Skip to main content

Table 3 Erroneous segments in which reconstructed MSA is “far-apart” from true MSA

From: Characterization of multiple sequence alignment errors using complete-likelihood score and position-shift map

 

w/ many block-wise steps a

w/ many site-wise steps b

long blocks c

Score category

MAFFT (E-INS-i)

Prank (Best-fit)

MAFFT (E-INS-i)

Prank (Best-fit)

MAFFT (E-INS-i)

Prank (Best-fit)

12 primates

 I

0.6%

1.7%

1.4%

1.8%

0.6%

0.9%

 D

2.0%

6.8%

4.1%

8.6%

1.7%

3.1%

 S

0.1%

0.1%

0.1%

0.2%

0.1%

0.1%

 Overall

0.5%

0.4%

1.1%

0.6%

0.6%

0.4%

15 mammals

 I

18.8%

27.2%

16.8%

22.5%

3.0%

5.7%

 D

36.1%

54.1%

33.6%

49.3%

5.8%

8.0%

 S

4.4%

7.7%

3.2%

5.3%

1.0%

1.6%

 Overall

22.3%

17.6%

20.3%

14.3%

4.7%

4.6%

9 fast-evolving mammals

 I

40.1%

51.5%

36.9%

47.7%

4.1%

7.1%

 D

68.9%

83.4%

66.5%

80.9%

7.1%

7.8%

 S

16.6%

24.1%

13.9%

19.6%

2.7%

2.9%

 Overall

50.7%

43.9%

48.2%

40.0%

6.5%

6.7%

  1. NOTE: Shown in each cell is the percentage of each specified score category (row) that the errors with each specified type of “apart”-ness (column) account for
  2. a The true and reconstructed MSAs are separated by 5 or more block-wise steps
  3. b The true and reconstructed MSAs are separated by 30 or more site-wise steps
  4. c Position-shift blocks that are 30 or more sites long