{"id":"W4411111057","doi":"10.1101/2025.06.03.657578","title":"Integrating multi-covariate disentanglement with counterfactual analysis on synthetic data enables cell type discovery and counterfactual predictions","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Cell Image Analysis Techniques","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research","funders":"NIHR Newcastle Biomedical Research Centre; Lister Institute of Preventive Medicine; National Institute for Health and Care Research; Wellcome Trust","keywords":"Counterfactual thinking; Covariate; Econometrics; Computer science; Type (biology); Statistics; Data mining; Mathematics; Psychology; Biology","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004101781,0.0006232516,0.0006220512,0.0003853931,0.0001985769,0.0004713516,0.0008340533,0.0003587083,0.00002412208],"category_scores_gemma":[0.0001613819,0.0005560257,0.0001370869,0.0005251167,0.0001787919,0.00003973502,0.001511833,0.0004796188,0.00000393761],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001351824,"about_ca_system_score_gemma":0.0004662266,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002175832,"about_ca_topic_score_gemma":0.0001270162,"domain_scores_codex":[0.9968624,0.0001500597,0.0005109676,0.001741,0.0003318361,0.000403766],"domain_scores_gemma":[0.9962052,0.00005396789,0.0003885034,0.00282757,0.0003879123,0.000136886],"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":[0.000331407,0.0008131906,0.01590219,0.0004141603,0.005165138,0.0000349139,0.00003365879,0.001206362,0.9740971,0.00003740218,0.00195881,0.000005650933],"study_design_scores_gemma":[0.000870514,0.0004521049,0.009360438,0.0005879634,0.005890781,4.977306e-8,0.00006791716,0.01728881,0.958779,4.655813e-7,0.005372941,0.001329037],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8584943,0.001366213,0.1360462,0.00005711251,0.0002098645,0.000848191,0.002738787,0.0001859675,0.0000533574],"genre_scores_gemma":[0.9923636,0.001001114,0.006053589,0.0001459455,0.000131825,0.00008558221,0.00006814866,0.00006623039,0.00008397899],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1338693,"threshold_uncertainty_score":0.9996891,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01624914438009599,"score_gpt":0.2541763764022088,"score_spread":0.2379272320221128,"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."}}