{"id":"W6902049475","doi":"10.6084/m9.figshare.26600317.v1","title":"Additional file 10 of MetaPro: a scalable and reproducible data processing and analysis pipeline for metatranscriptomic investigation of microbial communities","year":2024,"lang":"en","type":"dataset","venue":"Figshare","topic":"Diptera species taxonomy and behavior","field":"Agricultural and Biological Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; SickKids Foundation; Hospital for Sick Children","funders":"","keywords":"Pipeline (software); Table (database); Scalability; Set (abstract data type); Data set; Data processing","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002332614,0.002417409,0.001506448,0.003067124,0.001078958,0.002602634,0.003235473,0.001969075,0.3774222],"category_scores_gemma":[0.01143345,0.0008771134,0.001858666,0.004139034,0.0005207038,0.002242648,0.002303585,0.001900602,0.1324133],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00126289,"about_ca_system_score_gemma":0.002391307,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007626857,"about_ca_topic_score_gemma":0.01722436,"domain_scores_codex":[0.9988294,0.0001472433,0.0001651511,0.0004188224,0.0002709592,0.000168365],"domain_scores_gemma":[0.9955904,0.002340075,0.0003537145,0.0007408932,0.0006637746,0.0003111346],"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":[0.0002462786,0.00005907411,0.003107569,0.00222042,0.0001027247,0.00006863273,0.00005686359,0.000608208,0.0006041183,0.0005990824,0.987666,0.004661036],"study_design_scores_gemma":[0.001635135,0.0001287141,0.01481983,0.00101623,0.0001699879,0.000268094,0.0002273169,0.00236904,0.003013328,0.005823144,0.9703743,0.0001549197],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00008519383,0.00001296305,0.0001730241,0.00002614344,0.0000116427,0.00001643475,0.9987164,0.0007594308,0.0001988599],"genre_scores_gemma":[0.0009391059,0.00002777494,0.001461223,0.00007434322,0.00001106913,0.000198131,0.9962893,0.000444684,0.0005544742],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.3774222,"threshold_uncertainty_score":0.8880317,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1087905114344723,"score_gpt":0.2647005268028395,"score_spread":0.1559100153683672,"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."}}