{"id":"W3206711231","doi":"10.48550/arxiv.2110.07728","title":"Pre-training Molecular Graph Representation with 3D Geometry","year":2021,"lang":"en","type":"preprint","venue":"NPARC","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":154,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Molecular graph; Discriminative model; Computer science; Graph; Representation (politics); Consistency (knowledge bases); Theoretical computer science; Encoder; Training set; Artificial intelligence; Machine learning; Topology (electrical circuits); Mathematics; Combinatorics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005709988,0.0002990719,0.0003830223,0.0003485793,0.00008478893,0.0006423928,0.001158711,0.0001585907,0.00005450656],"category_scores_gemma":[0.0001870113,0.0003010454,0.0001710318,0.0009574539,0.00006785442,0.0003946185,0.001821446,0.0006066908,0.0000057394],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008151885,"about_ca_system_score_gemma":0.0006335197,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003190861,"about_ca_topic_score_gemma":0.000005105949,"domain_scores_codex":[0.9968105,0.0004399503,0.0003476821,0.001156206,0.0009173875,0.0003282678],"domain_scores_gemma":[0.9977033,0.0003333185,0.0002619566,0.001304817,0.0002744568,0.0001222179],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002312467,0.0001157957,0.0006700479,0.0001769812,0.0002911317,0.0004788105,0.004527446,0.8012118,0.005765845,0.03172506,0.0001845008,0.1548295],"study_design_scores_gemma":[0.000670446,0.0001139435,0.02330056,0.0006248528,0.00009317034,0.0001772399,0.0003034643,0.6726038,0.01232665,0.2883411,0.000200899,0.001243892],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2457707,0.0001097228,0.7458344,0.0004581572,0.0004884625,0.0002388262,0.000004897115,0.0001669852,0.006927889],"genre_scores_gemma":[0.434814,0.0000100342,0.5647305,0.0001932543,0.00006250317,0.00005700709,0.00005735665,0.00001984738,0.00005555424],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.256616,"threshold_uncertainty_score":0.9999442,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03343152167548662,"score_gpt":0.3201977049787594,"score_spread":0.2867661833032727,"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."}}