Skip to main content

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

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

0.422

0.425

0.355

0.340

0.435

0.325

 T.model

1

0

1

0.625

0.650

0.650

 Tint.model

1

0.075

1

0.625

0.650

0.650

 Sen.

1

0.783

1

0.974

0.977

0.976

 Spe.

1

0.999

0.954

0.964

0.775

0.970

 S.model

45

37.895

197.165

160.375

781.805

142.245

 Deviance

− 128.514

− 108.289

− 281.603

− 282.783

− 50.658

− 294.313

 c-index

0.925

0.891

0.984

0.983

0.855

0.985

censoring rate = 65%

 RMSE.M

0.421

0.424

0.375

0.364

0.436

0.348

 T.model

1

0

1

0.805

0.815

0.815

 Tint.model

1

0.070

1

0.805

0.815

0.815

 Sen.

1

0.764

1

0.986

0.988

0.987

 Spe.

1

0.999

0.962

0.965

0.686

0.968

 S.model

45

37.855

170.655

157.450

1072.52

148.745

 Deviance

−123.803

−102.527

−231.398

− 240.500

− 45.181

−250.026

 c-index

0.928

0.898

0.983

0.984

0.849

0.986