{"id":"W2795286036","doi":"10.1186/s12915-018-0486-7","title":"A software tool ‘CroCo’ detects pervasive cross-species contamination in next generation sequencing data","year":2018,"lang":"en","type":"article","venue":"BMC Biology","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":113,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Biotechnology and Biological Sciences Research Council; Directorate for Biological Sciences; Institut Universitaire de France; Agence Nationale de la Recherche","keywords":"Contamination; Biology; Transcriptome; Computational biology; Software; Multiplex; Bioinformatics; Computer science; Genetics; Gene; Ecology","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":[],"consensus_categories":[],"category_scores_codex":[0.0002324158,0.0001413284,0.0001480173,0.00004608095,0.0001265214,0.00003599204,0.0002902554,0.0001919881,0.00001950054],"category_scores_gemma":[0.0003980485,0.0001338267,0.00003237637,0.00006818905,0.000215933,0.000003523116,0.0003573527,0.00005457756,0.00001494044],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000035102,"about_ca_system_score_gemma":0.0001316095,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008605372,"about_ca_topic_score_gemma":0.004090193,"domain_scores_codex":[0.998898,0.00008264074,0.0002278583,0.0005033437,0.00004254647,0.0002456285],"domain_scores_gemma":[0.9992217,0.00002990177,0.00008397422,0.0004846169,0.0001538243,0.00002593505],"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":[0.00003152843,0.00001059814,0.1216419,0.00000798747,0.00001817748,7.353252e-7,0.0001160857,0.00001737236,0.8748645,0.0001036452,0.0001910181,0.002996407],"study_design_scores_gemma":[0.001656482,0.001073325,0.4423611,0.00001716159,0.00002602243,0.0000357194,0.000397735,0.002304552,0.5051967,0.0004258161,0.04587007,0.000635349],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.990288,0.001135099,0.007783918,0.0000268794,0.0003093551,0.0002095872,0.00008736375,0.000005658046,0.0001541001],"genre_scores_gemma":[0.992563,0.0002113153,0.005374295,0.0002273945,0.0008590763,0.0000298959,0.0005226958,0.00001357897,0.0001987173],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3696678,"threshold_uncertainty_score":0.5457293,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1263663421162203,"score_gpt":0.3130937625710435,"score_spread":0.1867274204548233,"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."}}