Results of RSV Analysis (237?) ======================= In this analysis we are using the data from all 20 subjects (no subjects disqualified). The classification that we are trying to learn and predict is the following: Subjects 1,2,3,6,7,8,11,12,14,20 are considered symptomatic and all others are considered asymptomatic. For the temporal abstraction, we define baseline, early, middle, and late intervals are -24...0, 5...29, 36...117.5, and 125...165.5, respectively. Results for Composite Single-Gene Predictors -------------------------------------------- Early Predictor: The best ternary logic predictor synthesized, which consists in average of 2.55 single gene-predictors, makes prediction in 9 out of 20 cases with 2 errors (which directly translates into an error rate of 22%). The normalized weighted error rate (based on an equivalent binary predictor) is 36%. The p-value determined using 100 randomization tests is ~ 0.3. The best predictor based on all 20 subjects (i.e. without crossvalidation) contains exactly the following three genes: NDUFA9 (negative), FUT9 and PLXDC1 (both positive predictors). Middle Predictor: The best predictor consists in average of 1.8 genes and makes predictions in 11 out of 20 cases with 5 errors (yielding a direct error rate of 45%). The normalized error rate is 46% and the p-value is ~ 0.36. The best predictor based on all 20 subjects consists of two genes: ITGB8 (positive) and TWSG1 (negative). Late Predictor: The best predictor consists in average of 1.95 genes and makes predictions in 13 out of 20 cases with 4 errors (translating into a direct error rate of 31%). The normalized error rate is 36% with a p-value of ~ 0.20. The best predictor based on all 20 subjects consists of two genes: STX17 and PML (both positive). Relevant Log-File Fragments Testing top composite predictors at time 17.0 ... Average Best Composite Predictor: Top-2.55 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 9 (0.45) Err: 2 (0.22) NErr: 0.38 WErr: 0.21 NWErr: 0.36 CScore: 0.36 Ok: 1,3,6,7,11,14,18 Err: 15,20 CompPred Top-1 - Sub: 20 Pred: 11 (0.55) Err: 4 (0.36) NErr: 0.42 WErr: 0.34 NWErr: 0.41 CScore: 0.41 Ok: 1,3,6,11,12,16,18 Err: 15,17,19,20 CompPred Top-2 - Sub: 20 Pred: 11 (0.55) Err: 4 (0.36) NErr: 0.42 WErr: 0.34 NWErr: 0.41 CScore: 0.41 Ok: 1,3,6,7,11,14,18 Err: 4,15,19,20 CompPred Top-3 - Sub: 16 Pred: 9 (0.56) Err: 2 (0.22) NErr: 0.34 WErr: 0.21 NWErr: 0.36 CScore: 0.36 Ok: 1,3,6,7,11,14,18 Err: 15,20 Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.00 p(e <= 0.20) = 0.01 p(e <= 0.25) = 0.10 p(e <= 0.30) = 0.13 p(e <= 0.35) = 0.27 p(e <= 0.40) = 0.41 p(e <= 0.45) = 0.48 p(e <= 0.50) = 0.61 Computing top composite predictors at time 17.0 (without crossvalidation) ... Computing predictors at time 17.0 ... Defining composite predictors ... 