Skip to main content

Table 4 The performances of tested scaffolders for A. thaliana, S. pennellii, and H. sapiens using Nanopore long reads

From: LRScaf: improving draft genomes using long noisy reads

Species

Method

Sum

NG50

NGA50

Longest

Mis. (#)

BUSCO (Complete)

CPU Time (Hours)

Peak Memory (GB)

A. thaliana (KBS-Mac-74)

DBG2OLC (SA)

150.4 Mbp

2.6 Mbp

–

12.2 Mbp

–

25.9%

2.9

18.9

MaSuRCA-Hy (MSR)

123.2 Mbp

2.5 Mb

2.3 Mbp

9.2 Mbp

211

98.3%

145.6 (17.9)

60.3

OPERA-LG (SA)

116.4 Mbp

7.3 Mbp

2.8 Mbp

14.3 Mbp

102

97.2%

188.9

45.9

SMIS (SOAP)

116.2 Mbp

10.3 Mbp

2.4 Mbp

15.0 Mbp

180

98.2%

2391.7

26.8

SSPACE-LR (PLA)

120.6 Mbp

3.2 Mbp

2.2 Mbp

6.8 Mbp

178

98.4%

104.6

9.0

LRScaf (DIS)

123.1 Mbp

9.0 Mbp

4.9 Mbp

12.4 Mbp

115

98.3%

0.5

21.4

S. pennellii

DBG2OLC (SA)

1.5 Gbp

243.7 kbp

–

1.7 Mbp

–

18.9%

27.0

21.6

MaSuRCA-Hy (MSR)

950.4 Mbp

331.1 kbp

159.6 kbp

3.1 Mbp

17,957

95.6%

1389.1 (202.8)

239.7

OPERA-LG (DIS)

952.0 Mbp

730.0 kbp

280.1 kbp

3.5 Mbp

11,404

92.1%

286.6

25.4

SSPACE-LR (DIS)

871.1 Mbp

82.9 kbp

69.5 kbp

1.3 Mbp

6796

94.7%

650.0

13.1

LRScaf (DIS)

952.8 Mbp

794.4 kbp

441.3 kbp

5.4 Mbp

5150

95.8%

3.1

36.8

H. sapiens (NA12878)

LRScaf

2.9 Gbp

17.4 Mbp

12.9 Mbp

64.2 Mbp

785

94.9%

2.1

62.6

  1. Note: The best genomic assembly metrics are highlighted in Bold; Mis. the number of misassemblies, SA SparseAssembler, MSR MaSuRCA de novo, SOAP SOAPdenovo2, DIS DISCOVAR de novo, PLA Platanus, SSPACE-LR SSPACE-LongRead, MaSuRCA-Hy MaSuRCA hybrid pipeline; ‘-’: not available; We report two CPU times for MaSuRCA-Hy (the first value is the time for hybrid pipeline and the second value enclosed in the parenthesis is the time for de novo pipeline); SMIS is excluded on S. pennellii because the run time exceeds the one-month time limit. SIMS and SSPACE-LongRead are excluded on H. sapiens (NA12878) because the run time exceeds the one-month time limit. OPERA-LG is excluded from H. sapiens (NA12878) because of the lack of NGS data