{"id":"W4415506496","doi":"10.1038/s41587-025-02859-7","title":"Democratizing protein language model training, sharing and collaboration","year":2025,"lang":"en","type":"article","venue":"Nature Biotechnology","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Language model; Training (meteorology); Language barrier; Training set; Deep learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["open_science"],"consensus_categories":[],"category_scores_codex":[0.02385193,0.001768258,0.002993037,0.001635623,0.002948944,0.005138083,0.007889543,0.003711318,0.006783774],"category_scores_gemma":[0.07367902,0.001831912,0.002481759,0.002180678,0.00325102,0.01502291,0.02192734,0.006925482,0.00507288],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001572095,"about_ca_system_score_gemma":0.007588598,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003566147,"about_ca_topic_score_gemma":0.006767613,"domain_scores_codex":[0.9757554,0.01421714,0.0008965367,0.00368668,0.003976503,0.001467788],"domain_scores_gemma":[0.9090077,0.03923197,0.001738961,0.04272025,0.004945824,0.002355261],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.002189794,0.002435988,0.008290753,0.0004261482,0.0006804387,0.0008130545,0.002220787,0.359298,0.03293113,0.08623897,0.03489282,0.469582],"study_design_scores_gemma":[0.0001176356,0.0001437317,0.0002763384,0.00002328151,0.00006092434,0.0001344132,0.0004030563,0.8922523,0.008221012,0.09244716,0.005888286,0.00003189075],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06372634,0.0003800809,0.9152598,0.004003838,0.0002587427,0.0002477435,0.0002839809,0.009917462,0.005921931],"genre_scores_gemma":[0.5950817,0.0003066025,0.3925907,0.001122224,0.0003198229,0.0004152796,0.001851122,0.001942452,0.006370075],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9921104,"threshold_uncertainty_score":0.1261426,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.003583905128090461,"score_gpt":0.2719175898338231,"score_spread":0.2683336847057326,"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."}}