{"id":"W4225380763","doi":"10.1101/2022.04.22.487554","title":"GeneVector: Identification of transcriptional programs using dense vector representations defined by mutual information","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Single-cell and spatial transcriptomics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Dimensionality reduction; Embedding; Curse of dimensionality; Computer science; Mutual information; Scalability; Computational biology; Leverage (statistics); Gene; Principal component analysis; Identification (biology); Feature vector; Artificial intelligence; Data mining; Biology; Genetics","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":[0.0007042832,0.0007333429,0.0007128151,0.001061063,0.000303838,0.001107469,0.0007501703,0.0005056973,0.002118472],"category_scores_gemma":[0.001379768,0.000388319,0.0007076534,0.001036271,0.0005953109,0.000917626,0.0008583676,0.000870839,0.0006603801],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008324296,"about_ca_system_score_gemma":0.0008446464,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003044578,"about_ca_topic_score_gemma":0.003992206,"domain_scores_codex":[0.9995958,0.00008345261,0.00001813342,0.0001542787,0.0001150167,0.00003337438],"domain_scores_gemma":[0.9995472,0.0001958181,0.00007586391,0.00006675789,0.0000804919,0.0000337961],"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":[0.001166818,0.0003717132,0.01103974,0.001033429,0.0004420532,0.0003246318,0.0005711539,0.3408555,0.1953093,0.05407034,0.03375325,0.361062],"study_design_scores_gemma":[0.00003368812,0.00005751319,0.001856021,0.00001370219,0.00001533675,0.00004533781,0.00003963117,0.9646983,0.01434517,0.01488903,0.003976764,0.00002951949],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08327752,0.000454929,0.8993017,0.0002699128,0.00005623758,0.0001358987,0.006139609,0.009361786,0.001002379],"genre_scores_gemma":[0.4446438,0.0004292313,0.5355457,0.0001520609,0.00005118459,0.000505664,0.01449816,0.001268991,0.002905236],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003044578,"threshold_uncertainty_score":0.007086992,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01833423391209078,"score_gpt":0.2317764538433078,"score_spread":0.213442219931217,"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."}}