{"id":"W4394930608","doi":"10.1101/2024.04.15.589672","title":"ProSST: Protein Language Modeling with Quantized Structure and Disentangled Attention","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Shanghai Jiao Tong University; Science and Technology Commission of Shanghai Municipality; National Natural Science Foundation of China; National Science Foundation","keywords":"Computer science; Linguistics; Natural language processing; Philosophy","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":[],"consensus_categories":[],"category_scores_codex":[0.0007711871,0.0009899782,0.0008324356,0.0005073519,0.0003019964,0.0006768171,0.001928751,0.001098457,0.002661362],"category_scores_gemma":[0.001787636,0.0004704386,0.001134703,0.0005574641,0.0006041149,0.001527683,0.001177605,0.001957538,0.001098141],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009107891,"about_ca_system_score_gemma":0.001267582,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008615067,"about_ca_topic_score_gemma":0.009700251,"domain_scores_codex":[0.9996334,0.000111281,0.00001625739,0.0001155506,0.00008616596,0.0000373957],"domain_scores_gemma":[0.9993749,0.0003769237,0.00004357966,0.0000706932,0.00008682585,0.00004698067],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002695826,0.0001383638,0.0007090445,0.0001072471,0.0001020098,0.0001497944,0.00005631628,0.904725,0.007903758,0.008746044,0.006678725,0.07041413],"study_design_scores_gemma":[0.000007766822,0.00001136126,0.000019849,0.000001212916,0.000002093037,0.000005449845,0.000001725039,0.9975019,0.000483612,0.001835223,0.0001277761,0.000002009706],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1066584,0.001186308,0.8722107,0.001110034,0.00016789,0.000111601,0.001567278,0.014281,0.002706678],"genre_scores_gemma":[0.7798246,0.0005445129,0.2062925,0.0008712314,0.0001102466,0.0002495776,0.004772658,0.0008362517,0.006498415],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008615067,"threshold_uncertainty_score":0.01712984,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005088720207963003,"score_gpt":0.2178787602609839,"score_spread":0.2127900400530209,"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."}}