{"id":"W4414084489","doi":"10.1038/s41551-025-01481-x","title":"Combinatorial prediction of therapeutic perturbations using causally inspired neural networks","year":2025,"lang":"en","type":"article","venue":"Nature Biomedical Engineering","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Harvard Data Science Initiative, Harvard University; Harvard University; AstraZeneca; Sanofi; Bill and Melinda Gates Foundation; Pfizer; Advanced Research Projects Agency; National Institutes of Health; National Science Foundation","keywords":"Embedding; Representation (politics); Artificial neural network; Graph; Perturbation (astronomy); Drug target; Clinical phenotype; Reversing","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004609347,0.0008734313,0.0005112916,0.0009673988,0.0001875338,0.0004497898,0.0007153899,0.00079411,0.001580755],"category_scores_gemma":[0.002262392,0.0003589015,0.0005615242,0.0004410885,0.0004598333,0.0006620797,0.0004177056,0.0008208053,0.0001772991],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001107488,"about_ca_system_score_gemma":0.000820188,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004134447,"about_ca_topic_score_gemma":0.007508047,"domain_scores_codex":[0.9998006,0.00006330167,0.000009206118,0.000054885,0.00005203727,0.00001996398],"domain_scores_gemma":[0.9991453,0.0005692759,0.0001222261,0.00004538301,0.00008223275,0.00003563472],"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":[0.00003425671,0.00003373919,0.001014152,0.0000346442,0.00002219704,0.00003290881,0.000004415981,0.9828714,0.001502046,0.002022514,0.0004542144,0.01197347],"study_design_scores_gemma":[0.000003726641,0.0000106517,0.00008084723,0.000001536064,0.000004092657,0.000004098109,8.434809e-7,0.9968566,0.000418768,0.002525676,0.0000915885,0.000001568942],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1898151,0.001045497,0.7986217,0.001111079,0.0000936662,0.000154541,0.001394003,0.002572271,0.005192195],"genre_scores_gemma":[0.9289837,0.0003681778,0.06757506,0.0002718835,0.00003226017,0.0001455529,0.0009878888,0.00007422086,0.001561266],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004134447,"threshold_uncertainty_score":0.008220792,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01050557665194205,"score_gpt":0.2734299059163521,"score_spread":0.2629243292644101,"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."}}