Skip to main content

Table 1 Time-consuming from five software on human HEK293 samples in silico

From: Systematic and benchmarking studies of pipelines for mammal WGBS data in the novel NGS platform

Software/version

Character of building index

Indexing strategy

Alignment softwarea

Strategy

Final report?

Sample data (pair-end reads/Mb)

Time of building reference index

Alignment/h

Deduplication

Extraction of paired_read

Extraction of methylation

Total time

BSMAP v2.90

Building reference index in process of alignment

Hash tables

SOAP

Wild-card

No

105.675

p = 0.0013

0

10.478

p = 0.0014

24.5 (m)

p = 0.0007

0

33 (m)

p = 0.0001

11h26m

Bismark v0.22.3

Building multi-reference index at a time

FM index

bowtie2

3-letter

Yes

1h51m

7.025

p = 0.0015

35 (m)

p = 0.0019

100 (m)

p = 0.0010

7.15 (h)

p = 0.0008

18h26m

BatMeth2

Building one reference index at a time

FM index

BatMeth2

3-letter

Yes

2h42m

13.07

p = 0.0017

113 (m)

p = 0.0109

17h39m

BS Seeker2 v2.1.0

Building one reference index at a time

FM index

bowtie2

3-letter

No

3h17m

111.08

p = 0.0011

36 (m)

p = 0.0003

0

30.85 (h)

p = 5.236e-05

145h44m

BSBolt v1.4.8

Building one reference index at a time

FM index

BWA-MEM

3-letter with forked version of BWA-MEM and HTSLIB

No

2h18m

7.775

p = 0.0059

47 (m)

p = 0.0025

0

7 (m)

p = 0.0022

10h58m

  1. aShort-read alignment pipeline used in this research for mapping bisulfite sequencing and extracting methylation pattern
  2. bThe parameters used in process was in default. The configuration of our operating system: Ubuntu v20.04 operating system, containing 2 physical CPUs, 48 logical CPUs, and 377G RAM size