{"id":"W3112662112","doi":"10.1038/s42256-020-00232-8","title":"Integration of mechanistic immunological knowledge into a machine learning pipeline improves predictions","year":2020,"lang":"en","type":"article","venue":"Nature Machine Intelligence","topic":"Single-cell and spatial transcriptomics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":84,"is_retracted":false,"has_abstract":false,"ca_institutions":"Lockheed Martin (Canada); National Research Council Canada","funders":"Stanford Maternal and Child Health Research Institute; National Institute of Dental and Craniofacial Research; National Institute of Allergy and Infectious Diseases; National Institute of Neurological Disorders and Stroke; National Institute of General Medical Sciences; National Heart, Lung, and Blood Institute; National Institute on Aging; Burroughs Wellcome Fund; U.S. Food and Drug Administration; American Heart Association; National Institutes of Health; U.S. Department of Health and Human Services; March of Dimes Foundation; Hamilton Health Sciences Foundation; Bill and Melinda Gates Foundation; Robertson Foundation; Doris Duke Charitable Foundation","keywords":"Mass cytometry; Computer science; Overfitting; Machine learning; Artificial intelligence; Profiling (computer programming); Immune system; Immunology; Medicine; Artificial neural network; Biology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00213925,0.001773498,0.001172583,0.001712314,0.0006881331,0.002041167,0.001512299,0.001579381,0.007250737],"category_scores_gemma":[0.005121722,0.0006763805,0.002015401,0.0008474418,0.0004142231,0.002697675,0.001251804,0.002529198,0.003462658],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005933651,"about_ca_system_score_gemma":0.001236821,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001482472,"about_ca_topic_score_gemma":0.002707081,"domain_scores_codex":[0.9994853,0.00009554522,0.00002904856,0.0001922828,0.0001231571,0.00007456143],"domain_scores_gemma":[0.9973399,0.001653938,0.000176605,0.0003373309,0.0003591757,0.0001330594],"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":[0.001636878,0.002351624,0.06379231,0.0009608223,0.001031612,0.0008880956,0.0001568691,0.283566,0.1973438,0.009974603,0.01319516,0.4251024],"study_design_scores_gemma":[0.00005210315,0.0001809605,0.004609896,0.00003483605,0.0002774356,0.0001877114,0.00003563871,0.9381119,0.03143705,0.02188039,0.003152963,0.00003918733],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2382391,0.001338946,0.7262487,0.001954836,0.0003569107,0.0002965724,0.003173638,0.01579394,0.01259743],"genre_scores_gemma":[0.82532,0.0005540816,0.166094,0.0006270893,0.0001661689,0.000102137,0.004303857,0.0003655208,0.002467151],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007250737,"threshold_uncertainty_score":0.02425611,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0137246454107532,"score_gpt":0.2678726170943851,"score_spread":0.2541479716836319,"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."}}