ahcm / qseq2fastq Goto Github PK

* ABOUT qseq2fastq

qseq2fastq converst Illumina Deep Sequencing Qseq files to more commonly
used FastQ format. This is mostly a transformation of selected columns
of the Qseq file. The Phred score is also transformed. Without -a option
reads the sequencer marks a bad are filtered out. Usually I would recommend
using -a and handling the quality later in the pipeline.
It is implemented in C for speed.

Written and designed by Andreas Hauser <[email protected]>, <[email protected]>.

It was initially based on qseq2fastq.pl: http://sourceforge.net/projects/qseq2fastq/


* INSTALLATION


** DOWNLOAD

http://www.splashground.de/software/qseq2fastq-andy/qseq2fastq-andy-3.0.tar.gz


** COMPILE

$ make

Note Linux x86_64 (Scientific Linux 6) is already containedprecompiled.


** INSTALL

$ make install INSTALL_PREFIX=/usr/local

where /usr/local is the path where the binary will be installed in the subfolder bin,
e.g. /usr/local/bin in this case.


* BUGS

Please do report them to Andreas Hauser <[email protected]>
or github: https://github.com/ahcm/qseq2fastq/issues


* USAGE

$ qseq2fastq -a s_1_1_*_qseq.txt > s_1_1.fastq

But in addition it can now do multiplexing. As you can see this much more complex.
E.g.
$ qseq2fastq -a -b ACTTGAA,GCCAATA,TGACCAA,ACAGTGA,CAGATCA -o out/s_1_1_0010_ -d s_1_2_0010_qseq.txt
s_1_1_0010_qseq.txt > out/s_1_1_0010_MISMATCH.fastq

Where

 -b gives the barcodes (the ones in the barcode lane, so be careful about
    inversion from your adapter sequences etc.)

 -d gives the qseq file with the barcode reads

 -o gives the prefix for the output files (one for each barcode).
    If the output files exist output will be appended.
    The prefix can like above contain a directory but that directory must exist.

The reads without a matching barcode are written to standard out so they can either
be discarded (> /dev/null) or like above be written to file (out/s_1_1_0010_MISMATCH.fastq).

It's easy to convert a tile like this as each tile has its own file.
To convert a whole lane one can either concatenate the files:

$ cat s_1_1_*_qseq.txt > s_1_1.qseq

$ cat s_1_2_*_qseq.txt > s_1_2.qseq

Since the star is usually expanded in alphabetical order (by your shell (e.g. zsh)
so be careful if you don't use such a shell) this will preserve the order and the
lines in the both files will correspond.

Another way is to use a loop and the appending feature of qseq2fastq.
$ for i in `seq -w 1 120`; do  qseq2fastq -a -b ACTTGAA,GCCAATA,TGACCAA,ACAGTGA,CAGATCA -o
out/s_1_1_ -d s_1_2_0${i}_qseq.txt  s_1_1_0${i}_qseq.txt >>  out/s_1_1_MISMATCH.fastq; done