{"id":"W4212859655","doi":"10.1101/2022.02.18.481053","title":"Is Over-parameterization a Problem for Profile Mixture Models?","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Tree (set theory); Amino acid; Sequence (biology); Phylogenetic tree; Set (abstract data type); Mathematics; Biological system; Network topology; Computer science; Combinatorics; Biology; Genetics","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":[0.02780983,0.001723102,0.002546329,0.001471958,0.001595909,0.002335837,0.003247127,0.003610434,0.002321416],"category_scores_gemma":[0.1412582,0.002023493,0.002018225,0.001380449,0.003343806,0.005353926,0.002936797,0.005681703,0.0007869333],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002059302,"about_ca_system_score_gemma":0.001567726,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01180192,"about_ca_topic_score_gemma":0.006988504,"domain_scores_codex":[0.9891855,0.007162465,0.0006360345,0.001402776,0.001121898,0.0004913277],"domain_scores_gemma":[0.841976,0.1349562,0.007475759,0.01002558,0.004611888,0.0009545715],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0006995792,0.0001859145,0.02250744,0.0004244628,0.0006506477,0.000813056,0.001004819,0.8417419,0.005027507,0.06000415,0.003981354,0.0629592],"study_design_scores_gemma":[0.00003446993,0.00005142023,0.001248165,0.00005589289,0.0000400541,0.0002306949,0.0001233592,0.9387305,0.001894244,0.05636075,0.001170108,0.00006038576],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05199784,0.0005076217,0.9442259,0.001443424,0.00006272078,0.00009783586,0.0001780817,0.0006304387,0.0008561449],"genre_scores_gemma":[0.7910164,0.0006328127,0.2028306,0.001238864,0.0001998219,0.0005223108,0.000994321,0.0006550151,0.001909878],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02780983,"threshold_uncertainty_score":0.1470742,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01621329503634905,"score_gpt":0.2284810281023328,"score_spread":0.2122677330659837,"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."}}