{"id":"W4416228039","doi":"10.1101/2025.11.07.687302","title":"Learning the Unseen: Data-Augmented Deep Learning for PTM Discovery with Prosit-PTM","year":2025,"lang":"","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"vaccines and immunoinformatics approaches","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Thermo Fisher Scientific (Canada)","funders":"Horizon 2020 Framework Programme; Elitenetzwerk Bayern; European Commission","keywords":"Deep learning; Identification (biology); Property (philosophy); Human proteome project; Feature (linguistics); Encoding (memory); Interpretation (philosophy); Proteome","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","sts","scholarly_communication"],"consensus_categories":["metaepi_narrow"],"category_scores_codex":[0.001834946,0.001360431,0.001075446,0.0002023257,0.001445242,0.001268489,0.002678845,0.0009655362,0.0000147541],"category_scores_gemma":[0.0009048164,0.001066376,0.0003770945,0.0006490991,0.0003008275,0.0001248138,0.003266946,0.001854446,0.00001434609],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001617515,"about_ca_system_score_gemma":0.001361341,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004999578,"about_ca_topic_score_gemma":0.00001209077,"domain_scores_codex":[0.9943578,0.0003818239,0.001328059,0.002094719,0.0005224196,0.00131515],"domain_scores_gemma":[0.9934782,0.0001996038,0.001405221,0.00378861,0.0008911673,0.0002372238],"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":[0.002804863,0.0009271974,0.01748302,0.006129188,0.007145965,0.00001918196,0.0002516431,0.03639495,0.9266322,0.0009223306,0.00099042,0.0002990524],"study_design_scores_gemma":[0.007166695,0.002970409,0.01743917,0.00256657,0.003260555,4.406454e-7,0.00077975,0.134338,0.4861696,0.000002167219,0.3402497,0.005056937],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7946416,0.01457302,0.1785112,0.001105891,0.001817983,0.007832369,0.001111841,0.0002937152,0.0001123164],"genre_scores_gemma":[0.9879329,0.003511464,0.005842008,0.0002946577,0.0007484056,0.0008388379,0.00009160423,0.00021854,0.0005216282],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4404626,"threshold_uncertainty_score":0.9999146,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01376455659092892,"score_gpt":0.2232471392514917,"score_spread":0.2094825826605628,"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."}}