Table 2

Accuracy results for the mean 30 AA COG simulation

range

ML μ

PP μ

ML σ

PP σ

ML FP

PP FP

ML #

PP #


0.0-0.1

-

-

-

-

-

-

0

0

0.1-0.2

3.67

3.94

3.23

3.31

0.09

0.05

7736

3583

0.2-0.3

3.24

3.48

3.26

3.23

0.16

0.11

17491

14308

0.3-0.4

2.64

2.98

3.23

3.26

0.26

0.17

17000

17600

0.4-0.5

2.51

2.46

3.30

3.11

0.33

0.25

11114

14572

0.5-0.6

2.27

2.27

3.26

3.10

0.40

0.33

8375

9894

0.6-0.7

2.11

2.03

3.14

3.08

0.45

0.41

6921

7771

0.7-0.8

1.83

1.76

3.06

2.98

0.52

0.50

6321

6530

0.8-0.9

1.51

1.44

2.92

2.83

0.62

0.60

7101

6873

0.9-1.0

0.22

0.20

1.22

1.17

0.94

0.94

66910

67838


Similar analysis as Table 1, but with a normally-distributed length with a mean of 30 amino acids, and a standard deviation of 7. In this case, the posterior probability calculation shows slightly superior ability to distinguish between accurate and inaccurate placements than the likelihood weight ratio.

Matsen et al. BMC Bioinformatics 2010 11:538   doi:10.1186/1471-2105-11-538

Open Data