Skip to main content

Table 1 Change in average Q-score according to the number of sequences being aligned

From: DECIPHER: harnessing local sequence context to improve protein multiple sequence alignment

 

Total sequences

Clustal Omega

DECIPHER

MAFFT

MUSCLE

PASTAa

HOMSTRAD-modb

Maximum score

0.898

0.910

0.920

0.881

0.921

Change from max

2

−0.013

−0.02

−0.045

0.000

N/A

125

0.000

0.000

0.000

−0.007

0.000

250

−0.009

−0.011

−0.028

−0.028

−0.003

500

−0.013

−0.014

−0.037

−0.028

−0.016

1000

−0.035

−0.011

−0.049

−0.048

−0.026

2000

−0.035

−0.019

−0.054

−0.088

−0.028

4000

−0.069

−0.030

−0.077

−0.128

−0.049

PREFAB-modb

Maximum score

0.875

0.908

0.923

0.855

0.900

Change from max

2

−0.067

−0.085

−0.123

−0.047

N/A

125

0.000

0.000

0.000

0.000

0.000

250

−0.022

−0.007

−0.057

−0.011

−0.015

500

−0.022

−0.004

−0.068

−0.026

−0.029

1000

−0.034

−0.023

−0.090

−0.056

−0.045

2000

−0.069

−0.037

−0.107

−0.093

−0.080

4000

−0.121

−0.026

−0.098

−0.156

−0.116

  1. aScores for aligning two sequences are listed as “N/A” because PASTA cannot perform pairwise alignment
  2. bResults for the subset of reference alignments with at least 4,000 reference sequences are shown (297 alignments for HOMSTRAD-mod and 201 alignments for PREFAB-mod)