{"id":"W4406859606","doi":"10.1101/2025.01.24.634804","title":"Refining sequence-to-activity models by increasing model resolution","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"vaccines and immunoinformatics approaches","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research","funders":"","keywords":"Chromatin; Computational biology; Computer science; Sequence (biology); Gene regulatory network; Artificial intelligence; Sequence motif; Biology; Genetics; Gene; Gene expression","routes":{"ca_aff":true,"ca_fund":false,"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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0007362992,0.0005547842,0.0004710908,0.0001735565,0.000227538,0.0001934106,0.0006277813,0.0007377214,0.000002156489],"category_scores_gemma":[0.0002045653,0.0006237188,0.0001709604,0.0002650404,0.00005001985,0.00002986001,0.001228649,0.0005463071,0.000007079077],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001923424,"about_ca_system_score_gemma":0.0007483797,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002242019,"about_ca_topic_score_gemma":0.00000369656,"domain_scores_codex":[0.997658,0.0001185184,0.0005021273,0.0008933742,0.0002603821,0.0005676456],"domain_scores_gemma":[0.9977059,0.00001562799,0.0003330492,0.001401029,0.0003364438,0.0002079804],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001260352,0.00007330836,0.0001610827,0.00022892,0.0001179672,0.00000112259,0.000009308839,0.03488139,0.9628807,0.0002206828,0.001288879,0.00001064721],"study_design_scores_gemma":[0.0005984665,0.0001069244,0.0008552499,0.0005076618,0.0001178809,4.074383e-8,0.000004704478,0.3998581,0.593089,0.0000096126,0.003592401,0.001259907],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9169406,0.001004994,0.08016612,0.0001689838,0.000303423,0.0004342578,0.0007160943,0.0001202556,0.0001452684],"genre_scores_gemma":[0.9703469,0.0003102146,0.02853731,0.0003373817,0.0001511241,0.000178705,0.000006700167,0.00007111409,0.00006061361],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3697916,"threshold_uncertainty_score":0.9996214,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02914561722415017,"score_gpt":0.2431225174466088,"score_spread":0.2139769002224586,"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."}}