{"id":"W4385349896","doi":"10.26434/chemrxiv-2023-jpwvn","title":"Regularized indirect learning improves phage display ligand discovery","year":2023,"lang":"en","type":"preprint","venue":"ChemRxiv","topic":"Monoclonal and Polyclonal Antibodies Research","field":"Medicine","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Novo Nordisk; Pharmaceutical Research and Manufacturers of America Foundation","keywords":"Random forest; Computational biology; Phage display; Artificial intelligence; Computer science; Machine learning; Peptide; Regularization (linguistics); Genetic programming; Fitness landscape; Peptide library; Biology; Peptide sequence; Genetics; Gene; Biochemistry","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":[0.0012702,0.0008150358,0.0009096144,0.000342108,0.0002040161,0.0006428304,0.0009982022,0.0008102545,0.001136978],"category_scores_gemma":[0.002410417,0.0003175206,0.0006469902,0.0002324635,0.0004972407,0.0008284036,0.0009859101,0.001312,0.0004639658],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006315311,"about_ca_system_score_gemma":0.0007511055,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0016208,"about_ca_topic_score_gemma":0.001976576,"domain_scores_codex":[0.9996217,0.0001179106,0.00001859975,0.0000864411,0.00009999258,0.00005530873],"domain_scores_gemma":[0.9993305,0.0003161761,0.00006759705,0.00009486153,0.0001458194,0.00004514888],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002840571,0.0002448763,0.002614982,0.0001026427,0.0001076396,0.00009659342,0.00005431686,0.865962,0.02778523,0.004237447,0.001024917,0.09748524],"study_design_scores_gemma":[0.000005778288,0.000055137,0.00007586971,0.000001579422,0.00000521047,0.000007234317,0.000001316346,0.996419,0.002535233,0.00076329,0.0001273609,0.000003107316],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4259883,0.001145282,0.5634358,0.0005351059,0.00007587288,0.00006285986,0.0001358383,0.004247033,0.004373907],"genre_scores_gemma":[0.9150627,0.0002471095,0.0805024,0.0002972362,0.00003072656,0.00007211826,0.0003513388,0.0001329578,0.003303363],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0016208,"threshold_uncertainty_score":0.006717563,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04396161294583192,"score_gpt":0.3259406145710054,"score_spread":0.2819790016251735,"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."}}