Trim to a specified length

[sjackman]

I'd like to trim a bunch of variable length reads to a fixed length. I read the help page for seqtk trimfq, but didn't see what I was looking for. Could this feature please be added?

[sjackman]

I found that FASTX-toolkit implements this feature.
http://hannonlab.cshl.edu/fastx_toolkit/commandline.html#fastx_trimmer_usage

   [-f N]       = First base to keep. Default is 1 (=first base).
   [-l N]       = Last base to keep. Default is entire read.

[vsbuffalo]

This would be useful — I'd be happy to implement this and issue a pull request if it'll be incorporated. Note that bioawk can also do this pretty easily:

$ echo ">test-1\\nAGTATCTAGCATATCGT\\n>test-B\\nGCTAGCACGATC" | \
  bioawk -cfastx '{print fastx($name, substr($seq, 0, 10))}' # note this depends on fastx(), a function I implemented
>test-1
AGTATCTAGC
>test-B
GCTAGCACGA

[lh3]

Release 1.0 does not support trimming to fixed length as adding the functionality may affect the stability of trimfq. You may use Vince's script to achieve the goal for now. I will leave the issue open.

[ndaniel]

Actually release 1.0 supports trimming to fixed length!
Just try "seqtk -q 0 -l N input.fq" (it keeps the first N bp)!

[ndaniel]

I have just commited some changes which does all these (i.e. shortcircuiting the q == 0, keep first last N bp).

[lh3]

Implemented via 46cd37e ... after 3 years. Sorry.

[sjackman]

Thanks, Heng!

[sjackman]

Would you please tag a new stable release tagged either 1.1 or 87? The last stable release 1.0 was tagged in 2013. An out-of-date tagged stable release is more confusing than having no tagged stable release at all.