Trimming Adapters For Paired-End Sequences

Question

[deepthithomaskannan]

Hi all,

I got illumina paired end fastq files. They told me to trim read 2 at the beginning for ~20 to 30 bp due to the WGA adapters.

Can we find the adapters by looking in to the quality? Which tool is good for trimming adapters by keeping the paired nature of the sequences? Is there any issues will come if I use bowtie2 in the downstream for aligning the trimmed sequences(trimming only one in pair) with ref?

Following is the example of read2:

@HWI-ST1162:139:C0H7WACXX:5:1101:1865:1112 2:N:0:CGATGTA GTCATGGTGTCTCTTCACAACAATGGAAACCCTAACTAAGACAAAGACTAATAGAAGTGTTTTTTTAGGAA

+

<9;>;>?1=>;=9=?########################################################

Thanks, Deepthi

[Jeremy Leipzig]

Cutadapt Or Fastx Clipper

Answer 1

[samsara]

I had encountered problem with Cutadapt and nowadays using trimgalore. Cutadapt did not remove both read mates based on quality after trimming in my case; I dont know why. Then later I had problem in downstream analysis. I did not encounter this problem in Trimgalore as it removes both mates based on quality.

READ1=r1.fastq
READ2=r2.fastq
$TRIMGALOREPATH --paired --fastqc_args "--outdir fastQC/outputDir/" -a AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC -a2 AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT $READ1 $READ2

Here, adapter for forward read is - AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC adapter for reverse read is - AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT

[nikhilvgbt]

Hi samsara;

first of all I would say thanks for your comment...

I am new in RNA-seq ... I have data Illumina hi-seq 2000, I used the fastqc to check quality so I found that there are overrepresented sequence in it.

read1:-GATCGGAAGAGCACACGTCTGAACTCCAGTCACTGACCAATCTCGTATGC
read2:-GATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCG

so I tried to trim those sequnces based on your comment above..

I want to know whether its right which I did OR ˆ should remove just 14bp of sequence taken as default in trimgalore software??

Answer 2

[Ashutosh Pandey]

This is sth that has been asked many times here and on other forums. I think you should just search smartly and u shud be able to find the solution. BTW, here is a page (http://intron.ccam.uchc.edu/groups/tgcore/wiki/013c0/Solexa_Library_Primer_Sequences.html) that lists standard adapters used by Illumina sequencing machine. I am sure there are tools like FastQC that can tell you about the overrepresented sequences in your reads. That may help too. For adapter trimming I use ea-utils fastqc tool that is much better than most of the adaptor trimming tool. I am sure there are some that are better than ea-utils but I prefer using ea-utils.

[deepthithomaskannan]

Thank you ashutoshmits. I think this is a good solution.

Answer 3

[Hongshan]

you may use skewer which is dedicated to trimming adapters for paired-end reads

Citation: BMC Bioinformatics.2014, 15:182

Answer 4

[Jordan]

If you know the adapter sequence, you can use a tool called cutadapt to trim the adapter sequences in high-throughput data. But I'm not sure how bowtie2 works.

cutadapt -a ATCGACT file.fastq > output.fastq

Here, ATCGACT is the adapter sequence.

The documentation can be found here: cutadapt

[deepthithomaskannan]

I have no idea about adapter sequence. Then what are my options?

[Jordan]

If you do not know the adapter sequence, I would go with the solution given by ashutoshmits as well.

Answer 5

[Jiacheng]

I'd recommend atria to determine and trim the adapter sequences. It is a newly-published cutting-edge trimmer with exceptional precision and speed. And if you do not know what adapter sequence should be used, Atria can detect adapters if adapter content is higher than 0.04%. (If <0.04%, no need to do adapter trimming.)

Eg: Finding adapters for R1 and R2

atria --detect-adapter -r read1.fastq -R read2.fastq

Adapter trimming and hard-clip 5' ends

atria -r read1.fastq -R read2.fastq -a ADAPTER_READ1 -A ADAPTER_READ2 --clip5 INT