{"id":"W4381741903","doi":"10.1101/2023.06.19.545575","title":"Seq-InSite: sequence supersedes structure for protein interaction site prediction","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Protein Structure and Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University; Western University","funders":"","keywords":"Computer science; Sequence (biology); Matching (statistics); Source code; Code (set theory); Quality (philosophy); Data mining; Artificial intelligence; Computational biology; Programming language; Biology; Mathematics; Genetics; Set (abstract data type)","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":[0.002290839,0.002238577,0.00159994,0.001815157,0.0009692578,0.001470149,0.002846317,0.001353219,0.009499392],"category_scores_gemma":[0.004252093,0.001014666,0.0019811,0.001776073,0.0006922656,0.001433899,0.001842251,0.00252072,0.008307391],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000780237,"about_ca_system_score_gemma":0.002367065,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005087174,"about_ca_topic_score_gemma":0.01055422,"domain_scores_codex":[0.9988695,0.0003388932,0.00006201339,0.0002965245,0.0003529165,0.00008004507],"domain_scores_gemma":[0.9986281,0.0007237473,0.0001146261,0.0002300651,0.0001838247,0.0001195687],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003727139,0.00131542,0.02356633,0.002548481,0.001274161,0.000907829,0.0004797069,0.2264322,0.06320149,0.03106966,0.4860142,0.1594633],"study_design_scores_gemma":[0.000260486,0.0001876668,0.002077257,0.00006644245,0.00007577712,0.000232849,0.00007845368,0.9191184,0.02218828,0.01632993,0.0393049,0.00007952075],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1030295,0.001143384,0.5409846,0.001131441,0.0006009478,0.0004005531,0.08176421,0.2611557,0.009789544],"genre_scores_gemma":[0.2152284,0.0005906323,0.5263918,0.0005313661,0.0001430504,0.0008284341,0.2323181,0.01878625,0.005182002],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009499392,"threshold_uncertainty_score":0.03177863,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01627701024441613,"score_gpt":0.2413829618523802,"score_spread":0.2251059516079641,"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."}}