{"id":"W4200623292","doi":"10.33774/chemrxiv-2021-9nhm1-v2","title":"Sanitize It Yourself: Web-based molecular sanitization for machine-generated chemical structures","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Chemical space; Web service; Software; Space (punctuation); Distributed computing; World Wide Web; Drug discovery; Bioinformatics; Operating system","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":[0.001778426,0.0008321453,0.0007564224,0.0009668809,0.0005229882,0.001048603,0.001574852,0.00110506,0.01685962],"category_scores_gemma":[0.004201982,0.000443476,0.0008682308,0.0007355434,0.0007821259,0.001322155,0.001952786,0.001719967,0.004596483],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005142367,"about_ca_system_score_gemma":0.0009834002,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008578181,"about_ca_topic_score_gemma":0.001296407,"domain_scores_codex":[0.9994086,0.0001539605,0.0000310165,0.00006486996,0.0002910857,0.00005056358],"domain_scores_gemma":[0.9984862,0.0006543971,0.0001179172,0.0004818558,0.0001870982,0.00007242744],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001972747,0.001486124,0.003554162,0.0009142405,0.000353957,0.0007701842,0.0003296185,0.2214012,0.0722111,0.04253668,0.05623139,0.5982386],"study_design_scores_gemma":[0.0004101065,0.0002207863,0.0003051387,0.00004735702,0.00004380944,0.0001702494,0.00002428736,0.8896649,0.0615834,0.01998849,0.02750443,0.00003693708],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07003777,0.0009479597,0.7919365,0.0007609991,0.0003323633,0.0005641528,0.00170395,0.1172704,0.01644605],"genre_scores_gemma":[0.3499311,0.001065447,0.6202402,0.0007951513,0.0001165328,0.0008775967,0.005387296,0.01051484,0.01107195],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01685962,"threshold_uncertainty_score":0.05640107,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02879964684128298,"score_gpt":0.3135630109130629,"score_spread":0.2847633640717799,"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."}}