This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How to remove empty read from fastq

Hello,

I would like to remove empty reads from fastq_file after trimming adapter sequencing. This fastq is from 454 GS-FLX.

I tried to remove that using following fastx_clipper(in fastx_toolkit)

fastx_clipper -Q33 -l 1 -i in.fastq -o out.fastq

But I received following error message:

Segmentation fault (core dumped)

Anybody has solution about this problem? Other software can remove these empty reads?

Thanks.

software-error
• 8,950 views • link
updated by Ram Ram • written by cfarmeri cfarmeri

In the latest documentation, I can't find a Q flag for the fastx_clipper command. Maybe try removing that flag?

• 2 views • link
written by James Ashmore James Ashmore

1 answer

If you just want to get rid of short sequences, you can use biowak

bioawk -cfastx 'length($seq) > 1 {print "@"$name"\n"$seq"\n+\n"$qual}'

Edit: FIXED based on comment below!

• 0 views • link
updated by Ram Ram • written by arnstrm arnstrm

Thanks, it works well. The empty reads are removed.

But the head @ characters of the read name line (line1) at each read were also removed.

So I couldn't FastQC these processed fastq...

• 0 views • link
updated by Ram Ram • written by cfarmeri cfarmeri

Oh yeah, I forgot. The command should include printing @ before the name:

bioawk -cfastx 'length($seq) > 1 {print "@"$name"\n"$seq"\n+\n"$qual}'
• 0 views • link
updated by Ram Ram • written by arnstrm arnstrm

Thank you so much!!

I can get processed fastq file trimmed correctly.

• 0 views • link
updated by Ram Ram • written by cfarmeri cfarmeri

I think we can add $comment if there are comments in your fastq files for specifying read number etc:

bioawk -cfastx 'length($seq) > 1 {print "@"$name" "$comment"\n"$seq"\n+\n"$qual}'
• 0 views • link
updated by Ram Ram • written by Dimas • Dimas

Hello,

I too want to remove the empty reads after the adapter trimming. Could you please elaborate your code

bioawk -cfastx 'length($seq) > 1 {print "@"$name"\n"$seq"\n+\n"$qual}'

I mean what part does what and where is the input file ?

• 2 views • link
updated by Ram Ram • written by saamar.rajput saamar.rajput

Hey,

bioawk works like a typical awk command but has been modified to understand some of the common ngs file formats (fasta, fastq, gff etc), hence we the -c flag (to consider the format as fastq). Like awk, it needs awk 'condition{action}' filename.

The condition here is length($seq) > 1 which means length of the sequence is greater than 1

The action here is {print "@"$name"\n"$seq"\n+\n"$qual}' which is to print the sequence back in fastq format (if the condition is satisfied).

You supply the filename after you close the single quote as shown above.

PS: you should not ask a question in a existing thread. This should have been simply a followup comment in the above answer.

• 1 views • link
updated by Ram Ram • written by arnstrm arnstrm

Log in to answer this question.

More posts like this