty_score":0.02510661,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09494131121400051,"score_gpt":0.2310760300847567,"score_spread":0.1361347188707561,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}