{"id":"W4413290922","doi":"10.1038/s42256-025-01088-6","title":"Boosting the predictive power of protein representations with a corpus of text annotations","year":2025,"lang":"en","type":"article","venue":"Nature Machine Intelligence","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"Vector Institute; University of Toronto","funders":"Vector Institute; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada","keywords":"Boosting (machine learning); Predictive power; Computer science; Natural language processing; Artificial intelligence; Physics","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":[0.00013679,0.00008033296,0.000103817,0.00004418878,0.00005884055,0.000005177458,0.0002297963,0.0001460527,0.000008733614],"category_scores_gemma":[0.0009395414,0.00004808392,0.00004273843,0.0002919702,0.0003517322,0.000001763814,0.00007621475,0.0002481591,4.048424e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000004243492,"about_ca_system_score_gemma":0.00009014233,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005500358,"about_ca_topic_score_gemma":0.000055436,"domain_scores_codex":[0.9993718,0.00004861964,0.0001900483,0.00017723,0.0001201282,0.00009216147],"domain_scores_gemma":[0.9992922,0.00009746958,0.0001183519,0.0002701937,0.0002046052,0.00001721344],"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":[0.00223779,0.0008529848,0.08150557,0.0003797465,0.001220643,0.00001486094,0.003647037,0.002707012,0.5621074,0.03323651,0.005070467,0.3070199],"study_design_scores_gemma":[0.0002229764,0.0008116686,0.01937882,0.0002482159,0.00005497137,0.00001270559,0.001467976,0.0006897225,0.9674702,0.002051007,0.00745109,0.0001405898],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6058798,0.01396238,0.3593812,0.003585322,0.0002613946,0.0009811833,0.0001832495,0.00004253655,0.01572284],"genre_scores_gemma":[0.9957582,0.00002868037,0.003460438,0.0001216882,0.00001498597,0.00002625326,0.00002081104,0.000004461765,0.0005644708],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4053628,"threshold_uncertainty_score":0.1960805,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005744969379399131,"score_gpt":0.2958628101583177,"score_spread":0.2901178407789186,"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."}}