About

In this tutorial we present the WS-PGRADE Job Wizard, through an example molecular docking application.

The application used is AutoDock Vina, and works as follows: for performing the molecular docking, AutoDock Vina expects a receptor molecula (receptor.pdbqt), a docking simulation configuration file (config.txt), and a set of ligands (ligands.tgz) to dock against the receptor molecule. The result of the docking is an output archive containing the docking results of the different ligands (output.tar).

The executable and the input files can be downloaded:

The example in this tutorial is running a parameter sweep experiment: as we will define 6 parametric input files, 6 job instances of the AutoDock Vina application will be executed in parallel on the LToS platform. Each job will receive the executable (vina_package.sh) and the static input files (config.txt and receptor.pdbqt), and each instance will receive exactly one of the parametric input files as ligands.tgz: instance one will receive the file 1.tgz as ligands.tgz, instance two will receivce 2.tgz as ligands.tgz, and so on.

Each of the parametric inputs contains multiple ligands - the AutoDock Vina application will perform the molecular docking of these against the receptor molecule. In our example, each parametric input contains 10 ligands, thus a total of 60 molecular dockings will be performed during the experiment.

Step 1: Upload the executable

As first step, open the Job Wizard, and upload you executable, as shown in the figure:

You have multiple possibilites to upload files:

  • You can click in the "Drop files here!" field, and select the files from the file browser dialog.
  • You can drag and drop files from your desktop/explorer window/finder window/etc. into the "Drop files here!" area.

Please note that in the executable upload step you can upload one single file:

Would you want to replace the uploaded executable, you can click the "Remove" button.

If you have finished uploading your executable, click the "Next" button to go to the next step:

Step 2: Define static inputs for your executable

You can upload the static inputs for you executable in a similar way to the executable definition method: either click in the gray "Drop files here!" field, or drag and drop files into that area. During this step you have the possibility to define multiple static input files for your executable - all of them will be transferred to each instance of your job.

Note: the name of the uploaded files cannot be overriden - please make sure you upload files using the name as your application expects them! 

The following figure shows the two static input files uploaded for the AutoDock Vina application:

Once you have defined all the static input files for you application, click the "Next" button to go the next step:

Step 3: Define parametric input files

As already discussed, if you define at least two parametric input files, then multiple instances of your application will be executed on the LToS platform: each instance will receive the static input files and one of the defined parametric input files. Exactly one instance will receive an element of your parametric input set, and every element will be assigned to one instance.

When defining parametric input files, you have to specify two things:

  • the filename using which your application will open these files,
  • the set of files to be used as parametric inputs. 

The following figure shows the parametric files defined for the AutoDock Vina application: the 6 example files have been uploaded (you can click into the gray area to open the file browser dialog or you can drag and drop files), and ligands.tgz has been set as the target filename.

Once you have uploaded all the parametric input files for you application, click the "Next" button to go the next step:

Step 4: Define command line arguments

If your application requires any command line arguments, you can define them in this step. Every instance of your application will receive exactly the same set of arguments, thus it is not possible to parameterize this property.

The AutoDock Vina application doesn't use any command line arguments. If you're ready with defining the arguments for your application, click the "Next" button to go to the next step:

Step 5 - Define execution resource

You need to select where to execute your application, and on what type of resource. The Job Wizard currently offers three different instance types:

  • Small: at least 1 vCPU and 2 GB RAM,
  • Medium: at least 2 vCPUs and 4 GB RAM,
  • Large: at least 4 vCPUs and 8 GB RAM.

If you're ready with resource selection, click the "Next" button to go to the next step:

Step 6 - Define output files

Finally, you have to define the name of the output files your application will produce:

Click on the "Add new!" button to add a new filed for an additional input file. If you think a defined output file is not necessary, just click the "Remove" link in the line of the file.

Finally, if all necessary properties of your application have been defined, click the "Start execution!" button!

Once the experiment is properly submitted, you will see a new entry in the right part of the Job Wizard:

Managing experiments

In the right part of the Job Wizard view, you can see the list of your experiments: name, start time, progress and status is presented.

After selecting an experiment with the radio button, you have the possibility to perform multiple operations with the experiment:

Select the experiment, and click on the triangle next to the "Download outputs" button to get the list of operations:

  • Download outputs: clicking this will download the outputs of your experiment,
  • Download all the workflow: clicking this will download the whole WS-PGRADE/gUSE workflow belonging to your experiment. This contains internal log files as well, so might help if debugging the reasons of an application failure,
  • Abort experiment: can be used to abort a running experiment,
  • Remove experiment: if you have an experiment in the list you're not interested in anymore, you can remove it. This will also remove all the files (both input and output) used and produced by the experiment,
  • Remove All experiments: all of the experiments will be removed.