{"id":"W7162829424","doi":"","title":"Minimal-Data Peptide Design and Accessible Protein Language Models for Protein Engineering","year":2025,"lang":"","type":"dissertation","venue":"TSpace","topic":"vaccines and immunoinformatics approaches","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Protein engineering; Protein sequencing; Synthetic biology; Pipeline (software); Stability (learning theory); Sequence (biology); Divergence (linguistics)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000832449,0.000774655,0.0005736144,0.0002864357,0.0002446274,0.000820907,0.001208776,0.0006646129,0.002293413],"category_scores_gemma":[0.002586156,0.0004095458,0.0006478834,0.00033194,0.0006573485,0.001530126,0.001272954,0.002139921,0.001303414],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005566934,"about_ca_system_score_gemma":0.00104803,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001158257,"about_ca_topic_score_gemma":0.002220578,"domain_scores_codex":[0.9995618,0.0001317112,0.00002679867,0.0000871248,0.0001418446,0.00005068765],"domain_scores_gemma":[0.9994524,0.0002619905,0.00004874752,0.00009768671,0.00009811058,0.00004108548],"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":[0.0003084879,0.0002166771,0.001009124,0.0002753557,0.00006514785,0.0001724205,0.0001616046,0.7405193,0.04399881,0.04869552,0.003930682,0.1606469],"study_design_scores_gemma":[0.00001337211,0.00006180334,0.0000436738,0.000006437427,0.000004700838,0.0000188069,0.000011972,0.9826758,0.005767472,0.009990381,0.001397047,0.000008594176],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01748464,0.0003270416,0.9791929,0.0002594811,0.00004258206,0.00004889158,0.0001268814,0.001282165,0.001235433],"genre_scores_gemma":[0.3829011,0.0005897058,0.6104403,0.0005002876,0.00006862093,0.000389096,0.0007696244,0.0004940903,0.003847245],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002293413,"threshold_uncertainty_score":0.007672191,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0368453139958262,"score_gpt":0.3076602548904919,"score_spread":0.2708149408946657,"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."}}