Skip to main content

Table 2 LS-SVM models for the prediction of WHEELER, pN-STAGE and CRM in rectal cancer

From: A kernel-based integration of genome-wide data for clinical decision support

Outcome

Model

NG*

NP†

AUC (SE)‡

p-value§

WHEELER

     

   A

MT 0

4

 

0.7538 (0.1085)

0.0987

 

MT 1

29

 

0.9038 (0.0502)

0.6861

 

PT 0

 

35

0.7423 (0.0867)

0.0540

 

PT 1

 

11

0.9038 (0.0575)

0.7273

   B

MT0-T1

32

 

0.6846 (0.1215)

0.0598

 

PT0-T1

 

5

0.8654 (0.0621)

0.4135

   C

MT 01

3¶

 

0.7808 (0.0985)

0.1320

 

PT 01

 

21¶

0.7692 (0.0831)

0.0831

 

MPT 0

3

35

0.8461 (0.0718)

0.2760

 

MPT 1

25

12

0.9269 (0.0425)

 
 

MPT 01

2¶

31¶

0.8846 (0.0558)

0.4858

 

MT 0 PT 1

2

4

0.9385 (0.0444)

0.8101Â¥

pN-STAGE

     

   A

MT 0

25

 

0.6493 (0.0914)

2.315e-4

 

MT 1

22

 

0.8506 (0.0665)

0.0362

 

PT 0

 

2

0.6753 (0.0906)

6.659e-4

 

PT 1

 

12

0.8409 (0.0652)

0.0238

   B

MT0-T1

4

 

0.6071 (0.0986)

1.359e-4

 

PT0-T1

 

9

0.7662 (0.0900)

0.0153

   C

MT 01

24¶

 

0.9286 (0.0450)

0.1998

 

PT 01

 

34¶

0.8182 (0.0695)

0.0145

 

MPT 0

27

27

0.9188 (0.0469)

0.1591

 

MPT 1

21

14

0.9870 (0.0135)

 
 

MPT 01

23¶

16¶

0.9610 (0.0280)

0.3421

 

MT 0 PT 01

26

20¶

1 (0)

0.3347Â¥

CRM

     

   A

MT 0

33

 

0.6790 (0.1016)

0.0072

 

MT 1

9

 

0.9259 (0.0472)

0.4955

 

PT 0

 

34

0.8518 (0.0624)

0.0935

 

PT 1

 

34

0.7654 (0.0831)

0.0281

   B

MT0-T1

6

 

0.9136 (0.0480)

0.4030

 

PT0-T1

 

2

0.8272 (0.0709)

0.0849

   C

MT 01

16¶

 

0.8066 (0.0846)

0.0468

 

PT 01

 

3¶

0.7531 (0.0865)

0.0227

 

MPT 0

7

27

0.8477 (0.0688)

0.1340

 

MPT 1

7

33

0.9630 (0.0344)

 
 

MPT 01

2¶

3¶

0.8230 (0.0771)

0.0973

 

MT 1 PT 0

16

14

0.9630 (0.0376)

1

 

MT 01 PT 1

9¶

29

0.9876 (0.0146)

0.4924Â¥

  1. *Number of genes selected in each LOO iteration. †Number of proteins selected in each LOO iteration. ‡Area under the ROC curve (standard error) obtained with leave-one-out. §Comparison of AUC between each model and the best model in bold [46]. ¶Number of features used at both time points. ¥This model is better than the model in bold we compare with.