To install q, extract the download file into some folder on your job server and run the configuration script:
$ cd /path/to/some/folder
$ tar -xzf q-<#.#.#>.tar.gz
$ cd q-<#.#.#>
$ perl configure.pl
reading system information ............................
generating q program target
done
created q program target:
/path/to/some/folder/q-<#.#.#>/q
where #.#.# is your q version (e.g. 1.0.2). The above steps are required to use both the command-line and web interfaces. Perform a quick check for appropriate installation by calling q with no commands or options:
$ ./q
q version #.#.#
q is a utility for submitting, monitoring and managing data analysis pipelines
usage: q <command> [options] <masterFile> [...]
masterFile = path to a master instructions file
use 'q --help' or 'q <command> --help' for extended help
In most cases you will want to add ‘/path/to/some/folder/q-<#.#.#>’ to your server system PATH variable so that it is available to you from any server directory by simply typing ‘q’. Details of how to do this will depend on your server and account configuration.
Familiarize yourself with q as follows:
$ q --help
$ q submit --help
$ q submit --dry-run q_example.q
and see also http://tewlab.org/q.
-------------
version 1.0.2
-------------
Added modules capability = q instructions file sets/packages implicitly available to all pipelines
Added modules for file handling, genome handling, and others
Added utilities = software called as a command in a worker script, designed with q workflows in mind
Added file slurping, stats management, and bedutil (similar to bedtools) utilities
Improved file access speed by caching encountered files.
Modified file search directories to include, in order:
master directory
current slave directory
q modules directory
current working directory (where q was called from)
absolute file path
Added support for optional variables in conditionals (exitIF, etc) and ternary operator
Added '#q option $VAR[XX]' directive to support optional variable in worker scripts
Modified environment search directories to include, in order:
<qDirectory>/environment.<type> (the q installation directory)
~/.q/environment.<type> (the user's .q direcory)
<masterDirectory>/../environment.<type> (the master classes directory)
<masterDirectory>/environment.<type> (the master directory)
Bug fixes/minor performance changes:
- If invoke uses a variable already defined in master, master variable resets to prior value after invoke.
- Added a syntax error if preserve attempts to use a variable name in use in an invoke statement.
- Other miscelleneaous tweaks.
-------------
version 1.0.1
-------------
Minor feature additions:
- Added option -u,--no-update to the status command to allow rapid retrieval of jobs in-process.
- Add option -c,--chain to the status command to allow retrieval of job dependency chains.
Bug fixes/minor performance changes:
- Replaced a non-standard dependency in snipStream and snipFile utilities.
- Lowered the job memory requirements on q_example jobs to 100M.
- Lowered the job memory requirements on protect and backup auto jobs to 500M.
- Removed an unnecessary status update called by report, script and environment commands.
- Added improved parsing to reporting of archived status lists.
-------------
version 1.0.0
-------------
The first public release of q. Included support for SGE and PBS job schedulers on a server cluster, using both the command line and q remote web interfaces.
