{"id":"W2054453573","doi":"10.1089/cmb.2007.0208","title":"New, Improved, and Practical k-Stem Sequence Similarity Measures for Probe Design","year":2008,"lang":"en","type":"article","venue":"Journal of Computational Biology","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of British Columbia","keywords":"Nearest neighbor search; Similarity (geometry); k-nearest neighbors algorithm; Sequence (biology); Computer science; Oligomer restriction; Data mining; Bit array; Algorithm; Pattern recognition (psychology); Oligonucleotide; Mathematics; Artificial intelligence; Biology; Genetics; Gene","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003277662,0.001005977,0.001297183,0.001954465,0.0005984212,0.001622629,0.001924284,0.001511806,0.002295093],"category_scores_gemma":[0.01923768,0.0004999223,0.0008206204,0.001959131,0.001314658,0.003850331,0.001601015,0.00190906,0.001374471],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001015302,"about_ca_system_score_gemma":0.001022314,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000367819,"about_ca_topic_score_gemma":0.0006571831,"domain_scores_codex":[0.994482,0.001481408,0.0005193331,0.0008291013,0.002525212,0.000162781],"domain_scores_gemma":[0.9907306,0.004608458,0.001320685,0.001112971,0.001941319,0.0002860847],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008221006,0.0005521444,0.006385787,0.0009920324,0.0001622956,0.0002002154,0.0004223802,0.1289238,0.2007561,0.0883155,0.003822469,0.5686453],"study_design_scores_gemma":[0.00009892802,0.001025496,0.001779881,0.00007148038,0.00006850863,0.0005293094,0.0001097142,0.8081781,0.1170755,0.06141549,0.00949042,0.0001570384],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01425765,0.0002420775,0.9835722,0.00006816956,0.00003316365,0.00009191851,0.0001107848,0.0008254076,0.0007985022],"genre_scores_gemma":[0.1183964,0.0001667823,0.8795698,0.000132161,0.00004598708,0.0004315557,0.0003620328,0.0002274936,0.0006678025],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003277662,"threshold_uncertainty_score":0.0173341,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04864507257986041,"score_gpt":0.3051138640791153,"score_spread":0.2564687914992549,"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."}}