{"id":"W4289261436","doi":"10.1101/2022.08.01.502183","title":"HSDatabase – a database of highly similar duplicate genes from plants, animals, and algae","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University; Dalhousie University","funders":"","keywords":"Genome; Gene; Biology; KEGG; Database; Computational biology; Gene duplication; Gene Annotation; Genetics; Computer science; Gene ontology; Gene expression","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":[0.001253586,0.002151552,0.002487881,0.009330292,0.001573601,0.002197961,0.002491373,0.001338336,0.01406126],"category_scores_gemma":[0.002747244,0.001057066,0.001116272,0.0120774,0.0004823244,0.002125693,0.002132305,0.001368732,0.01788387],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008077895,"about_ca_system_score_gemma":0.003178766,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002393365,"about_ca_topic_score_gemma":0.002181645,"domain_scores_codex":[0.9991196,0.00006807803,0.0002316332,0.0002786672,0.0002263137,0.00007568191],"domain_scores_gemma":[0.9986132,0.0002076649,0.0003342035,0.0002731709,0.0003048055,0.0002669778],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.004955081,0.0005447066,0.02099481,0.02564946,0.00102718,0.003028564,0.00202804,0.003154129,0.3008716,0.00876525,0.4230145,0.2059666],"study_design_scores_gemma":[0.0003474267,0.000315945,0.02649156,0.0008606481,0.0005342973,0.002270249,0.0003941702,0.002411897,0.03116361,0.00393406,0.9310634,0.0002128702],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0338995,0.008105397,0.01489628,0.0001858301,0.0003655039,0.0003031934,0.91303,0.02206688,0.007147363],"genre_scores_gemma":[0.009923038,0.00142444,0.01916821,0.00006557942,0.00003978602,0.0001742204,0.9672659,0.0009635485,0.000975271],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01406126,"threshold_uncertainty_score":0.04703963,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01474586987137131,"score_gpt":0.2181243133247198,"score_spread":0.2033784434533485,"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."}}