Jobs
Jobs are the record of server-side analyses in ProteinIQ. They connect inputs, settings, compute status, credits, downloadable outputs, and results.
A job represents one server-side execution of a tool in a workspace. Use this page to understand what a job stores, how statuses work, where results live, and what actions are available after a run starts.
Basic concepts
Job
A job is created when you submit a server-side tool. It stores the tool name, job name, submitter, workspace, status, progress, timestamps, credit usage, and references to stored inputs and outputs.
Browser-side tools can return results without a server-side compute job. Server-side tools create jobs so work can continue asynchronously and be reopened later from history.
Jobs store results and references to generated files so completed runs can be reopened from workspace history.
Workspace scope
Jobs belong to the active workspace. Switching workspaces changes the job history you see, the files available to you, and the credit balance used for new paid jobs.
Members with job read access can view workspace jobs. Owners, admins, and members can create jobs. Viewers can read jobs but cannot submit new ones.
Inputs and settings
Each job records the data and configuration used for the run. Depending on the tool, input can come from pasted text, uploaded files, external fetchers, or multiple molecular inputs.
When you reopen a job, ProteinIQ restores the run context so you can inspect the submitted inputs, settings, outputs, timing, and credit usage.
Job lifecycle
Jobs move through a small set of statuses as they are created, launched, processed, and finalized.
| Status | Meaning |
|---|---|
| Pending | The job was created and is waiting to start |
| Queued | The job is waiting for execution capacity |
| Processing | The tool is actively running |
| Completed | Results are ready |
| Failed | The run ended with an error |
| Timeout | The run exceeded its execution limit |
| Cancelled | The job was stopped before completion |
| Retry | The system is re-running the job after a transient failure |
Most runs move from Pending to Processing to Completed. Queueing, retries, timeouts, and cancellation appear only when the run or infrastructure state requires them.
Important
Paid jobs are charged when the job is created. Eligible jobs that end in Failed or Timeout are refunded automatically.
Viewing jobs
ProteinIQ gives you two main history views: workspace history and per-tool history.
Workspace history
Open My results to see jobs across the active workspace. This is the best view when you remember the project or target but not the exact tool.
The list shows the job name, status, duration, tool, and submitter. Opening a job shows the full run context and available outputs.
Tool history
Tool pages can include a History view for runs created by that tool. Use it when you are comparing repeated runs from the same analysis page.
Clicking a history item loads that job back into the tool view.
Job actions
Job actions are available from job menus in history and result views. Available actions depend on status, permissions, and whether output data exists.
- Rename: Give the job a descriptive name for easier search and review
- Delete: Remove the job and its associated inputs, outputs, and metadata
- Retry: Create a new run from a failed job's inputs and settings
- Duplicate: Start from the same inputs and settings, then adjust before running again
- Share: Change who can view the job
- Export: Download structured output as CSV or JSON when available
- Copy link: Copy a URL that opens the specific job
- Open in new tab: Review the job in a separate browser tab
Use names that include the target, molecule, tool, and purpose, such as 1ABC docking aspirin or BRCA1 AlphaFold2 test.
Warning
Deleting a job removes it from active history. Download important results or save reusable outputs to the Files library before deleting.
Sharing jobs
Jobs are private by default. Sharing changes who can view the job without changing who owns it or who can rerun it.
ProteinIQ supports these visibility levels:
- Private: Only you can view the job
- Public: Anyone with the link can view the job
- Logged-in users: Any ProteinIQ user can view the job
- Invited only: Only specific invited email addresses can view the job
Shared viewers can inspect inputs, settings, and results. They cannot edit or rerun the original job.
Credits and refunds
Each paid job has a credit cost calculated from the tool configuration and submitted input. The same cost model is used before submission and when the backend creates the job.
The workspace must have enough credits before a paid job starts. If the run is eligible for a refund, ProteinIQ records a refund entry and restores credits to the workspace balance.
Workflow jobs can have different refund handling because workflow runs manage credits at the workflow level.
Troubleshooting
- Job stuck in pending: The job is waiting to start. This can happen while waiting for compute capacity.
- Job stuck in queued: The job is waiting for an execution slot. It starts automatically when capacity is available.
- Job failed immediately: Check the error message. Invalid input, incompatible settings, or unsupported file content are common causes.
- Job timed out: The input or settings exceeded the execution limit for that tool. Reduce the input size or split the run.
- Cannot find a job: Confirm you are in the correct workspace, then check /app/results. Per-tool history only shows jobs for that tool.
- Results are missing after completion: Refresh the job. If results still do not appear, contact support with the job URL.