The following FAQ page describes how to distribute a pipeline script via pipeline reports and run a script on any client computer. Distribution is also possible via ex- and import of scripts. Distribution of scripts to multiple computer is needed for parallel processing of data. On computers without a graphical user interface (GUI) no script can be defined and ex- and/or imported.
Distributing a Script to Clients with GUI
The following steps are recommended if a GUI is available
on all computers for parallel processing of the same script.
For Linux computers without display a X11 forwarding to a desktop computer with display is required.
- Step 1: Start a SeqSphere+ Client on any of the computers with GUI.
- Step 2: Press Start Pipeline Mode
- Step 3: Create a named pipeline script to define the workflow.
- Step 4: Start the pipeline for the newly created pipeline script.
- Step 5: Start a SeqSphere+ Client on another computers and login.
- Step 6: Choose in the Tools menu: Browse Pipeline Reports.
- Step 7: Select the named pipeline script that was just started and press Open Report.
- Step 8: Switch to the tab Pipeline Script press Import to this Computer.
- Step 9: Step through the pipeline script definition and be sure to adjust the paths for the input sources (and assembly/read output folders) if the location is different on this computer.
- Step 10: Save changes, logout, start the pipeline mode, and run the newly imported pipeline script on this computer.
Distributing a Script to a Client without GUI
The following steps are needed to run a pipeline on a client without GUI (e.g., Linux) for single or parallel processing of a pipeline script.
- Step 1: Use a desktop computer with a GUI and start the SeqSphere+ Client.
- Step 2: Press the Start Pipeline Mode link.
- Step 3: Press the Create New Script button, define and save the pipeline script that should be run on the client computer without GUI.
- Step 4: Press the Manage Scripts button.
- Step 5: Select the just created pipeline script and press Export Script to export it to a file.
- Step 6: Transfer this file to the computer without GUI. The pipeline script file is an XML file. Open the file in a text editor (e.g, emacs, vi).
- Step 7: Replace the path in the XML-tag //PipelineScript/InputSources/InputSource/SourceURL/ with the directory where the input sequence files (FASTA/ACE/BAM/FASTQ) can be accessed by this computer.
- Step 8: Only required if assembly file storage was enabled:
- Replace the path in the XML-tag //PipelineScript/AssemblyFilesStorage/LocalDirectory/ with an output directory for ACE/BAM files on this computer.
- Step 9: Only required if read file storage was enabled:
- Replace the path in the XML-tag //PipelineScript/ReadFilesStorage/LocalDirectory/ with an output directory where the processed FASTQ files should be stored on this computer.
- Step 10: Store the XML file.
- Step 11: Finally run the pipeline with ./seqsphere64 -pipeline <pipeline-file.xml>
Hint: Internet access is required if public cgMLST schemes are used in the pipeline. If the computer without GUI connects to the Internet via a proxy, the proxy settings can be configured with the command:
./seqsphere64 -configureProxy 2>/dev/null