Skip to main content

Table 4 Number of incorrectly predicted speciesa for different abundance thresholdsb without clade exclusion. Fewer incorrectly predicted species are predicted with the in silico data that does not contain errors versus the in vitro data containing sequencing errors (Table 3)

From: Evaluation of shotgun metagenomics sequence classification methods using in silico and in vitro simulated communities

 

No cutoffb

Cutoff > 0.01 %b

Cutoff > 0.1 %b

Cutoff > 1 %b

Method

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

CARMA3

11

41

11

3

11

1

11

1

CLARK

11

0

11

0

11

0

11

0

DiScRIBinATE RAPSearch2c

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Kraken

11

0

11

0

11

0

11

0

Filtered Kraken

11

0

11

0

11

0

11

0

MEGAN4 BLASTN

11

0

11

0

11

0

10

0

MEGAN4 RAPSearch2

11

92

11

29

11

1

10

0

MetaBin

11

286

11

41

11

3

11

0

MetaCV

11

0

11

0

11

0

11

0

MetaPhyler

10

12

10

12

10

8

7

3

PhymmBLc

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

RITA

11

0

11

0

11

0

11

0

TACOAc

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

MG-RAST best hit

10

646

10

136

10

26

10

6

MG-RAST LCA

10

300

10

54

10

8

9

3

  1. aUsing the FW in silico dataset of sequenced reads from 11 species
  2. bA cutoff of > × %, for example 0.01 %, would indicate that only species with a predicted abundance of at least × % of the total set of predictions were considered
  3. cThese methods do not predict to the species level at this read length (they require longer read lengths). See additional analyses at other levels of clade exclusion
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us