1. Pred NDUFA9 - Pat: Neg - Sub: 20 Pred: 13 (0.65) Err: 2 (0.15) NErr: 0.28 WErr: 0.14 NWErr: 0.26 Ok: 1,3,6,7,10,11,12,13,14,16,18 Err: 15,19 2. Pred FUT9 - Pat: Pos - Sub: 20 Pred: 9 (0.45) Err: 2 (0.22) NErr: 0.38 WErr: 0.22 NWErr: 0.37 Ok: 2,4,7,11,12,13,18 Err: 16,20 3. Pred PLXDC1 - Pat: Pos - Sub: 20 Pred: 8 (0.40) Err: 3 (0.38) NErr: 0.45 WErr: 0.40 NWErr: 0.46 Ok: 4,5,10,19,20 Err: 11,12,15 Testing top composite predictors at time 17.0 ... Best Composite Predictor: Top-3 Score: 0.19 Testing top composite predictors at time 76.75 ... Average Best Composite Predictor: Top-1.80 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 11 (0.55) Err: 5 (0.45) NErr: 0.48 WErr: 0.42 NWErr: 0.46 CScore: 0.46 Ok: 3,6,7,9,11,12 Err: 2,5,13,15,16 CompPred Top-1 - Sub: 20 Pred: 11 (0.55) Err: 5 (0.45) NErr: 0.48 WErr: 0.44 NWErr: 0.47 CScore: 0.47 Ok: 3,6,7,9,10,12 Err: 2,11,13,15,16 CompPred Top-2 - Sub: 20 Pred: 11 (0.55) Err: 5 (0.45) NErr: 0.48 WErr: 0.43 NWErr: 0.46 CScore: 0.46 Ok: 3,6,7,9,12,18 Err: 2,5,13,15,16 CompPred Top-3 - Sub: 12 Pred: 10 (0.83) Err: 2 (0.20) NErr: 0.25 WErr: 0.18 NWErr: 0.33 CScore: 0.33 Ok: 1,3,7,9,11,12,19,20 Err: 4,16 Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.02 p(e <= 0.20) = 0.03 p(e <= 0.25) = 0.05 p(e <= 0.30) = 0.09 p(e <= 0.35) = 0.17 p(e <= 0.40) = 0.25 p(e <= 0.45) = 0.33 p(e <= 0.50) = 0.48 Computing top composite predictors at time 76.75 (without crossvalidation) ... Computing predictors at time 76.75 ... Defining composite predictors ... 1. Pred ITGB8 - Pat: Pos - Sub: 20 Pred: 12 (0.60) Err: 1 (0.08) NErr: 0.25 WErr: 0.08 NWErr: 0.24 Ok: 2,3,6,7,8,9,10,11,12,15,17 Err: 16 2. Pred TWSG1 - Pat: Neg - Sub: 20 Pred: 8 (0.40) Err: 1 (0.12) NErr: 0.35 WErr: 0.12 NWErr: 0.35 Ok: 5,6,7,9,11,17,18 Err: 8 3. Pred GDF11 - Pat: Neg - Sub: 20 Pred: 16 (0.80) Err: 7 (0.44) NErr: 0.45 WErr: 0.40 NWErr: 0.42 Ok: 1,3,6,7,9,12,15,19,20 Err: 4,5,8,10,16,17,18 Testing top composite predictors at time 76.75 ... Best Composite Predictor: Top-2 Score: 0.22 Testing top composite predictors at time 145.25 ... Average Best Composite Predictor: Top-1.95 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 13 (0.65) Err: 4 (0.31) NErr: 0.38 WErr: 0.29 NWErr: 0.36 CScore: 0.36 Ok: 1,2,3,4,6,7,11,12,18 Err: 10,14,16,17 CompPred Top-1 - Sub: 20 Pred: 10 (0.50) Err: 5 (0.50) NErr: 0.50 WErr: 0.46 NWErr: 0.48 CScore: 0.48 Ok: 1,2,3,7,12 Err: 10,13,14,16,17 CompPred Top-2 - Sub: 20 Pred: 14 (0.70) Err: 5 (0.36) NErr: 0.40 WErr: 0.33 NWErr: 0.38 CScore: 0.38 Ok: 1,2,3,4,6,7,11,12,18 Err: 5,10,14,16,17 CompPred Top-3 - Sub: 4 Pred: 4 (1.00) Err: 4 (1.00) NErr: 1.00 WErr: 1.00 NWErr: 0.60 CScore: 0.60 Ok: Err: 15,16,17,20 Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.02 p(e <= 0.15) = 0.02 p(e <= 0.20) = 0.05 p(e <= 0.25) = 0.09 p(e <= 0.30) = 0.13 p(e <= 0.35) = 0.17 p(e <= 0.40) = 0.28 p(e <= 0.45) = 0.41 p(e <= 0.50) = 0.49 Computing top composite predictors at time 145.25 (without crossvalidation) ... Computing predictors at time 145.25 ... Defining composite predictors ... 