{"id":"W4382344242","doi":"10.1186/s40168-023-01562-6","title":"MetaPro: a scalable and reproducible data processing and analysis pipeline for metatranscriptomic investigation of microbial communities","year":2023,"lang":"en","type":"article","venue":"Microbiome","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Hospital for Sick Children","funders":"Canadian Institutes of Health Research; University of Toronto","keywords":"Pipeline (software); Scalability; Computer science; Visualization; Modular design; Sequence assembly; Annotation; Benchmark (surveying); Metagenomics; Data mining; Computational biology; Biology; Information retrieval; Artificial intelligence; Database; Gene; Cartography","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":[0.00637031,0.003754538,0.001788475,0.003978355,0.001656937,0.004094862,0.005051031,0.001792244,0.005971119],"category_scores_gemma":[0.009419932,0.002059617,0.003102435,0.002855372,0.0012569,0.003714509,0.00566019,0.004769275,0.007016286],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001098995,"about_ca_system_score_gemma":0.00456237,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00242893,"about_ca_topic_score_gemma":0.003176544,"domain_scores_codex":[0.9964154,0.0004986578,0.0003361923,0.001136206,0.001324432,0.0002891776],"domain_scores_gemma":[0.9957098,0.001071973,0.0005749059,0.001271028,0.0008296967,0.0005426916],"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":[0.004640587,0.0008572073,0.01901335,0.003815572,0.002287666,0.001575876,0.001760218,0.03157081,0.3721772,0.01187025,0.2161594,0.3342719],"study_design_scores_gemma":[0.0009778579,0.0009982438,0.01923855,0.0005414427,0.0005563484,0.001463297,0.0004282393,0.3944251,0.3372971,0.03642338,0.2065448,0.001105746],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01811611,0.0007989296,0.6374509,0.0006733114,0.0003849654,0.0008755003,0.03342486,0.305782,0.002493537],"genre_scores_gemma":[0.07915065,0.0007121837,0.8119587,0.0007755263,0.0002054258,0.002347039,0.07675254,0.02569048,0.002407383],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00637031,"threshold_uncertainty_score":0.0336898,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05093893882885635,"score_gpt":0.2775620213990623,"score_spread":0.226623082570206,"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."}}