============================================================ BARNACLE ANALYSIS LOG Started: 2025-11-24 11:02:41 Log file: ../output/13.00-multiomics-barnacle/barnacle_analysis_log_20251124_110241.txt ============================================================ Working in output directory: ../output/13.00-multiomics-barnacle Loaded normalized data: Apul: (10223, 41) Peve: (10223, 37) Ptua: (10223, 37) Common genes across all species: 10223 Filtered to common genes: Apul: (10223, 40) Peve: (10223, 36) Ptua: (10223, 36) Parsing sample information for each species... apul: Samples: 10 (['ACR.139', 'ACR.145', 'ACR.150']...) Timepoints: [1, 2, 3, 4] peve: Samples: 9 (['POR.216', 'POR.245', 'POR.260']...) Timepoints: [1, 2, 3, 4] ptua: Samples: 9 (['POC.219', 'POC.222', 'POC.255']...) Timepoints: [1, 2, 3, 4] Timepoints found across all species: [1, 2, 3, 4] Maximum samples in any species: 10 Detailed sample structure: apul: 10 samples × 4 timepoints peve: 9 samples × 4 timepoints ptua: 9 samples × 4 timepoints Creating 3D tensor by combining species and samples... Note: R-level filtering has already removed samples without all 4 timepoints Processing apul: Added ACR.139 with 4 timepoints: [1, 2, 3, 4] Added ACR.145 with 4 timepoints: [1, 2, 3, 4] Added ACR.150 with 4 timepoints: [1, 2, 3, 4] Added ACR.173 with 4 timepoints: [1, 2, 3, 4] Added ACR.186 with 4 timepoints: [1, 2, 3, 4] Added ACR.225 with 4 timepoints: [1, 2, 3, 4] Added ACR.229 with 4 timepoints: [1, 2, 3, 4] Added ACR.237 with 4 timepoints: [1, 2, 3, 4] Added ACR.244 with 4 timepoints: [1, 2, 3, 4] Added ACR.265 with 4 timepoints: [1, 2, 3, 4] Processing peve: Added POR.216 with 4 timepoints: [1, 2, 3, 4] Added POR.245 with 4 timepoints: [1, 2, 3, 4] Added POR.260 with 4 timepoints: [1, 2, 3, 4] Added POR.262 with 4 timepoints: [1, 2, 3, 4] Added POR.69 with 4 timepoints: [1, 2, 3, 4] Added POR.72 with 4 timepoints: [1, 2, 3, 4] Added POR.73 with 4 timepoints: [1, 2, 3, 4] Added POR.74 with 4 timepoints: [1, 2, 3, 4] Added POR.83 with 4 timepoints: [1, 2, 3, 4] Processing ptua: Added POC.219 with 4 timepoints: [1, 2, 3, 4] Added POC.222 with 4 timepoints: [1, 2, 3, 4] Added POC.255 with 4 timepoints: [1, 2, 3, 4] Added POC.259 with 4 timepoints: [1, 2, 3, 4] Added POC.40 with 4 timepoints: [1, 2, 3, 4] Added POC.42 with 4 timepoints: [1, 2, 3, 4] Added POC.52 with 4 timepoints: [1, 2, 3, 4] Added POC.53 with 4 timepoints: [1, 2, 3, 4] Added POC.57 with 4 timepoints: [1, 2, 3, 4] Creating 3D tensor with shape: (10223, 28, 4) Combined samples from all species: 28 === TENSOR STATISTICS === Tensor shape: (10223, 28, 4) Total elements: 1144976 Finite values: 1144976 Missing/NaN values: 0 Missing percentage: 0.00% Filled 112 sample-timepoint combinations Missing 0 sample-timepoint combinations Non-zero finite values: 1128956 Zero finite values: 16020 Sparsity among finite values: 1.40% Sample mapping: combined_index sample_label species sample_id 0 0 apul_ACR.139 apul ACR.139 1 1 apul_ACR.145 apul ACR.145 2 2 apul_ACR.150 apul ACR.150 3 3 apul_ACR.173 apul ACR.173 4 4 apul_ACR.186 apul ACR.186 5 5 apul_ACR.225 apul ACR.225 6 6 apul_ACR.229 apul ACR.229 7 7 apul_ACR.237 apul ACR.237 8 8 apul_ACR.244 apul ACR.244 9 9 apul_ACR.265 apul ACR.265 ================================================================================ CROSS-VALIDATION STRUCTURE FOR TIMESERIES TENSOR ================================================================================ Tensor structure: genes × samples × timepoints - Total samples: 28 - Timepoints per sample: 4 HOW THE TENSOR IS ORGANIZED: • Each sample dimension = one COLONY (e.g., ACR-139) • That colony's data at each timepoint is in the 3rd dimension • Example: tensor[:, 0, :] contains all 4 timepoints for colony #0 • Example: tensor[:, 0, 2] is colony #0 at timepoint TP3 CV approach: Leave-one-COLONY-out • Train on: N-1 colonies (all their timepoints) • Test on: 1 held-out colony (all its timepoints) IMPORTANT CLARIFICATION: • Biological reality: Each sample is collected at ONE timepoint • Tensor organization: Timepoints grouped by colony • CV tests: Can model predict a NEW colony's timeseries? CV Groups (one per colony): apul: 10 colonies apul_ACR.139, apul_ACR.145, apul_ACR.150, apul_ACR.173, apul_ACR.186... peve: 9 colonies peve_POR.216, peve_POR.245, peve_POR.260, peve_POR.262, peve_POR.69... ptua: 9 colonies ptua_POC.219, ptua_POC.222, ptua_POC.255, ptua_POC.259, ptua_POC.40... Total CV folds: 28 CV method: Leave-one-colony-out (train on N-1 colonies, test on 1) WARNING: This is computationally expensive! - Each rank tested will require 28 model fits - Consider subsetting to fewer colonies or using species-level CV ================================================================================ Replicate groups stored for cross-validation Available for dissertation-validated rank selection No Python documentation found for '# - Tests species generalization: can model learn patterns that work across species'. Use help() to get the interactive help utility. Use help(str) for help on the str class. ================================================================================ SPECIES-LEVEL CV GROUPS (ALTERNATIVE APPROACH) ================================================================================ CV approach: