Skip to main content

Table 4 Frequencies of errors of different types in MSAs among 12 primates

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

Error type

MAFFT (E-INS-1)

(Percent)

MAFFT (E-INS-i)

(Percent)

Prank (Best-fit)

(Percent)

Shift

113,658

(79.00%)

110,890

(80.01%)

120,802

(89.78%)

Merge

3779

(2.63%)

3920

(2.83%)

1875

(1.39%)

Purge

2175

(1.51%)

2198

(1.59%)

742

(0.55%)

Split

6

(0.004%)

1

(0.001%)

383

(0.29%)

Ex-nihilo

0

(0%)

4

(0.003%)

34

(0.03%)

v-Merge a

926

(0.64%)

765

(0.55%)

489

(0.36%)

v-Split b

51

(0.04%)

28

(0.02%)

142

(0.11%)

CII c

1146

(0.80%)

1116

(0.81%)

110

(0.08%)

iCII d

2191

(1.52%)

2335

(1.69%)

40

(0.03%)

Others e

943

(0.66%)

900

(0.65%)

728

(0.54%)

Mixture f

7814

(5.43%)

7304

(5.27%)

3774

(2.80%)

(Paired) g

(5375)

(3.74%)

(5047)

(3.64%)

(2738)

(2.04%)

Complex h

11,183

(7.77%)

9127

(6.59%)

5428

(4.03%)

Total

143,872

(100%)

138,588

(100%)

134,547

(100%)

  1. NOTE: Shown in each cell is the number or the percentage of erroneous segments via a specified aligner (column) belonging to a specified error type (row). In each of the top 10 rows, the specified error type alone can explain each erroneous segment
  2. aVertical-merge
  3. bVertical-split
  4. cCreation of independent insertions
  5. dIncomplete creation of independent insertions
  6. eEach segment is explained solely by a non-complex error type other than the 9 types above
  7. fEach segment is explained by a mixture of two or more non-complex error types
  8. gEach segment is explained solely by non-complex errors associated intrinsically with a pair, or pairs, of blocks. This is included in the “Mixture” category
  9. hEach segment includes at least one complex error
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us