{"id":"W4414484211","doi":"10.1101/2025.09.22.677955","title":"Illuminating the Virosphere’s Dark Matter using Hierarchical Deep Learning","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Dark Matter and Cosmic Phenomena","field":"Physics and Astronomy","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Metagenomics; Genome; Human virome; Deep learning; Virus classification; Viral evolution; Genomics; Homology (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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006040279,0.0006869776,0.0005073234,0.001005377,0.0004722272,0.0012426,0.001007824,0.000778696,0.0008863701],"category_scores_gemma":[0.001349466,0.0003333741,0.0006879127,0.0006870823,0.001331393,0.001871239,0.001802285,0.001625491,0.0003723929],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001110993,"about_ca_system_score_gemma":0.0008926488,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005463991,"about_ca_topic_score_gemma":0.007045391,"domain_scores_codex":[0.9997059,0.00006904745,0.000009648155,0.00008204459,0.00006388141,0.00006949104],"domain_scores_gemma":[0.9994399,0.0002305057,0.00008608719,0.00009482026,0.00007517794,0.0000734667],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004593453,0.0002617383,0.03669529,0.0004969648,0.0003092206,0.0004669443,0.0007892633,0.5399115,0.0706643,0.07794583,0.01218273,0.2598168],"study_design_scores_gemma":[0.000006767249,0.0000322443,0.001221102,0.00002123315,0.00001934381,0.00003593101,0.00007309654,0.950733,0.004280547,0.04193375,0.001630165,0.00001292779],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3558949,0.003458505,0.62942,0.00192915,0.0001153789,0.00004304996,0.001416945,0.002606478,0.005115538],"genre_scores_gemma":[0.9444606,0.0007259029,0.0509884,0.0005827179,0.00006121446,0.00002356061,0.001689894,0.00009108176,0.001376576],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005463991,"threshold_uncertainty_score":0.01086438,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008449638257607589,"score_gpt":0.2170313288814933,"score_spread":0.2085816906238857,"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."}}