ProteinIQ
DocumentationAPI referenceChangelog
Talk to usGet started

Troubleshooting analyses

Use the job state and reported error to choose the next action.

An analysis can stop before submission, while waiting to run, during execution, or when displaying its output. Identify that stage first so you can address the cause without creating duplicate jobs.

Submission does not start

Check the message beside the input or submit control before changing settings:

  • Invalid input: Check the named field, file, or sequence against the tool's requirements. The input preparation guide explains common format problems.
  • Insufficient credits: Check the active workspace's balance and the displayed job credit requirement. A personal balance in another workspace does not fund this submission.
  • Usage or concurrency limit: Check the limit shown by the app. An available credit balance does not override plan limits or the number of jobs you may run at once.
  • Read-only access: A viewer cannot submit analyses. Ask a workspace owner or admin to check your role.
  • Sign-in required: Sign in, confirm the workspace, and review the input again before submitting.

If a click appears to do nothing after a network interruption, check Results and the tool's History before submitting again. The original request may already have created a job.

A job stays pending or queued

Pending and Queued describe a job that has not started scientific execution. Waiting time is separate from the runtime reported after processing begins.

Keep the existing job URL and check its status again. Closing the browser does not cancel a hosted job. Repeated submissions create additional work rather than moving the first job forward.

If the wait seems abnormal, include the job URL and submission time in a support ticket. A queue status alone does not establish the cause or provide a reliable completion time.

A job fails or times out

Open Details in the output panel and read the reported error. Inspect any available logs for the input, record, or setting involved.

  • Input-related failure: Correct the reported issue before starting another run. Do not repeatedly submit the same failing input without a reason to expect a different outcome.
  • Execution failure: Preserve the error and job URL. If the message does not identify an actionable correction, ask support to investigate.
  • Timed out: Review the tool's limits. Reduce or split the workload only when that preserves the scientific analysis you intend to perform.

Refund behavior depends on the job's final billing outcome. Check the recorded credit usage rather than inferring a charge from how long the page was open.

A job returns partial results

Partial results means the run stopped after reaching a spending limit, exhausting available workspace credits, or reaching its runtime cap. It does not guarantee that every requested output, or any usable output, was produced.

Read the stop reason and inspect the available files. Check which records or stages completed before deciding whether the output answers your question. Starting another run can incur a new charge; increasing a limit does not retroactively finish a stopped job.

For a paused workflow, use the action shown for that run. A review checkpoint, a credit pause, and a failed step require different responses.

Results or downloads are missing

Confirm the job has reached a state with available output, then refresh its page. Check the file list as well as the selected table, plot, or structure view; outputs differ by tool.

  • Access denied: Sign in with the expected account and check the job's sharing permissions. Shared output access does not grant access to submitted input or error files.
  • Download no longer works: Return to the original job page and request the download again. If it still fails, record which file and control failed.
  • Saved output disappeared: Check whether its source job was deleted. Saving a job output to Files keeps a reference, not an independent backup.
  • Job missing from Results: Check the active workspace and clear filters before searching by job name.

For scripted clients, the API error reference describes response codes and retry handling.

Contact support

Open Support, choose New ticket, and provide a Subject, Category, and Message. Select Create ticket to open the conversation and track replies. If you cannot sign in, use the contact form.

Include enough context to identify and reproduce the problem:

  • Job and workspace: The job URL or ID, tool name, and workspace name.
  • Timing: When the problem occurred, including your time zone.
  • Expected and actual behavior: What you tried and what happened instead.
  • Evidence: The exact error, relevant log excerpt, and the affected filename or input field.
  • Changes already tried: Any corrected inputs, changed settings, or repeated submissions.

Keep the original job available while the issue is investigated. Never include passwords, API keys, or payment credentials in a ticket.

PreviousWorkflowsNextOverview

Table of contents

Get started
WelcomeStart guideConceptsPreparing inputsRepair and validate a structure
Work and files
ProjectsJobsFiles and collectionsWorkflowsTroubleshooting analyses
Batches
OverviewCreating a batchCSV batchesPairing and combinationsBatch examplesCSV referenceCSV format and limits
Research tools
AI AssistantDatabase searchViewers
Plans and credits
CreditsJob credit usageWorkflow credit usagePricing and billingBilling and invoices
Workspace and account
Workspaces and seatsSharing and permissionsCancellation and renewalsAccount deletion
Appearance
Back to ProteinIQ