{"id":"W3109912085","doi":"10.1101/2020.12.02.406710","title":"Newfound coding potential of transcripts unveils missing members of human protein communities","year":2020,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Bioinformatics and Genomic Networks","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"PROTEO; Université de Sherbrooke","funders":"Canadian Institutes of Health Research; Fonds de Recherche du Québec - Santé; Compute Canada; Université de Sherbrooke","keywords":"Proteogenomics; ENCODE; Pseudogene; Computational biology; Open reading frame; Biology; Gene; Proteomics; Genetics; Genome; Genomics; Peptide sequence","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":[0.0003966406,0.0002137644,0.0002511025,0.001947306,0.0003928995,0.0006702116,0.000190886,0.000283274,0.002352964],"category_scores_gemma":[0.001942551,0.0001563042,0.0002276442,0.001706955,0.0003839118,0.0007366437,0.000530834,0.000283674,0.0004537453],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002277636,"about_ca_system_score_gemma":0.0001750665,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006339472,"about_ca_topic_score_gemma":0.0009762811,"domain_scores_codex":[0.9996095,0.00006314734,0.00001754401,0.000185151,0.00007522054,0.00004945741],"domain_scores_gemma":[0.9983797,0.000866865,0.0002900856,0.0001587868,0.0001366676,0.0001680185],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002308462,0.0001135554,0.3717192,0.0005332222,0.00026944,0.002000153,0.0009590286,0.007872423,0.5388657,0.01054407,0.002300279,0.06251437],"study_design_scores_gemma":[0.00008228655,0.0002675434,0.6875032,0.0001217819,0.0004593963,0.006518914,0.001919316,0.1119462,0.1025377,0.06853507,0.02002307,0.00008536105],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9841105,0.0007095006,0.01086438,0.0001756159,0.00001378257,0.000004384922,0.002954136,0.0001026569,0.001065088],"genre_scores_gemma":[0.9927938,0.0001995532,0.00410145,0.00002655433,0.00002475294,0.000006049753,0.002515854,0.00002347935,0.0003085515],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002352964,"threshold_uncertainty_score":0.007871389,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01923834607685197,"score_gpt":0.2266760145594615,"score_spread":0.2074376684826095,"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."}}