{"id":"W4413491758","doi":"10.1093/bioinformatics/btaf458","title":"An interpretable alphabet for local protein structure search based on amino acid neighborhoods","year":2025,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Protein Structure and Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Aeronautics and Space Institute; Wellcome Trust","keywords":"Alphabet; Computer science; Amino acid; Computational biology; Artificial intelligence; Biology; Genetics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001397181,0.0002165663,0.0001780729,0.00009402981,0.0001175916,0.00006963576,0.0003580037,0.0002838014,0.00001484001],"category_scores_gemma":[0.00005640815,0.0001843757,0.00009017152,0.0001289978,0.00008966478,0.00001389517,0.00007405635,0.0001411129,0.000003417065],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003556106,"about_ca_system_score_gemma":0.0002314805,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005497295,"about_ca_topic_score_gemma":0.00001468128,"domain_scores_codex":[0.9989861,0.00002867195,0.0002885729,0.0002145088,0.000150712,0.0003313893],"domain_scores_gemma":[0.9991241,0.00001128832,0.00006906837,0.0005897295,0.0001153553,0.00009052468],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002054317,0.0002699172,0.00116068,0.00156603,0.00026322,0.000003006688,0.0003301776,0.01874868,0.7405987,0.01144754,0.008146205,0.2154115],"study_design_scores_gemma":[0.001014253,0.000879531,0.0001469778,0.00006349325,0.00002201966,0.000001987214,0.0001191099,0.4995902,0.4880962,0.000690536,0.009108767,0.0002669722],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08928879,0.00006202787,0.9059467,0.000126217,0.0001776845,0.001014032,0.0003436074,0.00003652741,0.003004372],"genre_scores_gemma":[0.9537225,0.000002539285,0.04375068,0.001378424,0.00008211966,0.00005518643,0.0007261527,0.00001814385,0.000264286],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8644337,"threshold_uncertainty_score":0.7518623,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.004914227286936982,"score_gpt":0.2577339499135217,"score_spread":0.2528197226265847,"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."}}