{"id":"W4396922121","doi":"10.1101/2024.05.13.593988","title":"plantMASST - Community-driven chemotaxonomic digitization of plants","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genetics, Bioinformatics, and Biomedical Research","field":"Biochemistry, Genetics and Molecular Biology","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"Wilfrid Laurier University","funders":"","keywords":"Digitization; Botany; Biology; Computer science; Telecommunications","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000600122,0.0004050908,0.0004418317,0.0002266566,0.0001146501,0.0001356502,0.0008329137,0.0007542184,0.00001690683],"category_scores_gemma":[0.0002784954,0.0004037897,0.0001771454,0.000138684,0.0003141273,0.00000578556,0.001798087,0.0009441581,0.00007668595],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007746932,"about_ca_system_score_gemma":0.0006899884,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006075561,"about_ca_topic_score_gemma":0.000008580039,"domain_scores_codex":[0.9978749,0.0001442681,0.0006431649,0.0004706957,0.000399978,0.0004670454],"domain_scores_gemma":[0.997912,0.00004020512,0.0002982087,0.00120727,0.0002789899,0.0002633092],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00006546267,0.0001719541,0.001389455,0.001720188,0.0002652818,0.000008925444,0.00002031906,0.00009089662,0.9927036,0.00004951682,0.003488678,0.00002570766],"study_design_scores_gemma":[0.0004855797,0.0002448449,0.007828339,0.0006027149,0.00007370225,6.154141e-8,0.00001962364,0.001200094,0.969913,0.000009767215,0.01899485,0.000627408],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9953728,0.0007473699,0.0008748365,0.0001125254,0.000781592,0.000508226,0.001380471,0.00007176903,0.0001503541],"genre_scores_gemma":[0.9962705,0.001393488,0.001616651,0.00009793448,0.0004395386,0.0000490487,0.00002609269,0.00007867059,0.00002812565],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0227906,"threshold_uncertainty_score":0.9998414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01979758659384378,"score_gpt":0.2444697552074004,"score_spread":0.2246721686135566,"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."}}