Cutadapt Or Fastx Clipper

Question

[Nicolas Rosewick]

Hi,

Which one is the best adapter remover for small reads (36 bp and 72 bp long) ? Is there a study to compare adapter remover algorithms ?

cutadapt : http://code.google.com/p/cutadapt/ fastx clipper : http://hannonlab.cshl.edu/fastx_toolkit/index.html

Thanks,

N.

[Martin A Hansen]

Also consider find_adaptor: http://code.google.com/p/biopieces/wiki/find_adaptor

[brentp]

and consider: https://github.com/ucdavis-bioinformatics/scythe

[Steve Lianoglou]

@maasha: Thanks, I haven't stumbled on biopieces before ... looks interesting!

Answer 1

[Jeremy Leipzig]

cutadapt has more parameters

fastx is not pair-safe - it will discard sequences that are "all-adapter" which screws up the pairing

the fastx clipper is much more aggressive than cutadapt or scythe using default parameters, for example:

$ cat sample.fq 
@HWI-ST431_52:1:1:1259:1981/1
GTTCAGAGTTCTACAGTCCGACAATCGTATGCCGTNTTCTGCTTGAAAAAANNNANAANNNNANNTNNNNNGNGNNNNNNNNNNNNNNNNNNNNNNNNNNN
+
b`ZUYZKYUSV[[_[cad\\W\[X`ddadc^``^ZB][]\Tc\^`ccbdbBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB

fastx:

$ fastx_clipper -a ATCTCGTATGCCGTCTTCTGCTTG -i sample.fq 
@HWI-ST431_52:1:1:1259:1981/1
GTTCAGAGTTCTACAGTCCGA
+
b`ZUYZKYUSV[[_[cad\\W

cutadapt:

$ cutadapt -a ATCTCGTATGCCGTCTTCTGCTTG sample.fq 
@HWI-ST431_52:1:1:1259:1981/1
GTTCAGAGTTCTACAGTCCGACAATCGTATGCCGTNTTCTGCTTGAAAAAANNNANAANNNNANNTNNNNNGNGNNNNNNNNNNNNNNNNNNNNNNNNNNN
+
b`ZUYZKYUSV[[_[cad\\W\[X`ddadc^``^ZB][]\Tc\^`ccbdbBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB

scythe:

$ scythe -a adapter.fa sample.fq 
@HWI-ST431_52:1:1:1259:1981/1
GTTCAGAGTTCTACAGTCCGACAATCGTATGCCGTNTTCTGCTTGAAAAAANNNANAANNNNANNTNNNNNGNGNNNNNNNNNNNNNNNNNNNNNNNNNNN
+HWI-ST431_52:1:1:1259:1981/1
b`ZUYZKYUSV[[_[cad\\W\[X`ddadc^``^ZB][]\Tc\^`ccbdbBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB

trimmomatic

$trimmomatic SE sample.fq sample.out ILLUMINACLIP:adapter.fa:2:30:10
TrimmomaticSE: Started with arguments:
 sample.fq sample.out ILLUMINACLIP:adapter.fa:2:30:10
Automatically using 2 threads
Using Long Clipping Sequence: 'ATCTCGTATGCCGTCTTCTGCTTG'
ILLUMINACLIP: Using 0 prefix pairs, 1 forward/reverse sequences, 0 forward only sequences, 0 reverse only sequences
Quality encoding detected as phred64
Input Reads: 1 Surviving: 1 (100.00%) Dropped: 0 (0.00%)
TrimmomaticSE: Completed successfully
$ more sample.out
@HWI-ST431_52:1:1:1259:1981/1
GTTCAGAGTTCTACAGTCCGA
+
b`ZUYZKYUSV[[_[cad\\W

fastp

$ fastp -i sample.fq -n 50 -a ATCTCGTATGCCGTCTTCTGCTTG --stdout
Streaming uncompressed output to STDOUT...


@HWI-ST431_52:1:1:1259:1981/1
GTTCAGAGTTCTACAGTCCGACAATCGTATGCCGTNTTCTGCTTGAAAAAANNNANAANNNNANNTNNNNNGNGNNNNNNNNNNNNNNNNNNNNNNNNNNN
+
bZUYZKYUSV[[_[cad\\W\[Xddadc^`^ZB][]\Tc\^ccbdbBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB

[Jeremy Leipzig]

added recent entries trimmomatic (2014) and fastp (2018)