Leave-one-SPECIES-out - Train on 2 species - Test on 1 held-out species This tests: Can the decomposition generalize to a NEW SPECIES? Species groups: apul: 10 colonies Colonies: ACR.139, ACR.145, ACR.150, ACR.173, ACR.186... peve: 9 colonies Colonies: POR.216, POR.245, POR.260, POR.262, POR.69... ptua: 9 colonies Colonies: POC.219, POC.222, POC.255, POC.259, POC.40... Total CV folds: 3 This is MUCH faster than colony-level CV ================================================================================ ================================================================================ RECOMMENDATION FOR CV APPROACH: ================================================================================ 1. SPECIES-LEVEL CV (species_groups): → Use this for PRACTICAL rank selection → Fast: only 3 folds → Tests: Does model work on new species? → Similar to dissertation's leave-one-dataset-out approach 2. COLONY-LEVEL CV (replicate_groups): → Use for COMPREHENSIVE validation → Slow: ~30-40 folds → Tests: Does model work on new colony within same species? → More traditional biological replicate CV SUGGESTED: Start with species-level CV for initial rank selection ================================================================================ Full dissertation grid search functions loaded Ready to run complete rank × lambda grid search ================================================================================ STARTING FULL DISSERTATION GRID SEARCH Blaskowski (2024) Section 1.2.3 ================================================================================ Grid search configuration: Ranks to test (7): [5, 10, 15, 20, 25, 30, 35] Lambdas to test (8): [0.0, 0.01, 0.05, 0.1, 0.5, 1.0, 2.0, 5.0] Total combinations: 7 × 8 = 56 CV folds: 3 Max iterations: 10000 Random state: 42 ================================================================================ Estimated workload: Models to fit: 168 (7 ranks × 8 lambdas × 3 CV folds) Estimated time: ~5.6-14.0 minutes (assuming 2-5 min per model) ================================================================================ ================================================================================ STARTING GRID SEARCH... ================================================================================ ================================================================================ FULL DISSERTATION GRID SEARCH Blaskowski (2024) Section 1.2.3 ================================================================================ Testing 7 ranks: [5, 10, 15, 20, 25, 30, 35] Testing 8 lambdas: [0.0, 0.01, 0.05, 0.1, 0.5, 1.0, 2.0, 5.0] Total combinations: 56 Cross-validation: Leave-one-group-out (3 folds) Random state: 42 ================================================================================ For each (R, λ) combination, calculating: • 3 × 2 = 6 SSE scores (each model evaluated on 2 held-out groups) • 3 FMS scores (pairwise comparisons between 3 models) ================================================================================ ================================================================================ Combination 1/56: Rank=5, Lambda=0.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.23e+05 SSE on others: 5.06e+05 ± 4.88e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.65e+05 SSE on others: 5.22e+05 ± 5.70e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.94e+05 SSE on others: 5.01e+05 ± 4.17e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.10e+05 ± 4.19e+04 (6 scores) CV-FMS: 0.3631 ± 0.0539 (3 scores) ================================================================================ Combination 2/56: Rank=5, Lambda=0.01 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.18e+05 SSE on others: 5.16e+05 ± 4.70e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.63e+05 SSE on others: 5.11e+05 ± 5.58e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.99e+05 SSE on others: 4.93e+05 ± 4.89e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.07e+05 ± 4.40e+04 (6 scores) CV-FMS: 0.4108 ± 0.0418 (3 scores) ================================================================================ Combination 3/56: Rank=5, Lambda=0.05 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.18e+05 SSE on others: 5.16e+05 ± 4.90e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.63e+05 SSE on others: 5.12e+05 ± 5.56e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.99e+05 SSE on others: 4.93e+05 ± 4.91e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.07e+05 ± 4.41e+04 (6 scores) CV-FMS: 0.4224 ± 0.0484 (3 scores) ================================================================================ Combination 4/56: Rank=5, Lambda=0.1 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.18e+05 SSE on others: 5.16e+05 ± 5.01e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.63e+05 SSE on others: 5.13e+05 ± 5.54e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.99e+05 SSE on others: 4.93e+05 ± 4.93e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.08e+05 ± 4.41e+04 (6 scores) CV-FMS: 0.4504 ± 0.0674 (3 scores) ================================================================================ Combination 5/56: Rank=5, Lambda=0.5 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.18e+05 SSE on others: 5.16e+05 ± 5.23e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.66e+05 SSE on others: 5.19e+05 ± 5.51e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.99e+05 SSE on others: 4.94e+05 ± 5.06e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.10e+05 ± 4.48e+04 (6 scores) CV-FMS: 0.4499 ± 0.0707 (3 scores) ================================================================================ Combination 6/56: Rank=5, Lambda=1.