Skip to main content

Table 2 Results of the evolutionary simulations

From: Accurate and fast methods to estimate the population mutation rate from error prone sequences

Evolutionary simulations

Coalescent/mutation models

Watterson and Tajima estimators

θ

n

ε

θ FW

θ KM

θ' KM

θ W

θ' W

θ T

θ' T

A1: 1

8

0.0

0.96 ± 0.10

0.76 ± 0.10

1.10 ± 0.15

0.98 ± 0.10

1.12 ± 0.16

1.02 ± 0.12

1.10 ± 0.15

A2: 1

8

0.5

3.49 ± 0.23

0.90 ± 0.14

0.96 ± 0.15

2.63 ± 0.1 5

0.94 ± 0.15

2.04 ± 0.13

0.95 ± 0.16

A3: 1

16

0.0

1.02 ± 0.10

0.84 ± 0.09

1.01 ± 0.12

1.00 ± 0.09

0.99 ± 0.11

0.97 ± 0.10

0.96 ± 0.12

A4: 1

16

0.5

5.10 ± 0.28

0.99 ± 0.12

1.02 ± 0.13

3.43 ± 0.16

1.03 ± 0.13

2.01 ± 0.13

1.01 ± 0.13

B1: 2

8

0.0

1.92 ± 0.17

1.51 ± 0.17

1.80 ± 0.25

1.92 ± 0.18

1.81 ± 0.25

1.89 ± 0.20

1.82 ± 0.26

B2: 2

8

0.5

4.38 ± 0.26

1.80 ± 0.20

1.92 ± 0.26

3.44 ± 0.19

1.91 ± 0.26

2.91 ± 0.20

1.92 ± 0.25

B3: 2

16

0.0

2.06 ± 0.16

1.78 ± 0.15

2.02 ± 0.19

2.04 ± 0.16

2.04 ± 0.20

2.09 ± 0.19

2.10 ± 0.22

B4: 2

16

0.5

6.19 ± 0.33

1.88 ± 0.17

1.92 ± 0.19

4.32 ± 0.20

1.92 ± 0.19

2.95 ± 0.20

1.96 ± 0.22

C1: 4

8

0.0

-

-

4.05 ± 0.44

4.27 ± 0.35

4.27 ± 0.53

4.31 ± 0.39

4.32 ± 0.54

C2: 4

8

0.5

-

-

3.89 ± 0.46

5.62 ± 0.33

4.09 ± 0.53

5.12 ± 0.38

4.13 ± 0.54

C3: 4

16

0.0

-

-

4.00 ± 0.31

4.12 ± 0.28

4.21 ± 0.38

4.24 ± 0.36

4.28 ± 0.41

C4: 4

16

0.5

-

-

3.82 ± 0.31

6.23 ± 0.27

3.87 ± 0.35

4.84 ± 0.32

3.86 ± 0.38

D1: 8

8

0.0

-

-

7.88 ± 0.71

7.75 ± 0.49

7.87 ± 0.78

7.78 ± 0.55

7.85 ± 0.77

D2: 8

8

0.5

-

-

7.90 ± 0.85

9.67 ± 0.63

8.38 ± 1.03

9.22 ± 0.73

8.39 ± 1.03

D3: 8

16

0.0

-

-

7.82 ± 0.54

8.08 ± 0.49

8.20 ± 0.72

8.26 ± 0.69

8.32 ± 0.82

D4: 8

16

0.5

-

-

8.00 ± 0.58

10.73 ± 0.52

8.31 ± 0.71

9.57 ± 0.68

8.60 ± 0.81

  1. These results are for the standard FW model (θ FW ) and the original and new Watterson estimators (θ W and θ' W ), Tajima estimators (θ T and θ' T ), and KM and KM' models (θ KM and θ' KM ). They are summarized as the means ± twice their standard deviations for 200 simulated datasets apiece for each of the 16 different combinations of θ, n (number of sampled sequences), and ε (A1 to D4). Mean estimates that are significantly greater than or less than the true θ are highlighted in italics and boldface, respectively. No results are provided for the FW and KM models with θ = 4 or 8 (lower left corner), since these calculations are too computationally intensive (see text).