1. Pred STX17 - Pat: Pos - Sub: 20 Pred: 10 (0.50) Err: 0 (0.00) NErr: 0.25 WErr: 0.00 NWErr: 0.24 Ok: 1,2,3,5,7,12,14,15,17,20 Err: 2. Pred PML - Pat: Pos - Sub: 20 Pred: 13 (0.65) Err: 2 (0.15) NErr: 0.28 WErr: 0.14 NWErr: 0.26 Ok: 1,2,3,4,5,6,7,11,12,18,20 Err: 10,16 Testing top composite predictors at time 145.25 ... Best Composite Predictor: Top-2 Score: 0.18 Results for Composite Single-Gene Predictors with Preselected Genes ------------------------------------------------------------------- Early Predictor: The best ternary logic predictor synthesized, which consists in average of 1.75 single gene-predictors, makes prediction in 9 out of 20 cases with 2 errors (which directly translates into an error rate of 22%). The normalized weighted error rate (based on an equivalent binary predictor) is 38%. The p-value determined using 100 randomization tests is ~ 0.29. The best predictor based on all 20 subjects (i.e. without crossvalidation) contains exactly the following two genes: SAMHD1 (negative) and IGHA1 (positive). Middle Predictor: The best predictor consists in average of 1.1 genes and makes predictions in 10 out of 20 cases with 4 errors (yielding a direct error rate of 40%). The normalized error rate is 42% and the p-value is ~ 0.39. The best predictor based on all 20 subjects consists of one gene: IRF7 (positive). Late Predictor: The best predictor consists in average of 1.00 genes and makes predictions in 10 out of 20 cases with 2 errors (translating into a direct error rate of 20%). The normalized error rate is 32% with a p-value of ~ 0.18. The best predictor based on all 20 subjects consists of one gene: TAP1 (positive). Relevant Log-File Fragments Testing top composite predictors at time 17.0 ... Average Best Composite Predictor: Top-1.75 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 9 (0.45) Err: 2 (0.22) NErr: 0.38 WErr: 0.23 NWErr: 0.38 CScore: 0.38 Ok: 1,4,9,10,14,18,19 Err: 8,20 CompPred Top-1 - Sub: 20 Pred: 11 (0.55) Err: 1 (0.09) NErr: 0.27 WErr: 0.10 NWErr: 0.29 CScore: 0.29 Ok: 1,4,9,10,12,14,15,16,18,19 Err: 20 CompPred Top-2 - Sub: 20 Pred: 10 (0.50) Err: 4 (0.40) NErr: 0.45 WErr: 0.43 NWErr: 0.46 CScore: 0.46 Ok: 4,9,10,14,18,19 Err: 3,6,8,11 CompPred Top-3 - Sub: 7 Pred: 1 (0.14) Err: 0 (0.00) NErr: 0.43 WErr: 0.00 NWErr: 0.47 CScore: 0.47 Ok: 1 Err: Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.00 p(e <= 0.20) = 0.00 p(e <= 0.25) = 0.05 p(e <= 0.30) = 0.11 p(e <= 0.35) = 0.24 p(e <= 0.40) = 0.29 p(e <= 0.45) = 0.41 p(e <= 0.50) = 0.50 Computing top composite predictors at time 17.0 (without crossvalidation) ... Computing predictors at time 17.0 ... Defining composite predictors ... 