{"id":"W4321498777","doi":"10.21203/rs.3.rs-2604981/v1","title":"Something for Nothing: Improved Solvation Free Energy Prediction with Δ-Learning","year":2023,"lang":"en","type":"preprint","venue":"Research Square","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; Canada Research Chairs; Canarie","keywords":"Solvation; Gaussian; Regression; Solubility; Linear regression; Gaussian process; Kriging; Bioavailability; Computer science; Machine learning; Artificial intelligence; Statistical physics; Chemistry; Computational chemistry; Mathematics; Molecule; Statistics; Physics; Physical chemistry; Bioinformatics; Organic chemistry","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002616806,0.0006269842,0.001048217,0.0008084213,0.0003058002,0.0007251819,0.001816237,0.0009971828,0.001450785],"category_scores_gemma":[0.006159121,0.0002806912,0.0005558167,0.0007381598,0.0005754157,0.001447376,0.001000088,0.001124783,0.0002829487],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007607065,"about_ca_system_score_gemma":0.001051817,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007640286,"about_ca_topic_score_gemma":0.004385056,"domain_scores_codex":[0.9994092,0.0002866922,0.00003022587,0.0001237593,0.0001031394,0.00004689848],"domain_scores_gemma":[0.9966995,0.002256689,0.0001695307,0.0003133436,0.0004093527,0.0001515414],"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":[0.0002438296,0.0001931374,0.002179816,0.0000449424,0.00005523973,0.00002909029,0.00001416913,0.9643074,0.0008025978,0.001253629,0.001037777,0.02983833],"study_design_scores_gemma":[0.000006260801,0.00001012196,0.0000433,0.000001014643,0.000001529698,0.000001262021,0.000001158599,0.9993343,0.0001878215,0.0003880539,0.000023661,0.000001521005],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.6955671,0.001383458,0.2939699,0.00123319,0.0001335807,0.00007855069,0.0005426015,0.003609382,0.003482201],"genre_scores_gemma":[0.9468445,0.00009455359,0.05159812,0.0001609696,0.00002309912,0.00003381342,0.0004077929,0.0001213928,0.0007156889],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.007640286,"threshold_uncertainty_score":0.01519161,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1217663387769186,"score_gpt":0.4140187468894607,"score_spread":0.2922524081125421,"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."}}