{"id":"W4415517900","doi":"10.1186/s13321-025-01104-2","title":"VNFlow: integration of variational autoencoders and normalizing flows for novel molecular design","year":2025,"lang":"en","type":"article","venue":"Journal of Cheminformatics","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Defence Research and Development Canada; National Research Council Canada","funders":"National Research Council Canada; Defence Research and Development Canada","keywords":"Generative grammar; Key (lock); Inverse; Range (aeronautics); Sampling (signal processing); Generative model; Inverse problem; Chemical space","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001894864,0.00009405724,0.0002331777,0.0001423177,0.00006425909,0.00009000773,0.0002519023,0.00005584731,0.00002887346],"category_scores_gemma":[0.001282755,0.00007590708,0.00005188013,0.000127932,0.00005961798,0.0005241969,0.00004781344,0.00008512984,8.918957e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004144602,"about_ca_system_score_gemma":0.0001990902,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003033836,"about_ca_topic_score_gemma":2.95518e-7,"domain_scores_codex":[0.9987095,0.00001519872,0.0007846018,0.0000589667,0.0003003756,0.0001313577],"domain_scores_gemma":[0.9984316,0.0003035891,0.0007477268,0.0001002691,0.0003755798,0.00004122661],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0000523574,0.0000204866,0.00001541515,0.0001765892,0.00000963911,2.934797e-7,0.0008842716,0.05089595,0.9441741,0.002856659,0.0003207279,0.0005935193],"study_design_scores_gemma":[0.0004519081,0.00007391313,0.00009418636,0.000156755,0.00003279199,0.00002315474,0.0001508222,0.4154783,0.5816693,0.00157962,0.0002258421,0.00006332162],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.125631,0.00003284385,0.87323,0.0001870204,0.0003225364,0.0001207057,0.000006824235,0.000007024952,0.0004619859],"genre_scores_gemma":[0.2387244,0.000007175257,0.761066,0.0001328413,0.00002878172,0.000003537681,0.000001719955,0.000004199567,0.00003138454],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.3645824,"threshold_uncertainty_score":0.3095401,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0154423853595157,"score_gpt":0.2801444683022938,"score_spread":0.2647020829427781,"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."}}