1. Pred SAMHD1 - Pat: Neg - Sub: 20 Pred: 11 (0.55) Err: 1 (0.09) NErr: 0.27 WErr: 0.10 NWErr: 0.29 Ok: 1,4,9,10,12,14,15,16,18,19 Err: 20 2. Pred IGHA1 - Pat: Pos - Sub: 20 Pred: 9 (0.45) Err: 4 (0.44) NErr: 0.48 WErr: 0.41 NWErr: 0.46 Ok: 1,3,6,11,20 Err: 8,12,15,16 Testing top composite predictors at time 17.0 ... Best Composite Predictor: Top-2 Score: 0.27 Testing top composite predictors at time 76.75 ... Average Best Composite Predictor: Top-1.10 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 10 (0.50) Err: 4 (0.40) NErr: 0.45 WErr: 0.35 NWErr: 0.42 CScore: 0.42 Ok: 1,2,3,7,11,12 Err: 10,13,16,18 CompPred Top-1 - Sub: 20 Pred: 11 (0.55) Err: 5 (0.45) NErr: 0.48 WErr: 0.41 NWErr: 0.45 CScore: 0.45 Ok: 1,2,3,7,11,12 Err: 5,10,13,16,18 Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.00 p(e <= 0.20) = 0.01 p(e <= 0.25) = 0.01 p(e <= 0.30) = 0.05 p(e <= 0.35) = 0.12 p(e <= 0.40) = 0.23 p(e <= 0.45) = 0.39 p(e <= 0.50) = 0.60 Computing top composite predictors at time 76.75 (without crossvalidation) ... Computing predictors at time 76.75 ... Defining composite predictors ... 1. Pred IRF7 - Pat: Pos - Sub: 20 Pred: 12 (0.60) Err: 3 (0.25) NErr: 0.35 WErr: 0.22 NWErr: 0.33 Ok: 1,2,3,5,7,9,11,12,18 Err: 10,13,16 Testing top composite predictors at time 76.75 ... Best Composite Predictor: Top-1 Score: 0.33 Testing top composite predictors at time 145.25 ... Average Best Composite Predictor: Top-1.00 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 10 (0.50) Err: 2 (0.20) NErr: 0.35 WErr: 0.17 NWErr: 0.32 CScore: 0.32 Ok: 1,2,3,6,7,11,12,20 Err: 10,16 CompPred Top-1 - Sub: 20 Pred: 10 (0.50) Err: 2 (0.20) NErr: 0.35 WErr: 0.17 NWErr: 0.32 CScore: 0.32 Ok: 1,2,3,6,7,11,12,20 Err: 10,16 CompPred Top-2 - Sub: 3 Pred: 2 (0.67) Err: 1 (0.50) NErr: 0.50 WErr: 0.45 NWErr: 0.49 CScore: 0.49 Ok: 14 Err: 10 Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.00 p(e <= 0.20) = 0.02 p(e <= 0.25) = 0.07 p(e <= 0.30) = 0.13 p(e <= 0.35) = 0.18 p(e <= 0.40) = 0.32 p(e <= 0.45) = 0.43 p(e <= 0.50) = 0.54 Computing top composite predictors at time 145.25 (without crossvalidation) ... Computing predictors at time 145.25 ... Defining composite predictors ... 1. Pred TAP1 - Pat: Pos - Sub: 20 Pred: 11 (0.55) Err: 1 (0.09) NErr: 0.27 WErr: 0.08 NWErr: 0.25 Ok: 1,2,3,6,7,9,11,12,14,20 Err: 16 Testing top composite predictors at time 145.25 ... Best Composite Predictor: Top-1 Score: 0.25 Preliminary Results for Composite Cluster Predictors ---------------------------------------------------- Early Predictors: The best computed ternary predictor consists in average of 3.95 (sub)clusters. It makes predictions in 14 out of 20 cases with 8 errors (translating into 57% error rate). The normalized error rate is 57%. The best predictor based on all 20 subjects consists of 5 (sub)clusters named 0.48, 2.2 (both negative), 0.1, 0.70, 0.74 (all positive). See table below for the corresponding sets of genes. Middle Predictors: The best ternary predictor consists in average of 4.60 (sub)clusters. It makes predictions in 14 out of 20 cases with 4 errors (yielding an error rate of 29%). The normalized error rate is 33% and the p-value is ~ 0.15. The best predictor based on all 20 subjcts has 5 (sub)clusters: 0.54, 0.17, 3.1, 1.0 (all positive), and 8.0 (negative). Late Predictors: The best ternary predictor consists in average of 2.05 (sub)clusters. It makes predictions in 20 out of 20 cases with 7 errors (yielding an error rate of 35%). The normalized error rate is 33% with a p-value of ~ 0.16. The best predictor based on all 20 subjects has 2 (sub)clusters: 0.55 and 1.0 (positive). Table of Relevant Clusters subcluster 0.48: { FKBP5 SAMHD1 } subcluster 2.2: { GPR64 DKK2 IL17A NDN } subcluster 0.1: { WDR6 ITPR3 EVL SIGIRR TNFRSF25 CDC25B } subcluster 0.70: { IGKV4-1 IGKC TXNDC5 } subcluster 0.74: { C2CD2 NFS1 } subcluster 0.54: { TREX1 LY6E } subcluster 0.17: { IRF9 IFI44 TRAFD1 TDRD7 DDX58 LOC26010 GBP1 LAMP3 IFI35 IFIT1 IRF7 MX1 OAS1 OAS3 LAP3 HERC5 RTP4 IFIH1 PARP12 OASL RSAD2 UBE2L6 ISG15 SCO2 } subcluster 3.1: { GML APOD NF1 } subcluster 1.0: { SAMD4A RHBDD3 } subcluster 8.0: { TSTA3 CTNNAL1 } subcluster 0.55: { TREX1 CNP ALDH3B1 LY6E MARCKSL1 SRC } Relevant Log-File Fragments Testing top composite predictors at time 17.0 ... Average Best Composite Predictor: Top-3.95 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 14 (0.70) Err: 8 (0.57) NErr: 0.55 WErr: 0.59 NWErr: 0.57 CScore: 0.57 Ok: 1,5,9,10,14,15 Err: 2,3,7,11,12,13,17,20 CompPred Top-1 - Sub: 20 Pred: 9 (0.45) Err: 2 (0.22) NErr: 0.38 WErr: 0.25 NWErr: 0.39 CScore: 0.39 Ok: 1,9,10,14,15,18,19 Err: 12,20 CompPred Top-2 - Sub: 20 Pred: 11 (0.55) Err: 4 (0.36) NErr: 0.42 WErr: 0.39 NWErr: 0.44 CScore: 0.44 Ok: 1,5,9,10,14,15,19 Err: 2,8,11,20 CompPred Top-3 - Sub: 20 Pred: 15 (0.75) Err: 7 (0.47) NErr: 0.48 WErr: 0.48 NWErr: 0.49 CScore: 0.49 Ok: 1,5,6,9,10,14,15,19 Err: 2,7,11,12,13,17,20 CompPred Top-4 - Sub: 20 Pred: 16 (0.80) Err: 10 (0.62) NErr: 0.60 WErr: 0.64 NWErr: 0.61 CScore: 0.61 Ok: 1,5,9,10,14,15 Err: 2,3,4,7,11,12,13,17,18,20 CompPred Top-5 - Sub: 19 Pred: 14 (0.74) Err: 8 (0.57) NErr: 0.55 WErr: 0.59 NWErr: 0.56 CScore: 0.56 Ok: 1,5,9,10,14,15 Err: 2,3,4,11,12,13,17,20 CompPred Top-6 - Sub: 8 Pred: 7 (0.88) Err: 2 (0.29) NErr: 0.31 WErr: 0.30 NWErr: 0.43 CScore: 0.43 Ok: 1,6,10,14,15 Err: 3,20 CompPred Top-7 - Sub: 1 Pred: 1 (1.00) Err: 0 (0.00) NErr: 0.00 WErr: 0.00 NWErr: 0.47 CScore: 0.47 Ok: 1 Err: Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.00 p(e <= 0.20) = 0.02 p(e <= 0.25) = 0.04 p(e <= 0.30) = 0.08 p(e <= 0.35) = 0.18 p(e <= 0.40) = 0.30 p(e <= 0.45) = 0.47 p(e <= 0.50) = 0.57 Computing top composite predictors at time 17.0 (without crossvalidation) ... Computing predictors at time 17.0 ... Defining composite predictors ... 