Split fastq to fasta and qual file?

[ewilbanks]

Hi all,

Does anyone have or know about good scripts to split a sanger format fastq file into the corresponding fasta and qual file?? I have a dataset that I'd like to quality trim with LUCY but I can't figure out how to get it split apart! I've tried using the app on the galaxy page -- but its producing weird errors that I don't understand. Any help much appreciated!!

-Lizzy

[maasha]

This can be done with Biopieces (www.biopieces.org):

read_fastq -i test.fq | write_454 -o test.fna -q test.fna.qual -x

Cheers,


Martin

[maubp]

In Biopython the simplest way to do it is like this:

from Bio import SeqIO
SeqIO.convert("example.fastq", "fastq", "example.fasta", "fasta")
SeqIO.convert("example.fastq", "fastq", "example.qual", "qual")

You can be more cunning if you want to avoid making two passes through the FASTQ, but the above should be pretty fast anyway.

See also http://dx.doi.org/10.1093/nar/gkp1137 - I'd have suggested using EMBOSS seqret which can do FASTQ to FASTA, but I don't think it supports the QUAL format.

[ewilbanks]

thank you!! The Biopython script did the trick-- even for a python newbie!

[ewilbanks]

Do you know how to use Biopython to do the reverse? Fasta +qual = fastq?

[maasha]

Well, Biopieces can do that as well:

read_454 -i test.fna -q test.qual | write_fastq -o test.fq -x

In fact, Biopieces can also trim sequences based on quality scores by using trim_seq:

read_454 -i test.fna -q test.qual | trim_seq | write_fastq -o test.fq -x

Martin

[maubp]

Do you know how to use Biopython to do the reverse? Fasta +qual = fastq?

Since you asked, yes, most easily done with the PairedFastaQualIterator function in the Bio.SeqIO.QualityIO module:

from Bio import SeqIO
from Bio.SeqIO.QualityIO import PairedFastaQualIterator
rec_iter = PairedFastaQualIterator(open("Quality/example.fasta"),
                                   open("Quality/example.qual"))
SeqIO.write(rec_iter, "Quality/temp.fastq", "fastq")

This isn't quite as easy as the reverse since we need to take two input files and read over them in sync - and the high level functions in Bio.SeqIO are all intended for just one file. This example is based on the example in the documentation here:

http://www.biopython.org/DIST/docs/api/Bio.SeqIO.QualityIO-module.html#PairedFastaQualIterator

[ewilbanks]

Thanks everyone!

@maasha, I'll have to check it out! Does trim_seq accept sanger format qualities or only Solexa?

[maasha]

trim_seq works on Illumina type qualities.

read_fastq and read_454 convert to Illumina type qualities per default. Phred scores are automagically detected and converted. If you have Solexa scores there is a switch.

write_fastq output Illumina type qualities.

write_454 automagically convertes to decimal scores.



Cheers,


Martin