Skip to main content

Table 2 Correction power/sensitivity (in terms of total corrected bases and corrected read length) and accuracy (in terms of identity and indel rate) of seven methods over M. abscessus dataset (5 Mb genome, 251 bp read length)

From: An efficient error correction algorithm using FM-index

 

Corrected bases (bp)

Corrected length (bp)

Identity (%)

Indel rate (10−5)

Raw data

353.43 Mbp

180.77

98.80%

4.37

Coral

358.65 Mbp

182.72

99.19%

4.59

Karect

387.17 Mbp

196.62

99.42%

2.91

FMOE

492.95 Mbp

249.31

99.85%

1.37

QuorUm

361.02 Mbp

182.55

99.96%

0.56

RACER

381.11 Mbp

192.49

99.89%

1.23

BLESS2

358.02 Mbp

182.35

99.87%

0.68

SGA

366.05 Mbp

182.16

99.46%

2.35

  1. Coral, Karect, and our FMOE are overlap-based correction methods whereas QuorUm, RACER, BLESS2, and SGA belong to k-mer spectrum approach