Running Pipelines in Quark as a Bench Scientist
This document explains how bench scientists can launch and configure pipelines in Quark using the Launchpad, without needing any coding experience. It covers locating a pipeline, reviewing its details, configuring inputs and run settings, validating parameters, and submitting a pipeline run.
Step 1: Open the Launchpad
Navigate to the Pipelines tab in the navigation pane.
Select Launchpad to view all available pipelines.
The Launchpad displays a brief overview of every pipeline available on the platform.
Step 2: Search and Filter for the Desired Pipeline
Use the search field to search across all pipelines, or apply category filters to narrow down the list.
For example, if you are looking for a Boltzgen-related pipeline, select nf-core from the available category filters.
You can also search directly for a pipeline by name using the search field. For instance, search for "Boltzgen" to locate that specific pipeline.
In the About section for the selected pipeline, you will find: - A brief overview of what the pipeline does - The requirements needed to run or load the pipeline - The current pipeline version - Tags associated with the pipeline (e.g., category)
Step 3: Configure Run Settings
In the Run Settings area, provide a descriptive name for your pipeline run so it can be easily identified later in your run history.
When you are ready to proceed, click Next.
Step 4: Provide Pipeline Inputs
On the Inputs page, configure all required scientific inputs for your run:
- PDB file — Select the PDB file you will use to model your peptide binder.
- Chain and sequence — Specify the chain identifier and the amino acid sequence.
- Cyclic peptide — Indicate whether the peptide is cyclic.
- Full chain visibility — Indicate whether you need visibility across the entire polypeptide chain.
Next, specify the protocol and any additional run parameters. Use the dropdown to choose from the available protocol options.
Then configure the following run parameters: - Number of designs — The total number of design iterations to generate. - Fold number sequences — The number of sequences to fold per design. - Batch sizes — The number of designs to process in each batch. Larger batches may reduce overall runtime but require more compute resources.
Step 5: Review Inputs and Submit the Pipeline Run
When you have finished configuring all inputs, click Review to validate your setup before submitting.
If any required inputs are missing — for example, if you have not yet uploaded the required PDB file — an error message will be displayed. Review all input parameters carefully and correct any issues before proceeding.
Once you have confirmed that all parameters are accurate and complete, click Next to submit your pipeline run.
Tip: After submission, your run will appear in the Runs, tab where you can monitor its progress and access results once it completes.