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.18e+05 SSE on others: 5.17e+05 ± 5.32e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.67e+05 SSE on others: 5.18e+05 ± 5.44e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.99e+05 SSE on others: 4.94e+05 ± 5.09e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.10e+05 ± 4.44e+04 (6 scores) CV-FMS: 0.4185 ± 0.0666 (3 scores) ================================================================================ Combination 7/56: Rank=5, Lambda=2.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.18e+05 SSE on others: 5.17e+05 ± 5.45e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.65e+05 SSE on others: 5.19e+05 ± 5.81e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.94e+05 ± 5.04e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.10e+05 ± 4.59e+04 (6 scores) CV-FMS: 0.4161 ± 0.0762 (3 scores) ================================================================================ Combination 8/56: Rank=5, Lambda=5.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.25e+05 SSE on others: 5.08e+05 ± 3.55e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.63e+05 SSE on others: 5.20e+05 ± 5.35e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.93e+05 SSE on others: 5.01e+05 ± 4.27e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 5.10e+05 ± 4.04e+04 (6 scores) CV-FMS: 0.3870 ± 0.0821 (3 scores) ================================================================================ Combination 9/56: Rank=10, Lambda=0.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.25e+05 SSE on others: 4.98e+05 ± 4.08e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.71e+05 SSE on others: 5.00e+05 ± 3.62e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.87e+05 ± 3.55e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.95e+05 ± 2.99e+04 (6 scores) CV-FMS: 0.4850 ± 0.0350 (3 scores) ================================================================================ Combination 10/56: Rank=10, Lambda=0.01 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 5.00e+05 ± 4.63e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.69e+05 SSE on others: 5.01e+05 ± 4.53e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.95e+05 SSE on others: 4.88e+05 ± 3.46e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.96e+05 ± 3.35e+04 (6 scores) CV-FMS: 0.4862 ± 0.0118 (3 scores) ================================================================================ Combination 11/56: Rank=10, Lambda=0.05 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 5.00e+05 ± 4.03e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.68e+05 SSE on others: 5.01e+05 ± 4.52e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.88e+05 ± 3.58e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.96e+05 ± 3.39e+04 (6 scores) CV-FMS: 0.4807 ± 0.0059 (3 scores) ================================================================================ Combination 12/56: Rank=10, Lambda=0.1 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 5.00e+05 ± 3.61e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.69e+05 SSE on others: 5.01e+05 ± 4.48e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.88e+05 ± 3.58e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.96e+05 ± 3.37e+04 (6 scores) CV-FMS: 0.4742 ± 0.0026 (3 scores) ================================================================================ Combination 13/56: Rank=10, Lambda=0.5 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.26e+05 SSE on others: 5.01e+05 ± 3.63e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.67e+05 SSE on others: 5.02e+05 ± 3.72e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.94e+05 SSE on others: 4.88e+05 ± 3.54e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.97e+05 ± 3.04e+04 (6 scores) CV-FMS: 0.4556 ± 0.0232 (3 scores) ================================================================================ Combination 14/56: Rank=10, Lambda=1.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.25e+05 SSE on others: 4.99e+05 ± 7.75e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.67e+05 SSE on others: 5.03e+05 ± 3.70e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.95e+05 SSE on others: 4.88e+05 ± 3.57e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.97e+05 ± 3.06e+04 (6 scores) CV-FMS: 0.4791 ± 0.0469 (3 scores) ================================================================================ Combination 15/56: Rank=10, Lambda=2.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.99e+05 ± 8.69e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.66e+05 SSE on others: 5.03e+05 ± 3.75e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.93e+05 SSE on others: 4.90e+05 ± 3.45e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.97e+05 ± 3.04e+04 (6 scores) CV-FMS: 0.4588 ± 0.0017 (3 scores) ================================================================================ Combination 16/56: Rank=10, Lambda=5.