{"id":"W4402441163","doi":"10.1038/s41467-024-52060-8","title":"Automated design of multi-target ligands by generative deep learning","year":2024,"lang":"en","type":"article","venue":"Nature Communications","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Innovative Medicines Initiative; European Commission; Deutsche Forschungsgemeinschaft; European Federation of Pharmaceutical Industries and Associations; Diamond Light Source; McGill University","keywords":"Pharmacophore; Computer science; Drug discovery; Generative model; Small molecule; String (physics); Generative grammar; Similarity (geometry); Deep learning; Dual (grammatical number); Computational biology; Artificial intelligence; Chemistry; Bioinformatics; Biology; Stereochemistry; Physics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006568435,0.0001265754,0.0001509403,0.0001634023,0.0002144646,0.0001589694,0.001855613,0.0001513696,0.000005711223],"category_scores_gemma":[0.0003293418,0.0001209151,0.00006876313,0.0009652625,0.00009053041,0.0004005599,0.0006178067,0.0008504388,0.00001944865],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004871624,"about_ca_system_score_gemma":0.0001535255,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006285491,"about_ca_topic_score_gemma":0.000006029473,"domain_scores_codex":[0.9985504,0.0005663275,0.000256273,0.0002599289,0.0002224343,0.0001446308],"domain_scores_gemma":[0.996954,0.001715911,0.00008118161,0.000997757,0.0002003402,0.00005084688],"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":[0.000007800289,0.0004327936,0.0002819337,0.00005390412,0.0002902852,0.000006849488,0.006610156,0.6968082,0.03626836,0.2112548,0.01706497,0.03091999],"study_design_scores_gemma":[0.00009326157,0.00002659513,0.0003630934,0.0000303666,0.00000876673,0.000005649562,0.00002454661,0.9801741,0.007741737,0.001542402,0.00986933,0.00012014],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0003868783,0.02944877,0.9665186,0.002132634,0.0002251686,0.0001696686,0.00001159176,0.0007676214,0.0003390527],"genre_scores_gemma":[0.5038525,0.0001802135,0.495703,0.00009165503,0.00001048568,0.00002235336,0.00004407625,0.000009285355,0.00008632994],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5034657,"threshold_uncertainty_score":0.4930774,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03808914279555357,"score_gpt":0.3636508624338755,"score_spread":0.3255617196383219,"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."}}