{"id":"W4365456885","doi":"10.1021/acs.jctc.3c00031","title":"Geometric Deep Learning for Molecular Crystal Structure Prediction","year":2023,"lang":"en","type":"article","venue":"Journal of Chemical Theory and Computation","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":46,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Division of Chemistry; Natural Sciences and Engineering Research Council of Canada; Division of Materials Research; Deutsche Forschungsgemeinschaft","keywords":"Crystal structure prediction; Computer science; Ranking (information retrieval); Crystal (programming language); Artificial intelligence; Machine learning; Stability (learning theory); Pipeline (software); Test set; Chemical space; Graph; Filter (signal processing); Data mining; Algorithm; Crystal structure; Theoretical computer science; Chemistry; Crystallography; Computer vision","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001167668,0.00008662793,0.0001509717,0.0003102131,0.00006402247,0.00009509458,0.0001843259,0.00005611717,0.000002055936],"category_scores_gemma":[0.0006451409,0.00007819344,0.00008873372,0.0007166233,0.00002920114,0.000377,0.00007687925,0.0001838182,7.500765e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003144719,"about_ca_system_score_gemma":0.00003408007,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":8.411993e-8,"about_ca_topic_score_gemma":6.83078e-9,"domain_scores_codex":[0.998929,0.0002213112,0.0003003437,0.0001476438,0.0002738741,0.0001278384],"domain_scores_gemma":[0.9980299,0.001401666,0.0002305607,0.00005264809,0.0002135848,0.00007160904],"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":[0.0001620696,0.00003044491,0.00007817085,0.00005499702,0.00007231958,0.00001301367,0.0005108449,0.6078271,0.08942913,0.06211421,0.00008115535,0.2396265],"study_design_scores_gemma":[0.0004458887,0.0001267933,0.000865015,0.00001790429,0.00001623436,0.00009769705,0.00004032718,0.5022427,0.0151639,0.4808126,0.00009679441,0.00007404106],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4402396,0.00009409343,0.5593591,0.0000782474,0.0001400315,0.00004468154,0.000001180503,0.00002764947,0.00001542741],"genre_scores_gemma":[0.9609213,0.00000738782,0.0388662,0.00005357749,0.0001255149,0.000001720324,0.00001090845,0.000007229262,0.000006155844],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5206817,"threshold_uncertainty_score":0.3188636,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01173941113275205,"score_gpt":0.2890998000662413,"score_spread":0.2773603889334892,"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."}}