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.20e+05 SSE on others: 5.01e+05 ± 4.55e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.66e+05 SSE on others: 5.03e+05 ± 3.80e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.94e+05 SSE on others: 4.90e+05 ± 3.41e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.98e+05 ± 3.02e+04 (6 scores) CV-FMS: 0.4475 ± 0.0191 (3 scores) ================================================================================ Combination 17/56: Rank=15, Lambda=0.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.94e+05 ± 5.50e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.95e+05 ± 3.97e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.87e+05 ± 3.35e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.92e+05 ± 3.04e+04 (6 scores) CV-FMS: 0.5651 ± 0.0325 (3 scores) ================================================================================ Combination 18/56: Rank=15, Lambda=0.01 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.20e+05 SSE on others: 4.95e+05 ± 5.17e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.72e+05 SSE on others: 4.96e+05 ± 4.07e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.87e+05 ± 3.29e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.93e+05 ± 3.06e+04 (6 scores) CV-FMS: 0.4889 ± 0.0399 (3 scores) ================================================================================ Combination 19/56: Rank=15, Lambda=0.05 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.20e+05 SSE on others: 4.95e+05 ± 5.19e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.72e+05 SSE on others: 4.96e+05 ± 4.08e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.87e+05 ± 3.31e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.93e+05 ± 3.08e+04 (6 scores) CV-FMS: 0.4991 ± 0.0231 (3 scores) ================================================================================ Combination 20/56: Rank=15, Lambda=0.1 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.20e+05 SSE on others: 4.95e+05 ± 5.24e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.71e+05 SSE on others: 4.96e+05 ± 4.01e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.87e+05 ± 3.39e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.93e+05 ± 3.08e+04 (6 scores) CV-FMS: 0.5013 ± 0.0371 (3 scores) ================================================================================ Combination 21/56: Rank=15, Lambda=0.5 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.95e+05 ± 2.30e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.71e+05 SSE on others: 4.97e+05 ± 3.69e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.87e+05 ± 3.37e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.93e+05 ± 2.92e+04 (6 scores) CV-FMS: 0.4479 ± 0.0466 (3 scores) ================================================================================ Combination 22/56: Rank=15, Lambda=1.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.96e+05 ± 1.92e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.71e+05 SSE on others: 4.95e+05 ± 3.86e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.87e+05 ± 3.47e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.93e+05 ± 3.02e+04 (6 scores) CV-FMS: 0.4654 ± 0.0736 (3 scores) ================================================================================ Combination 23/56: Rank=15, Lambda=2.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.21e+05 SSE on others: 4.94e+05 ± 2.87e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.68e+05 SSE on others: 4.99e+05 ± 4.24e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.87e+05 ± 3.46e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.93e+05 ± 3.20e+04 (6 scores) CV-FMS: 0.5105 ± 0.0240 (3 scores) ================================================================================ Combination 24/56: Rank=15, Lambda=5.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.20e+05 SSE on others: 4.96e+05 ± 4.98e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.69e+05 SSE on others: 4.97e+05 ± 4.04e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.88e+05 ± 3.24e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.93e+05 ± 3.03e+04 (6 scores) CV-FMS: 0.5285 ± 0.0124 (3 scores) ================================================================================ Combination 25/56: Rank=20, Lambda=0.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.92e+05 ± 1.50e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.72e+05 SSE on others: 4.95e+05 ± 3.95e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.94e+05 SSE on others: 4.87e+05 ± 3.07e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.92e+05 ± 2.91e+04 (6 scores) CV-FMS: 0.5554 ± 0.0412 (3 scores) ================================================================================ Combination 26/56: Rank=20, Lambda=0.01 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.19e+05 SSE on others: 4.93e+05 ± 4.84e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.95e+05 ± 3.82e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.86e+05 ± 3.42e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.91e+05 ± 3.00e+04 (6 scores) CV-FMS: 0.5714 ± 0.0235 (3 scores) ================================================================================ Combination 27/56: