{"id":"W4407578746","doi":"10.1093/bioinformatics/btaf071","title":"<i>PNL</i>: a software to build polygenic risk scores using a super learner approach based on PairNet, a Convolutional Neural Network","year":2025,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"Academia Sinica; Institute of Biomedical Sciences, Academia Sinica; Natural Sciences and Engineering Research Council of Canada; National Science and Technology Council","keywords":"Convolutional neural network; Computer science; Polygenic risk score; Software; Artificial intelligence; Machine learning; Artificial neural network; Programming language; Biology; Single-nucleotide polymorphism; Genetics","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":[0.001855128,0.001428082,0.0005267958,0.001425167,0.0003991971,0.001062639,0.002620063,0.0009947261,0.033776],"category_scores_gemma":[0.006424342,0.0009145139,0.001242884,0.0006919282,0.0005483248,0.001828147,0.00183952,0.002220258,0.01140504],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007463618,"about_ca_system_score_gemma":0.001216599,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005245471,"about_ca_topic_score_gemma":0.006852927,"domain_scores_codex":[0.9995762,0.00008260579,0.00004035045,0.0001396739,0.0001322454,0.00002896024],"domain_scores_gemma":[0.99857,0.000810632,0.0001667018,0.0001902668,0.0001895838,0.00007280042],"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":[0.0007368085,0.0002821599,0.01573052,0.00116677,0.0008765794,0.001256047,0.0005236272,0.1085716,0.01472808,0.02544575,0.4029171,0.427765],"study_design_scores_gemma":[0.0001689782,0.00009081441,0.003016169,0.00009485184,0.00007447477,0.0003754059,0.00003035688,0.900413,0.01532845,0.02053267,0.05975622,0.0001184905],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.005910428,0.0001278584,0.6215118,0.0004145188,0.0001316096,0.0001905936,0.007121705,0.3601464,0.00444521],"genre_scores_gemma":[0.1100191,0.0004178653,0.8015007,0.0009740391,0.000124617,0.001174028,0.0304024,0.04313995,0.01224728],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.033776,"threshold_uncertainty_score":0.112992,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0198151624764901,"score_gpt":0.2710087187799389,"score_spread":0.2511935563034488,"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."}}