Skip to main content

Table 5 Results of Simulation 1 (3): The performances of OGS compared with other approaches under coexistence of within- and between-pathway gene-gene interactions

From: Overlapping group screening for detection of gene-gene interactions: application to gene expression profiles with survival trait

 

Oracle

Uni. Sel.

Ordinary

Lasso

TS-GSIS

Lasso

OGS

Ridge

OGS

Lasso

censoring rate = 50%

 RMSE.M

0.457

0.463

0.400

0.430

0.471

0.393

 T.model

1

0

1

0.500

0.525

0.525

 Tint.model

1

0

1

0.500

0.525

0.525

 Sen.

1

0.708

1

0.959

0.966

0.963

 Spe.

1

0.999

0.956

0.968

0.672

0.969

 S.model

48

36.970

191.050

151.135

1119.675

146.475

 Deviance

−127.602

−95.760

− 266.727

− 206.468

−46.208

− 254.344

 c-index

0.924

0.871

0.983

0.961

0.826

0.978

censoring rate = 65%

 RMSE.M

0.455

0.461

0.418

0.412

0.471

0.401

 T.model

1

0

1

0.675

0.715

0.715

 Tint.model

1

0

1

0.675

0.715

0.715

 Sen.

1

0.694

1

0.973

0.980

0.977

 Spe.

1

0.999

0.963

0.968

0.614

0.969

 S.model

48

37.070

168.335

150.420

1308.87

147.785

 Deviance

− 126.277

−92.721

− 220.297

− 223.262

−43.936

−235.996

 c-index

0.929

0.878

0.979

0.978

0.831

0.982