Rank=20, Lambda=0.05 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.19e+05 SSE on others: 4.93e+05 ± 4.95e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.95e+05 ± 3.89e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.86e+05 ± 3.45e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.91e+05 ± 3.04e+04 (6 scores) CV-FMS: 0.5711 ± 0.0428 (3 scores) ================================================================================ Combination 28/56: Rank=20, Lambda=0.1 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.19e+05 SSE on others: 4.93e+05 ± 5.01e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.72e+05 SSE on others: 4.96e+05 ± 3.84e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.86e+05 ± 3.47e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.92e+05 ± 3.03e+04 (6 scores) CV-FMS: 0.5690 ± 0.0409 (3 scores) ================================================================================ Combination 29/56: Rank=20, Lambda=0.5 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.19e+05 SSE on others: 4.93e+05 ± 5.67e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.72e+05 SSE on others: 4.95e+05 ± 3.91e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.85e+05 ± 3.49e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.91e+05 ± 3.08e+04 (6 scores) CV-FMS: 0.5530 ± 0.0057 (3 scores) ================================================================================ Combination 30/56: Rank=20, Lambda=1.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.19e+05 SSE on others: 4.93e+05 ± 5.97e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.94e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.86e+05 ± 3.64e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.91e+05 ± 3.14e+04 (6 scores) CV-FMS: 0.5185 ± 0.0073 (3 scores) ================================================================================ Combination 31/56: Rank=20, Lambda=2.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.20e+05 SSE on others: 4.93e+05 ± 2.69e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.72e+05 SSE on others: 4.94e+05 ± 3.93e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.95e+05 SSE on others: 4.87e+05 ± 3.48e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.91e+05 ± 3.06e+04 (6 scores) CV-FMS: 0.5336 ± 0.0355 (3 scores) ================================================================================ Combination 32/56: Rank=20, Lambda=5.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.19e+05 SSE on others: 4.94e+05 ± 5.14e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.71e+05 SSE on others: 4.96e+05 ± 3.73e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.95e+05 SSE on others: 4.87e+05 ± 3.42e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.92e+05 ± 2.97e+04 (6 scores) CV-FMS: 0.5730 ± 0.0301 (3 scores) ================================================================================ Combination 33/56: Rank=25, Lambda=0.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.92e+05 ± 3.59e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.92e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.84e+05 ± 3.29e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 2.99e+04 (6 scores) CV-FMS: 0.5889 ± 0.0118 (3 scores) ================================================================================ Combination 34/56: Rank=25, Lambda=0.01 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.92e+05 ± 4.27e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.76e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.84e+05 ± 3.51e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 3.01e+04 (6 scores) CV-FMS: 0.5521 ± 0.0397 (3 scores) ================================================================================ Combination 35/56: Rank=25, Lambda=0.05 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.92e+05 ± 4.02e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.82e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.85e+05 ± 3.57e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.91e+05 ± 3.05e+04 (6 scores) CV-FMS: 0.5274 ± 0.0203 (3 scores) ================================================================================ Combination 36/56: Rank=25, Lambda=0.1 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.92e+05 ± 4.02e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.81e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.85e+05 ± 3.54e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 3.04e+04 (6 scores) CV-FMS: 0.5136 ± 0.0185 (3 scores) ================================================================================ Combination 37/56: Rank=25, Lambda=0.5 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.92e+05 ± 4.15e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.94e+05 ± 3.88e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.85e+05 ± 3.52e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 3.06e+04 (6 scores) CV-FMS: 0.5324 ± 0.0227 (3 scores) ================================================================================ Combination 38/56: Rank=25, Lambda=1.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.92e+05 ± 2.65e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.92e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.96e+05 SSE on others: 4.85e+05 ± 3.50e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 3.07e+04 (6 scores) CV-FMS: 0.5521 ± 0.0214 (3 scores) ================================================================================ Combination 39/56: Rank=25, Lambda=2.