Skip to main content

Table 8 Scores obtained for the primary dataset using cross validation

From: Environmental metagenome classification for constructing a microbiome fingerprint

  

AKL

HAM

NYC

OFA

PXO

SAC

SCL

TOK

Total

Ryan [21]

#correct

7

10

25

20

60

16

18

20

\(\sum =176\)

 

PPV

0.54

0.56

0.96

0.95

0.98

1

1

1

Ns=193

 

TPR

0.47

0.63

0.96

1

1

1

0.9

1

ACC=0.912

Sanchez et al. [24]

#correct

9

11

110

17

60

34

17

20

\({\sum }=278\)

 

PPV

0.69

0.73

0.95

0.89

1

0.83

0.89

0.71

Ns=311

 

TPR

0.6

0.69

0.87

0.85

1

1

0.85

1

ACC=0.894

Harris et al. [32]

—

—

—

—

—

—

—

—

—

Ns=N/A

          

ACC=0.897

Walker and Datta [22]

TPR (median)

0.6

0.62

0.58

0.95

0.87

0.76

0.3

0.7

Ns=211

 

—

—

—

—

—

—

—

—

—

ACC=0.71

Zhu [25]

#correct

5

3

114

14

51

31

17

15

\({\sum }=250\)

 

TPR

0.33

0.19

0.9

0.74

0.85

0.91

0.85

0.75

Ns=310

          

ACC=0.81

Chierici et al. [23]

—

—

—

—

—

—

—

—

—

Ns=311

          

ACC=0.894

Our method using Mash

#correct

15

15

50

20

60

31

19

20

\({\sum }=230\)

sketch size=1000

PPV

0.34

0.26

1.00

0.67

1.00

1.00

1.00

1.00

Ns=311

 

TPR

1.00

0.94

0.40

1.00

1.00

0.91

0.95

1.00

ACC=0.740

Our method using Mash

#correct

15

16

42

20

60

34

20

20

\({\sum }=227\)

sketch size=10000

PPV

0.65

0.18

1.00

0.83

1.00

1.00

1.00

1.00

Ns=311

 

TPR

1.00

1.00

0.33

1.00

1.00

1.00

1.00

1.00

ACC=0.730

Our method using Mash

#correct

15

16

44

20

60

34

19

20

\({\sum }=228\)

sketch size=100000

PPV

0.60

0.18

1.00

1.00

1.00

1.00

1.00

1.00

Ns=311

 

TPR

1.00

1.00

0.35

1.00

1.00

1.00

0.95

1.00

ACC=0.733

Our method using CoMeta

#correct

4

12

116

20

37

34

13

20

\({\sum }=256\)

(class-level filtering)

PPV

0.67

0.63

0.92

0.74

1.00

0.97

1

0.42

Ns=311

 

TPR

0.27

0.75

0.92

1.00

0.62

1.00

0.65

1.00

ACC=0.823

Our method using CoMeta

#correct

4

13

113

20

57

30

16

19

\({\sum }=272\)

(sample-level filtering)

PPV

0.57

0.65

0.92

0.65

0.92

1.00

0.94

0.9

Ns=311

 

TPR

0.27

0.81

0.9

1.00

0.95

0.88

0.8

0.95

ACC=0.875

  1. We report the number of correctly classified samples (#correct), precision (PPV), and recall (TPR) for each class, as well as the overall accuracy (ACC). Some of the values are missing, as they were not reported in the referenced papers. Also, we show the number of samples (Ns), as in some works, the results for a subset of all of Ns=311 samples were reported