{"id":"W4410501150","doi":"10.1002/anie.202503259","title":"Mitigating Molecular Aggregation in Drug Discovery With Predictive Insights From Explainable AI","year":2025,"lang":"en","type":"article","venue":"Angewandte Chemie International Edition","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Winnipeg; University of Manitoba","funders":"Canadian Institutes of Health Research; Bundesministerium für Bildung und Forschung","keywords":"Leverage (statistics); Counterfactual conditional; Drug discovery; Identification (biology); False positive paradox; Computer science; Machine learning; Artificial intelligence; Data science; Chemistry; Counterfactual thinking; Psychology; Biology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002229822,0.0001544302,0.0001536616,0.0001664597,0.00009956455,0.0002999595,0.0003367709,0.00006100177,0.0002286519],"category_scores_gemma":[0.0002798388,0.0001373332,0.00002975427,0.0002443232,0.0001178633,0.001424523,0.0001055324,0.0001614291,0.00002483075],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00025651,"about_ca_system_score_gemma":0.0000982213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001097893,"about_ca_topic_score_gemma":0.0001576092,"domain_scores_codex":[0.9985403,0.00007124159,0.000286905,0.0004477841,0.0004712854,0.0001824605],"domain_scores_gemma":[0.9993087,0.0001376898,0.0001671448,0.0001777687,0.0001738199,0.00003486193],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001116921,0.00005782846,0.001875935,0.00001963986,0.00001178092,0.00002896124,0.0005594983,0.004224248,0.9896698,0.002421073,0.0009339555,0.00008559054],"study_design_scores_gemma":[0.0005284694,0.00002218265,0.003587428,0.0004580522,0.00001005772,0.0000022545,0.0002492861,0.002673861,0.9797838,0.01191421,0.0006275237,0.0001428377],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9355153,0.00007134052,0.05401003,0.001252671,0.001357021,0.0001767523,0.00006770255,0.0000812827,0.007467915],"genre_scores_gemma":[0.9959224,0.000009423865,0.002262476,0.0006094759,0.0003017644,0.0001031355,0.0003144771,0.00001113193,0.0004656599],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06040717,"threshold_uncertainty_score":0.5600286,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.003727183954696887,"score_gpt":0.2374616992526132,"score_spread":0.2337345152979163,"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."}}