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.92e+05 ± 3.61e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.92e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.95e+05 SSE on others: 4.85e+05 ± 3.58e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 3.09e+04 (6 scores) CV-FMS: 0.6152 ± 0.0251 (3 scores) ================================================================================ Combination 40/56: Rank=25, Lambda=5.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.23e+05 SSE on others: 4.92e+05 ± 4.35e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.71e+05 SSE on others: 4.94e+05 ± 3.87e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.95e+05 SSE on others: 4.84e+05 ± 3.37e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 3.00e+04 (6 scores) CV-FMS: 0.5974 ± 0.0182 (3 scores) ================================================================================ Combination 41/56: Rank=30, Lambda=0.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.91e+05 ± 2.84e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.94e+05 ± 3.80e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.83e+05 ± 3.18e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.89e+05 ± 2.90e+04 (6 scores) CV-FMS: 0.6081 ± 0.0251 (3 scores) ================================================================================ Combination 42/56: Rank=30, Lambda=0.01 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.90e+05 ± 5.22e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.87e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.84e+05 ± 3.36e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.89e+05 ± 3.00e+04 (6 scores) CV-FMS: 0.5654 ± 0.0223 (3 scores) ================================================================================ Combination 43/56: Rank=30, Lambda=0.05 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.90e+05 ± 5.32e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.87e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.84e+05 ± 3.32e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.89e+05 ± 2.99e+04 (6 scores) CV-FMS: 0.5136 ± 0.0477 (3 scores) ================================================================================ Combination 44/56: Rank=30, Lambda=0.1 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.90e+05 ± 5.20e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.85e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.83e+05 ± 3.32e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.89e+05 ± 2.97e+04 (6 scores) CV-FMS: 0.5344 ± 0.0455 (3 scores) ================================================================================ Combination 45/56: Rank=30, Lambda=0.5 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.24e+05 SSE on others: 4.91e+05 ± 4.77e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.89e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.84e+05 ± 3.39e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.89e+05 ± 3.02e+04 (6 scores) CV-FMS: 0.5815 ± 0.0217 (3 scores) ================================================================================ Combination 46/56: Rank=30, Lambda=1.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.23e+05 SSE on others: 4.91e+05 ± 4.17e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.83e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.83e+05 ± 3.24e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.89e+05 ± 2.94e+04 (6 scores) CV-FMS: 0.5984 ± 0.0179 (3 scores) ================================================================================ Combination 47/56: Rank=30, Lambda=2.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.23e+05 SSE on others: 4.91e+05 ± 4.36e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.93e+05 ± 3.91e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.83e+05 ± 3.20e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.89e+05 ± 2.96e+04 (6 scores) CV-FMS: 0.5948 ± 0.0128 (3 scores) ================================================================================ Combination 48/56: Rank=30, Lambda=5.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.22e+05 SSE on others: 4.91e+05 ± 5.17e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.72e+05 SSE on others: 4.94e+05 ± 3.84e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.95e+05 SSE on others: 4.84e+05 ± 3.53e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.90e+05 ± 3.05e+04 (6 scores) CV-FMS: 0.6018 ± 0.0170 (3 scores) ================================================================================ Combination 49/56: Rank=35, Lambda=0.