Skip to main content

Table 4 Quartiles for πcc values when classifying 23 non-sex-specific tumor types separately using male and female samples across 1000 GA/KNN runs for each of five training/testing partitions (5000 runs total)

From: A comprehensive genomic pan-cancer classification using The Cancer Genome Atlas gene expression data

Type

Minimum

1st Quartile

3rd Quartile

Maximum

Modal Prediction Accuracy

F

M

F

M

F

M

F

M

F

M

ACC

0.01

0.01

0.83

0.61

0.95

0.84

1.00

0.95

0.97

0.93

BLCA

0.01

0.04

0.46

0.66

0.91

0.96

1.00

1.00

0.89

0.93

COAD

0.22

0.12

0.83

0.83

0.93

0.91

1.00

0.96

1.00

0.99

GBM

0.35

0.59

0.91

0.96

0.99

1.00

1.00

1.00

1.00

1.00

HNSC

0.54

0.06

0.93

0.96

1.00

1.00

1.00

1.00

1.00

1.00

KIRC

0.00

0.00

0.98

0.98

1.00

1.00

1.00

1.00

0.95

0.98

KIRP

0.00

0.00

0.51

0.93

0.92

1.00

1.00

1.00

0.89

0.92

LAML

0.93

0.83

1.00

0.99

1.00

1.00

1.00

1.00

1.00

1.00

LGG

0.72

0.16

0.99

0.99

1.00

1.00

1.00

1.00

1.00

1.00

LIHC

0.01

0.01

0.93

0.98

1.00

1.00

1.00

1.00

0.99

0.99

LUAD

0.06

0.05

0.90

0.80

0.99

0.96

1.00

1.00

0.96

0.97

LUSC

0.00

0.03

0.49

0.85

0.94

0.99

1.00

1.00

0.86

0.94

MESO

0.03

0.00

0.50

0.78

0.79

0.95

0.89

1.00

0.92

0.95

PAAD

0.00

0.08

0.84

0.83

0.99

0.99

1.00

1.00

0.95|

0.93

PCPG

0.14

0.80

0.98

0.97

1.00

1.00

1.00

1.00

0.98

1.00

READ

0.01

0.03

0.08

0.09

0.15

0.17

0.33

0.23

0.00

0.00

SARC

0.17

0.06

0.88

0.83

0.98

0.95

1.00

1.00

1.00

0.99

SKCM

0.15

0.01

0.89

0.90

0.98

0.98

1.00

1.00

0.98

0.97

THCA

0.81

0.71

1.00

0.96

1.00

1.00

1.00

1.00

1.00

1.00

THYM

0.09

0.08

0.85

0.89

1.00

0.99

1.00

1.00

0.98

0.92

UVM

0.76

0.28

0.95

0.92

0.99

0.99

1.00

1.00

1.00

0.97

ESCA

0.00

0.05

0.02

0.35

0.61

0.97

0.80

1.00

0.38

0.64

STAD

0.07

0.08

0.84

0.80

0.98

0.96

1.00

1.00

0.98

0.97

  1. The rightmost column labeled “overall” is not based on πcc but instead on a prediction using the tumor type to which each sample was assigned most often