{"id":"W2111527303","doi":"10.1186/1742-4682-4-47","title":"Measuring the functional sequence complexity of proteins","year":2007,"lang":"en","type":"article","venue":"Theoretical Biology and Medical Modelling","topic":"Fractal and DNA sequence analysis","field":"Biochemistry, Genetics and Molecular Biology","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Sequence (biology); Computational biology; Measure (data warehouse); Protein sequencing; Computer science; Variable (mathematics); Biology; Bioinformatics; Algorithm; Data mining; Mathematics; Peptide sequence; Genetics; Gene","routes":{"ca_aff":true,"ca_fund":true,"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":["sts"],"consensus_categories":[],"category_scores_codex":[0.00117967,0.0000872933,0.0001324599,0.00002066286,0.0001012417,0.000003230779,0.0001406269,0.0001965964,0.000110762],"category_scores_gemma":[0.0002650832,0.00005188142,0.00006174925,0.00006299756,0.00315801,0.000002100455,0.00008843486,0.0001738026,0.000002686171],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000003827904,"about_ca_system_score_gemma":0.0000447688,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002563739,"about_ca_topic_score_gemma":0.000009463029,"domain_scores_codex":[0.9991463,0.00007447764,0.0002143839,0.0002059758,0.0001487456,0.0002101003],"domain_scores_gemma":[0.9995345,0.00009146267,0.00004409825,0.0001441885,0.00006110818,0.0001246005],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000217723,0.00004235192,0.001242462,0.0000125366,0.0000532961,0.000003085717,0.00003127629,0.0001653149,0.1486237,0.8444924,0.000005534826,0.005110342],"study_design_scores_gemma":[0.000698372,0.0005674572,0.0008979019,0.00007577736,0.00008964841,0.0001064121,0.0001851291,0.04619326,0.281235,0.668384,0.001210151,0.000356938],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5719723,0.0003667623,0.4259628,0.0006239867,0.00003136727,0.0000529662,0.000003080571,0.000004740518,0.0009820241],"genre_scores_gemma":[0.998123,0.0001823545,0.001066344,0.0004241357,0.0001541814,0.000003828769,0.00002650168,0.000004307088,0.00001528257],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4261508,"threshold_uncertainty_score":0.9995548,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05477837189285917,"score_gpt":0.2846450181320267,"score_spread":0.2298666462391675,"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."}}