{"id":"W3130053421","doi":"10.1101/2021.02.23.432558","title":"MetaPro: A scalable and reproducible data processing and analysis pipeline for metatranscriptomic investigation of microbial communities","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Hospital for Sick Children","funders":"University of Toronto; Canada First Research Excellence Fund; Government of Ontario; Compute Canada; Hospital for Sick Children; Ministry of Agriculture, Food and Rural Affairs; Alberta Livestock and Meat Agency; Ontario Ministry of Agriculture, Food and Rural Affairs; Canadian Poultry Research Council","keywords":"Computer science; Pipeline (software); Scalability; Modular design; Visualization; Sequence assembly; Annotation; Data mining; Benchmark (surveying); Metagenomics; Computational biology; Information retrieval; Data science; Artificial intelligence; Database; Biology; 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":[0.005601908,0.003389023,0.001737015,0.003381814,0.00142693,0.004096214,0.004552951,0.001566524,0.007580322],"category_scores_gemma":[0.007690697,0.002122052,0.002839539,0.002584295,0.001079484,0.003498968,0.004783737,0.004484006,0.008227195],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009543479,"about_ca_system_score_gemma":0.003623923,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002077999,"about_ca_topic_score_gemma":0.002380513,"domain_scores_codex":[0.9968964,0.0004387778,0.000318035,0.0009893387,0.001101558,0.0002559242],"domain_scores_gemma":[0.9964013,0.0009921314,0.000446181,0.001074061,0.0006609244,0.0004255025],"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":[0.003825204,0.0006182468,0.01253101,0.003595991,0.00164021,0.001445973,0.001391386,0.0287142,0.3643456,0.01187841,0.2283852,0.3416286],"study_design_scores_gemma":[0.0007923681,0.0007716231,0.01396205,0.0004703148,0.0004007034,0.00131832,0.0003344023,0.3505583,0.3824755,0.03501984,0.2128756,0.001020831],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.01400912,0.0006566786,0.643605,0.0005831106,0.0003179281,0.0006279237,0.02594784,0.3121212,0.002131144],"genre_scores_gemma":[0.07232651,0.0006231546,0.8279637,0.0007215572,0.0001819389,0.002101843,0.06183902,0.03133537,0.002906869],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.007580322,"threshold_uncertainty_score":0.02962607,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03365124690847148,"score_gpt":0.2431432492097994,"score_spread":0.2094920023013279,"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."}}