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.25e+05 SSE on others: 4.90e+05 ± 5.84e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.84e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.82e+05 ± 3.23e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.95e+04 (6 scores) CV-FMS: 0.5606 ± 0.0456 (3 scores) ================================================================================ Combination 50/56: Rank=35, Lambda=0.01 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 4.89e+05 ± 6.55e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.75e+05 SSE on others: 4.93e+05 ± 3.88e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.83e+05 ± 3.21e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.96e+04 (6 scores) CV-FMS: 0.6160 ± 0.0160 (3 scores) ================================================================================ Combination 51/56: Rank=35, Lambda=0.05 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 4.89e+05 ± 6.50e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.85e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.83e+05 ± 3.17e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.93e+04 (6 scores) CV-FMS: 0.5783 ± 0.0135 (3 scores) ================================================================================ Combination 52/56: Rank=35, Lambda=0.1 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 4.89e+05 ± 6.35e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.85e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.83e+05 ± 3.15e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.92e+04 (6 scores) CV-FMS: 0.5874 ± 0.0053 (3 scores) ================================================================================ Combination 53/56: Rank=35, Lambda=0.5 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 4.89e+05 ± 6.52e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.86e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.83e+05 ± 3.15e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.93e+04 (6 scores) CV-FMS: 0.6110 ± 0.0114 (3 scores) ================================================================================ Combination 54/56: Rank=35, Lambda=1.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.27e+05 SSE on others: 4.88e+05 ± 7.31e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.74e+05 SSE on others: 4.93e+05 ± 3.87e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.82e+05 ± 3.08e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.92e+04 (6 scores) CV-FMS: 0.6220 ± 0.0088 (3 scores) ================================================================================ Combination 55/56: Rank=35, Lambda=2.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.26e+05 SSE on others: 4.89e+05 ± 7.37e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.93e+05 ± 3.89e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.98e+05 SSE on others: 4.82e+05 ± 3.13e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.95e+04 (6 scores) CV-FMS: 0.6236 ± 0.0154 (3 scores) ================================================================================ Combination 56/56: Rank=35, Lambda=5.0 ================================================================================ Fold 1/3: Training without apul ✓ SSE on apul: 6.26e+05 SSE on others: 4.89e+05 ± 5.82e+03 Fold 2/3: Training without peve ✓ SSE on peve: 5.73e+05 SSE on others: 4.93e+05 ± 3.84e+04 Fold 3/3: Training without ptua ✓ SSE on ptua: 5.97e+05 SSE on others: 4.83e+05 ± 3.31e+04 Calculating FMS between 3 successful models... RESULTS: CV-SSE: 4.88e+05 ± 2.97e+04 (6 scores) CV-FMS: 0.6291 ± 0.0079 (3 scores) ================================================================================ STAGE 1: RANK SELECTION Dissertation: 'Select R at minimum CV-SSE for λ=0.0 models' ================================================================================ ✓ OPTIMAL RANK: 35 CV-SSE: 4.88e+05 ± 2.95e+04 All λ=0.0 results: rank mean_cv_sse std_cv_sse n_successful_folds 35 4.88e+05 2.95e+04 3 30 4.89e+05 2.90e+04 3 25 4.90e+05 2.99e+04 3 20 4.92e+05 2.91e+04 3 15 4.92e+05 3.04e+04 3 10 4.95e+05 2.99e+04 3 5 5.10e+05 4.19e+04 3 ================================================================================ STAGE 2: LAMBDA SELECTION Dissertation: 'Select maximum λ where CV-FMS ≥ (max_FMS - 1SE)' ================================================================================ Maximum FMS: 0.6291 ± 0.0046 (at λ=5.0) 1SE Threshold: 0.6246 ✓ OPTIMAL LAMBDA: 5.0 CV-FMS: 0.6291 CV-SSE: 4.88e+05 (Maximum λ where FMS ≥ 0.6246) All results at rank=35: lambda mean_cv_fms se_cv_fms mean_cv_sse 0.0000 0.5606 0.0263 488328.1202 0.0100 0.6160 0.0092 488000.6664 0.0500 0.5783 0.0078 487899.4275 0.1000 0.5874 0.0031 487920.7548 0.5000 0.6110 0.0066 487980.5383 1.0000 0.6220 0.0051 487835.9215 2.0000 0.6236 0.0089 487951.7320 5.0000 0.6291 0.0046 488151.6329 ================================================================================ FINAL SELECTED PARAMETERS ================================================================================ ✓ Rank: 35 ✓ Lambda: 5.0 ================================================================================ ================================================================================ GRID SEARCH COMPLETE ================================================================================ Total time: 3052.5 minutes Time per combination: 3270.5 seconds ================================================================================ ✓ Full grid results saved to: full_grid_results.csv ✓ Optimal parameters saved to: optimal_parameters.json ================================================================================ Results saved to directory: ../output/13.00-multiomics-barnacle/dissertation_grid_search ================================================================================