{"id":"W4226049133","doi":"10.1089/big.2020.0383","title":"A Survey of Biological Data in a Big Data Perspective","year":2022,"lang":"en","type":"review","venue":"Big Data","topic":"Genetics, Bioinformatics, and Biomedical Research","field":"Biochemistry, Genetics and Molecular Biology","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Genome British Columbia","funders":"","keywords":"Big data; NoSQL; Computer science; Data science; Biological data; Context (archaeology); Relation (database); Data mining; Bioinformatics","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":["metaresearch","metaepi_narrow","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.004762772,0.0004103601,0.001201455,0.0002331558,0.0000644,0.00004781355,0.01358706,0.000555521,0.00008780612],"category_scores_gemma":[0.01122258,0.0003093949,0.00008670637,0.0006478622,0.000472234,0.00001085192,0.03882292,0.0005930641,0.00003158573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005198405,"about_ca_system_score_gemma":0.001979196,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001999017,"about_ca_topic_score_gemma":0.003431977,"domain_scores_codex":[0.9952658,0.0008433639,0.001025854,0.001689633,0.0006228529,0.0005524412],"domain_scores_gemma":[0.9873558,0.000314578,0.0003941826,0.01162778,0.000115214,0.0001924941],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00003541267,0.0002235686,0.00009515993,0.00157974,0.0002012768,0.00001356797,0.00001126233,2.488691e-8,0.0000115879,0.000002871549,0.02473667,0.9730889],"study_design_scores_gemma":[0.0002901357,0.000235811,0.0002844023,0.0003853829,0.0000884191,0.00001795925,0.0001073899,0.00008630041,0.000002801367,0.00000679493,0.9981382,0.0003563362],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00002108548,0.8875751,0.0001822716,0.000035825,0.0005203583,0.0006282428,0.1106477,0.000006672142,0.0003827921],"genre_scores_gemma":[0.0000911907,0.6926478,0.0002357824,0.00003170287,0.0003978157,0.00001439582,0.3064708,0.00002488701,0.00008560127],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9734016,"threshold_uncertainty_score":0.9999358,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.539427162647309,"score_gpt":0.4484749691020821,"score_spread":0.09095219354522699,"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."}}