Understanding the Velvet configuration

The Velvet configuration file is explained here as a specific example of all external application configuration files.

Going back to figure 15.25, there is a text field at the top. This is where the command expression is created, in this case:

    /opt/local/velvet/clcbio/velvet.sh {hash size} {read type}
            {reads} {expected coverage} {contigs}
The first part is the path to the script. The following parts are parameters that are interpreted by the server when calling the script. Parameters to be interpreted are surrounded by curly brackets { }. Note that each parameter entered in curly brackets gets an entry in the panel below the command line expression.

The first parameter, hash size, can be entered as a Double (which is a number that can take decimal values in computer parlance). The user provides a value when they launch Velvet. A default value is provided in the configuration (31).

The second parameter is read type, which has been configured as a CSV enum which means a list of possible values that the user can choose from. The first part of the configuration information consists of the parameters to be used when calling the velvet script (-short, -shortPaired, -long, -longPaired), and the second part is the more human-readable representation that is to be shown in the Workbench (Short, Short Paired, Long, Long Paired).

The third parameter is reads which is used for the input data. When the User-selected input data option is chosen, a list of all the available export formats is presented. In this case, Velvet expects a fasta file. When a user starts Velvet from the Workbench, the server starts exporting the selected input data from clc format to a temporary fasta file before running the script.

The expected coverage parameter is similar to hash size.

The last parameter, contigs, represents the output file, as indicated by the choice of "Output file from CL" as the type. Here, you specify how the output from velvet should be handled. A list of standard import formats is provided, as well as the option not to import using those tools. Choosing not to import using those tools means that you can choose a high throughput importer instead from the section High-throughput sequencing import/ Post-processing section.

For this example, Do not import is the action set for the contigs parameter. Then, below, in the High-throughput sequencing import/ Post-processing section, the Fasta High-throughput Sequencing Import tool has been selected. Thus, when the results from velvet are ready, they are imported into the CLC Server using that tool and saved where the user indicates when they run the job.