Skip to main content

Table 1 Comparison of estimation and prediction performance of different approaches.

From: Supervised group Lasso with applications to microarray data analysis

  

Lasso

Simple

GLasso

SGLasso

Colon

Nonzero

19

500

500

22

 

Cluster

-

9

9

8

 

Prediction

0.129

0.226

0.161

0.129

Nodal

Nonzero

37

500

233

66

 

Cluster

-

20

9

17

 

Prediction

0.245

0.163

0.122

0.122

Follicular

Nonzero

15

729

233

79

 

Cluster

-

34

2

13

 

Prediction

5.9

2.3

0.5

6.5

MCL

Nonzero

15

834

132

28

 

Cluster

-

30

3

3

 

Prediction

8.2

6.2

19.3

20.3

  1. Nonzero: number of genes in the final models. Cluster: number of clusters in the final models. Prediction: for Colon and Nodal, Leave-One-Out prediction error; For Follicular and MCL, the logrank statistic.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us