{"id":"W4415147619","doi":"10.1038/s41467-025-64177-5","title":"Empathi: embedding-based phage protein annotation tool by hierarchical assignment","year":2025,"lang":"en","type":"article","venue":"Nature Communications","topic":"Bacteriophages and microbial interactions","field":"Environmental Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval; Canadian Nutrition Society","funders":"Fonds de Recherche du Québec - Santé; Natural Sciences and Engineering Research Council of Canada; Fonds de recherche du Québec – Nature et technologies; Alliance de recherche numérique du Canada","keywords":"Annotation; Homology (biology); Phage display; ENCODE; Protein sequencing; Repertoire; Classifier (UML); Protein function; Genome","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":[0.001249329,0.001666088,0.0007373168,0.003211161,0.0007399427,0.001403182,0.001580932,0.001425978,0.006709483],"category_scores_gemma":[0.003964938,0.0005628201,0.001315969,0.001825778,0.0004076357,0.002255589,0.002301954,0.00169436,0.005324416],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005830439,"about_ca_system_score_gemma":0.000838565,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001524831,"about_ca_topic_score_gemma":0.001954985,"domain_scores_codex":[0.9991722,0.000140333,0.00006333776,0.0002992231,0.0002391747,0.00008570808],"domain_scores_gemma":[0.9990227,0.0004143515,0.000118197,0.000191602,0.0001544601,0.00009873544],"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":[0.002047558,0.0007747319,0.02353153,0.002220207,0.0005130338,0.0008620578,0.0008355012,0.02183338,0.07826601,0.008895264,0.2087865,0.6514342],"study_design_scores_gemma":[0.0002706563,0.0006429375,0.01268513,0.0002252166,0.0001724376,0.001195325,0.0004040494,0.7609028,0.06518956,0.0292143,0.1288901,0.0002075837],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07548995,0.00131087,0.6023672,0.0005708226,0.0003808791,0.0003829073,0.05070002,0.2630409,0.005756448],"genre_scores_gemma":[0.2136681,0.0007051731,0.6510374,0.0004419351,0.0001280952,0.0009589685,0.1179482,0.008173048,0.006939016],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006709483,"threshold_uncertainty_score":0.02244544,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007108957200631578,"score_gpt":0.2958185854841351,"score_spread":0.2887096282835035,"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."}}