1. Pred 0.48 - Pat: Neg - Sub: 20 Pred: 9 (0.45) Err: 1 (0.11) NErr: 0.32 WErr: 0.12 NWErr: 0.33 Ok: 1,9,10,12,14,15,18,19 Err: 20 2. Pred 2.2 - Pat: Neg - Sub: 20 Pred: 11 (0.55) Err: 3 (0.27) NErr: 0.38 WErr: 0.28 NWErr: 0.38 Ok: 1,5,6,8,9,10,19,20 Err: 11,12,18 3. Pred 0.1 - Pat: Pos - Sub: 20 Pred: 8 (0.40) Err: 2 (0.25) NErr: 0.40 WErr: 0.26 NWErr: 0.41 Ok: 10,13,17,18,19,20 Err: 8,12 4. Pred 0.70 - Pat: Pos - Sub: 20 Pred: 11 (0.55) Err: 4 (0.36) NErr: 0.42 WErr: 0.38 NWErr: 0.43 Ok: 1,3,6,14,15,16,20 Err: 2,7,11,12 5. Pred 0.74 - Pat: Pos - Sub: 20 Pred: 11 (0.55) Err: 4 (0.36) NErr: 0.42 WErr: 0.40 NWErr: 0.44 Ok: 2,4,9,10,17,19,20 Err: 3,6,11,12 Testing top composite predictors at time 17.0 ... Best Composite Predictor: Top-5 Score: 0.24 0.48 { FKBP5 SAMHD1 } 2.2[2.1[2.0]] { GPR64 DKK2 IL17A NDN } 0.1 { WDR6 ITPR3 EVL SIGIRR TNFRSF25 CDC25B } 0.70[0.51] { IGKV4-1 IGKC TXNDC5 } 0.74 { C2CD2 NFS1 } Testing top composite predictors at time 76.75 ... Average Best Composite Predictor: Top-4.60 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 14 (0.70) Err: 4 (0.29) NErr: 0.35 WErr: 0.26 NWErr: 0.33 CScore: 0.33 Ok: 1,2,3,4,7,9,11,14,17,18 Err: 10,15,16,19 CompPred Top-1 - Sub: 20 Pred: 9 (0.45) Err: 4 (0.44) NErr: 0.48 WErr: 0.42 NWErr: 0.46 CScore: 0.46 Ok: 2,3,5,7,11 Err: 10,12,13,16 CompPred Top-2 - Sub: 20 Pred: 14 (0.70) Err: 5 (0.36) NErr: 0.40 WErr: 0.33 NWErr: 0.38 CScore: 0.38 Ok: 1,2,3,4,5,7,9,11,17 Err: 10,13,15,16,19 CompPred Top-3 - Sub: 20 Pred: 16 (0.80) Err: 7 (0.44) NErr: 0.45 WErr: 0.41 NWErr: 0.43 CScore: 0.43 Ok: 1,2,3,4,7,9,11,12,17 Err: 10,13,15,16,18,19,20 CompPred Top-4 - Sub: 20 Pred: 15 (0.75) Err: 5 (0.33) NErr: 0.38 WErr: 0.30 NWErr: 0.35 CScore: 0.35 Ok: 1,2,3,4,7,9,11,12,14,17 Err: 10,13,15,16,19 CompPred Top-5 - Sub: 20 Pred: 18 (0.90) Err: 8 (0.44) NErr: 0.45 WErr: 0.43 NWErr: 0.43 CScore: 0.43 Ok: 1,2,3,4,7,9,11,14,17,18 Err: 5,6,10,13,15,16,19,20 CompPred Top-6 - Sub: 15 Pred: 12 (0.80) Err: 3 (0.25) NErr: 0.30 WErr: 0.23 NWErr: 0.34 CScore: 0.34 Ok: 2,3,4,7,9,11,14,17,18 Err: 5,10,19 CompPred Top-7 - Sub: 3 Pred: 3 (1.00) Err: 0 (0.00) NErr: 0.00 WErr: 0.00 NWErr: 0.43 CScore: 0.43 Ok: 7,9,18 Err: Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.00 p(e <= 0.20) = 0.00 p(e <= 0.25) = 0.01 p(e <= 0.30) = 0.09 p(e <= 0.35) = 0.15 p(e <= 0.40) = 0.28 p(e <= 0.45) = 0.39 p(e <= 0.50) = 0.47 Computing top composite predictors at time 76.75 (without crossvalidation) ... Computing predictors at time 76.75 ... Defining composite predictors ... 1. Pred 0.54 - Pat: Pos - Sub: 20 Pred: 12 (0.60) Err: 3 (0.25) NErr: 0.35 WErr: 0.23 NWErr: 0.34 Ok: 2,3,4,5,7,9,11,12,17 Err: 10,13,16 2. Pred 0.17 - Pat: Pos - Sub: 20 Pred: 11 (0.55) Err: 3 (0.27) NErr: 0.38 WErr: 0.24 NWErr: 0.35 Ok: 1,2,3,5,7,11,12,14 Err: 10,13,16 3. Pred 3.1 - Pat: Pos - Sub: 20 Pred: 11 (0.55) Err: 3 (0.27) NErr: 0.38 WErr: 0.27 NWErr: 0.37 Ok: 7,9,10,11,13,14,18,20 Err: 12,16,19 4. Pred 1.0 - Pat: Pos - Sub: 20 Pred: 10 (0.50) Err: 3 (0.30) NErr: 0.40 WErr: 0.29 NWErr: 0.39 Ok: 2,3,7,9,14,17,18 Err: 8,16,20 5. Pred 8.0 - Pat: Neg - Sub: 20 Pred: 11 (0.55) Err: 4 (0.36) NErr: 0.42 WErr: 0.39 NWErr: 0.44 Ok: 5,9,10,13,15,17,20 Err: 3,11,12,19 6. Pred 0.69 - Pat: Pos - Sub: 20 Pred: 13 (0.65) Err: 5 (0.38) NErr: 0.42 WErr: 0.42 NWErr: 0.45 Ok: 4,5,7,9,13,15,17,18 Err: 6,11,14,16,20 Testing top composite predictors at time 76.75 ... Best Composite Predictor: Top-5 Score: 0.24 0.54 { TREX1 LY6E } 0.17[0.15[0.12[0.10[0.5[0.3[0.0]]]]]] { IRF9 IFI44 TRAFD1 TDRD7 DDX58 LOC26010 GBP1 LAMP3 IFI35 IFIT1 IRF7 MX1 OAS1 OAS3 LAP3 HERC5 RTP4 IFIH1 PARP12 OASL RSAD2 UBE2L6 ISG15 SCO2 } 3.1[3.0] { GML APOD NF1 } 1.0 { SAMD4A RHBDD3 } 8.0 { TSTA3 CTNNAL1 } 0.69 { MED22 GSDMDC1 } Testing top composite predictors at time 145.25 ... Average Best Composite Predictor: Top-2.05 Validation results for top composite predictors ... CompPred Best - Sub: 20 Pred: 20 (1.00) Err: 7 (0.35) NErr: 0.35 WErr: 0.33 NWErr: 0.33 CScore: 0.33 Ok: 1,2,3,4,5,6,7,9,11,12,15,18,20 Err: 8,10,13,14,16,17,19 CompPred Top-1 - Sub: 20 Pred: 14 (0.70) Err: 2 (0.14) NErr: 0.25 WErr: 0.13 NWErr: 0.24 CScore: 0.24 Ok: 2,3,4,5,6,7,9,11,12,15,18,20 Err: 10,16 CompPred Top-2 - Sub: 20 Pred: 20 (1.00) Err: 7 (0.35) NErr: 0.35 WErr: 0.33 NWErr: 0.33 CScore: 0.33 Ok: 1,2,3,4,5,6,7,9,11,12,15,18,20 Err: 8,10,13,14,16,17,19 CompPred Top-3 - Sub: 20 Pred: 16 (0.80) Err: 6 (0.38) NErr: 0.40 WErr: 0.34 NWErr: 0.37 CScore: 0.37 Ok: 1,2,3,4,6,7,9,11,12,15 Err: 8,10,13,16,17,19 CompPred Top-4 - Sub: 1 Pred: 1 (1.00) Err: 0 (0.00) NErr: 0.00 WErr: 0.00 NWErr: 0.48 CScore: 0.48 Ok: 9 Err: Randomization Summary ... Nullhypothesis: computed predictor can predict random classification with error rate <= x Composite Predictor Best - Tests: 100 - p(e <= 0.05) = 0.00 p(e <= 0.10) = 0.00 p(e <= 0.15) = 0.01 p(e <= 0.20) = 0.01 p(e <= 0.25) = 0.03 p(e <= 0.30) = 0.09 p(e <= 0.35) = 0.16 p(e <= 0.40) = 0.26 p(e <= 0.45) = 0.36 p(e <= 0.50) = 0.50 Computing top composite predictors at time 145.25 (without crossvalidation) ... Computing predictors at time 145.25 ... Defining composite predictors ... 1. Pred 0.55 - Pat: Pos - Sub: 20 Pred: 14 (0.70) Err: 2 (0.14) NErr: 0.25 WErr: 0.13 NWErr: 0.24 Ok: 2,3,4,5,6,7,9,11,12,15,18,20 Err: 10,16 2. Pred 1.0 - Pat: Pos - Sub: 20 Pred: 12 (0.60) Err: 3 (0.25) NErr: 0.35 WErr: 0.22 NWErr: 0.33 Ok: 1,2,3,6,7,9,11,14,15 Err: 8,10,16 3. Pred 0.15 - Pat: Pos - Sub: 20 Pred: 15 (0.75) Err: 5 (0.33) NErr: 0.38 WErr: 0.29 NWErr: 0.34 Ok: 1,2,3,6,7,9,11,12,14,20 Err: 10,13,16,17,19 Testing top composite predictors at time 145.25 ... Best Composite Predictor: Top-2 Score: 0.20 0.55[0.54,0.44[0.24]] { TREX1 CNP ALDH3B1 LY6E MARCKSL1 SRC } 1.0 { SAMD4A RHBDD3 } 0.15[0.12[0.10[0.5[0.3[0.0]]]]] { IRF9 IFI44 TRAFD1 DDX58 LOC26010 GBP1 LAMP3 IFI35 IRF7 MX1 OAS1 OAS3 LAP3 HERC5 RTP4 IFIH1 PARP12 OASL RSAD2 UBE2L6 ISG15 SCO2 }