{"id":"W3118964695","doi":"10.1101/2020.12.31.425014","title":"A Practical Guide to Sparse K-Means Clustering for Studying Molecular Development of the Human Brain","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Cluster analysis; Lasso (programming language); Computational biology; Curse of dimensionality; Computer science; Biology; Elastic net regularization; Contrast (vision); Sample (material); Pattern recognition (psychology); Feature selection; Artificial intelligence","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003307856,0.001940639,0.001156858,0.002959689,0.0007859883,0.002331687,0.003531636,0.003265211,0.03591787],"category_scores_gemma":[0.01657381,0.001417731,0.001311201,0.003054892,0.00121406,0.001804039,0.001978223,0.0052148,0.03166438],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000916463,"about_ca_system_score_gemma":0.002249711,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006232381,"about_ca_topic_score_gemma":0.008617332,"domain_scores_codex":[0.9983033,0.0007191738,0.0002120089,0.0002742569,0.0004490837,0.000042225],"domain_scores_gemma":[0.9935424,0.00384611,0.0002152161,0.0004267741,0.001760849,0.0002086471],"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":[0.00008071666,0.0001438713,0.0005423808,0.001516514,0.0001205864,0.0005947597,0.0004397391,0.02355617,0.005114154,0.1182871,0.6189293,0.2306747],"study_design_scores_gemma":[0.0001010073,0.00008870044,0.0006706564,0.0005936601,0.00002577237,0.0005789244,0.0001624908,0.1143917,0.001725748,0.1385342,0.7429785,0.0001486317],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0002221231,0.001996313,0.9826873,0.001778558,0.0007571963,0.0003376853,0.001530011,0.004397676,0.006293073],"genre_scores_gemma":[0.001477155,0.001818867,0.9865891,0.0006833852,0.0003134614,0.0007128563,0.0007747223,0.001246658,0.006383763],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.03591787,"threshold_uncertainty_score":0.1201573,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03434401772304783,"score_gpt":0.3033454376972105,"score_spread":0.2690014199741626,"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."}}