Skip to main content

Table 5 Analysis of hot spot predictions for each amino acid type

From: Prediction of hot spot residues at protein-protein interfaces by combining machine learning and energy-based methods

Mutated amino acid

N mut

TP

TN

FP

FN

P

R

F1

A

Arg

33

5

16

10

2

0.33

0.71

0.45

0.64

Asn

22

3

15

1

3

0.75

0.50

0.60

0.82

Asp

29

7

15

5

2

0.58

0.78

0.67

0.76

Cys

1

0

1

0

0

NA

NA

NA

1.00

Gln

21

1

17

2

1

0.33

0.50

0.40

0.86

Glu

31

1

22

4

4

0.20

0.20

0.20

0.74

His

13

1

11

1

0

0.50

1.00

0.67

0.92

Ile

15

0

11

0

4

NA

0.00

NA

0.73

Leu

10

0

9

0

1

NA

0.00

0.00

0.90

Lys

32

10

20

1

1

0.91

0.91

0.91

0.94

Met

2

0

1

1

0

0.00

NA

NA

0.50

Phe

11

1

8

1

1

0.50

0.50

0.50

0.82

Ser

28

1

22

5

0

0.17

1.00

0.29

0.82

Thr

24

0

21

2

1

0.00

0.00

0.00

0.88

Trp

23

5

13

1

4

0.83

0.56

0.67

0.78

Tyr

44

17

17

7

3

0.71

0.85

0.77

0.77

Val

10

1

7

1

1

0.50

0.50

0.50

0.80

All

349

53

226

42

28

0.56

0.65

0.60

0.80

  1. N mut : number of mutations, TP: true positives, TN: true negatives, FP: false positives, FN = false negatives; P: precision, R: recall, F1: F1 score, A: accuracy = .