{"id":"W4413477453","doi":"10.1101/2025.08.19.670115","title":"FAVABEAN and FALAPhyl: Open-Source Pipelines for Scalable 16s rRNA Microbiome Data Processing and Visualization","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Oral microbiology and periodontitis research","field":"Dentistry","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Nova Scotia Hospital; Dalhousie University; University of Alberta","funders":"University of Alberta; Dalhousie Medical Research Foundation","keywords":"Microbiome; Amplicon; Computer science; Computational biology; Workflow; Metagenomics; Preprocessor; Data mining; Scalability; Biology; Bioinformatics; Database; Artificial intelligence; Genetics; Polymerase chain reaction","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":[0.006654537,0.004260963,0.001967849,0.003384996,0.001749802,0.003758189,0.005042932,0.002079888,0.01980954],"category_scores_gemma":[0.01409352,0.002663906,0.005055729,0.002265686,0.001430252,0.004732781,0.006570683,0.004756477,0.01241802],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001296417,"about_ca_system_score_gemma":0.004149179,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01127831,"about_ca_topic_score_gemma":0.01126043,"domain_scores_codex":[0.9976555,0.0003883422,0.0002343756,0.0007968238,0.0006321633,0.0002928091],"domain_scores_gemma":[0.9956723,0.001694241,0.0003727379,0.00113153,0.000712621,0.0004165193],"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":[0.004150354,0.0004760839,0.01211698,0.004085163,0.001878716,0.001526684,0.002907689,0.02094353,0.09237624,0.01824049,0.5009149,0.3403832],"study_design_scores_gemma":[0.001342062,0.0005250165,0.0123354,0.0007872664,0.0004251406,0.001052968,0.0006437702,0.4064621,0.1214078,0.08056219,0.3730107,0.001445528],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"software","genre_gemma":"software","genre_scores_codex":[0.003927976,0.0004735148,0.3867034,0.0004569814,0.0003012695,0.0004128095,0.01634176,0.5898653,0.001516926],"genre_scores_gemma":[0.04315464,0.0008931063,0.7926212,0.00108693,0.0001436308,0.003058453,0.07055936,0.08457552,0.003907187],"genre_candidate":"software","genre_consensus":"software","teacher_disagreement_score":0.01980954,"threshold_uncertainty_score":0.06626952,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0396159238902001,"score_gpt":0.3216922498134997,"score_spread":0.2820763259232996,"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."}}