Skip to main content

Table 4 Log-rank test p-values for combinations of methods with 500 features

From: Comparison of cancer subtype identification methods combined with feature selection methods in omics data analysis

FS

Dataset

Subtyping Identification Methods

CC

PINS

NMF

ICB

SNF

NEMO

MED

AML

0.42 (4)

0.38 (3)

0.13 (3)

0.12 (4)

0.17 (2)

2.0E-03 (7)

GBM

0.01 (4)

0.40 (3)

0.02 (3)

0.01 (4)

0.83 (2)

6.6E-04 (6)

BIC

0.35 (4)

0.37 (2)

0.45 (2)

0.17 (5)

0.30 (2)

0.35 (6)

COAD

0.39 (6)

0.77 (2)

0.19 (3)

0.18 (5)

0.86 (2)

0.31 (8)

VAR

AML

9.0E-04 (3)

0.10 (5)

0.06 (5)

3.0E-03 (4)

0.18 (2)

0.01 (5)

GBM

0.09 (3)

0.73 (2)

0.09 (2)

0.06 (5)

0.73 (2)

7.0E-07 (6)

BIC

0.04 (4)

0.52 (2)

0.53 (2)

0.14 (5)

0.06 (4)

0.09 (4)

COAD

0.05 (4)

0.96 (2)

0.01 (2)

0.32 (4)

0.44 (2)

0.29 (8)

MAD

AML

1.0E-03 (4)

0.05 (2)

0.04 (4)

1.0E-04 (4)

0.21 (2)

0.08 (6)

GBM

0.09 (3)

2.2E-05 (5)

0.05 (2)

0.05 (5)

0.04 (3)

1.0E-03 (6)

BIC

0.32 (6)

0.24 (3)

0.06 (2)

0.09 (4)

0.06 (4)

0.12 (4)

COAD

0.55 (2)

0.09 (3)

2.0E-03 (3)

0.06 (5)

0.10 (3)

0.46 (9)

DIP

AML

0.01 (6)

0.75 (2)

0.66 (2)

0.39 (4)

0.62 (2)

0.58 (2)

GBM

0.01 (4)

0.02 (6)

0.23 (5)

0.02 (5)

0.20 (2)

0.08 (9)

BIC

0.06 (2)

0.92 (2)

0.49 (2)

0.68 (4)

0.74 (2)

0.48 (3)

COAD

0.27 (2)

4.0E-03 (2)

0.02 (2)

0.02 (3)

0.05 (2)

0.00 (5)

  1. Value in parentheses indicates the number of clusters \(k\)