ngsutils / xsqutils

Utilities for converting SOLiD XSQ files

xsqutils

Converts XSQ format files to FASTQ format files.

Requires the pytables library and HDF5-devel libraries to be installed. (Note: pytables also requires numpy, numexpr, and cython)

HDF5 libraries can be downloaded from http://www.hdfgroup.org/HDF5/. They can also be found in the EPEL yum repository. Pytables requires HDF5 1.6.10 or better.

The installation script install.sh will take care of setting up a virtualenv environment and installing Pytables and it's dependencies. The virtualenv will be setup in the 'env' directory. The default xsq driver script expects this setup and will automatically setup the virtualenv accordingly.

When converting a sample or entire file, multiple processors can be used to speed the conversion.

Usage

xsq cmd {opts} filename.xsq

Commands:
    info      - Lists all of the data associated with the XSQ file
    list      - Lists the samples and tags (R3/F3/etc) present in the file
        Options:
          -c           Show the number of reads present for each tag
          -min {val}   Hide samples that have less than {val} reads

    convert   - Converts XSQ samples and fragments to FASTQ format
        Options:
          -a           Convert all samples (saves to sample_name.fastq.gz)
              [-a additional options]
              -desc          Use descriptions for the sample name
              -f             Overwrite existing files
              -min {val}     Skip samples that have less than {val} reads
              -noz           Don't compress the output FASTQ files with gzip
              -fsuf {val}    Add suffix to file name
              -unclassified  Export "Unclassified" library (usually skipped)

          -n name        Convert only sample "name" (writes to stdout)
                         (can be only one, written uncompressed)
          -procs {val}   Use {val} number of threads (CPUs) to convert one
                         region at a time. (default 1)
          -s suffix      Append a suffix to all read names
          -t tag         Convert only this tag (can be more than one)
                         If more than one tag is given, the sequences for
                         each read will be written out together.
                           For example:
                           @read1 F3
                           F3 seq
                           +
                           F3 qual
                           @read1 R5
                           R5 seq
                           +
                           R5 qual
                           @read2
                           ...

        The default is to convert all samples and all fragments/tags.