{"id":"W4220903645","doi":"10.26434/chemrxiv-2022-x647j","title":"MACAW: an accessible tool for molecular embedding and inverse molecular design","year":2022,"lang":"en","type":"preprint","venue":"ChemRxiv","topic":"Chemical Synthesis and Analysis","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Lawrence Berkeley National Laboratory; Sandia National Laboratories; U.S. Department of Energy; Laboratory Directed Research and Development; National Nuclear Security Administration; Central South University; University of Toronto","keywords":"chEMBL; Computer science; Embedding; Chemical space; Theoretical computer science; Algorithm; Artificial intelligence; Drug discovery; Bioinformatics; Biology","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":[0.002200164,0.002711524,0.001485895,0.001798735,0.0007468634,0.002114263,0.002894034,0.001971556,0.04588166],"category_scores_gemma":[0.00526239,0.001422247,0.002173539,0.001202377,0.0007374982,0.002122415,0.002815057,0.002926198,0.01532083],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000739856,"about_ca_system_score_gemma":0.001343514,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001156347,"about_ca_topic_score_gemma":0.00223968,"domain_scores_codex":[0.99915,0.0001655874,0.00006736888,0.0001826645,0.0003663684,0.00006791903],"domain_scores_gemma":[0.9982511,0.001130998,0.0001664628,0.0002673528,0.0001112438,0.00007272417],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00104722,0.0004626662,0.002912714,0.005456344,0.0007629022,0.001199071,0.0004843991,0.08540884,0.05404243,0.0977321,0.2535878,0.4969034],"study_design_scores_gemma":[0.0006760189,0.0002846283,0.0007656255,0.000504586,0.0001807755,0.0006184062,0.00008519018,0.3880248,0.04825851,0.109413,0.4510046,0.0001837753],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002844967,0.001161776,0.7968526,0.0003025386,0.000239135,0.0002304786,0.009268893,0.1828837,0.006215933],"genre_scores_gemma":[0.04236894,0.0020601,0.895101,0.0006807416,0.0001110774,0.002054708,0.02145581,0.0292305,0.006937107],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04588166,"threshold_uncertainty_score":0.1534894,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0240658272084067,"score_gpt":0.2991952824512442,"score_spread":0.2751294552428374,"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."}}