Run Your First Experiment
A complete walkthrough from logging into MINT to running your first analysis plugin — about 5 minutes.
[Screenshot: full MINT window showing the home dashboard, ready to start]
Prerequisites
- A running MINT instance (hosted, direct, or Docker — see Get Started)
- An account with at least the Member role
- At least one analysis plugin installed and visible to your role (your admin can install one from the marketplace if not)
Step 1: Create a project
From the home dashboard, click New project.
[Screenshot: New-project modal with name and description fields]
| Field | What it's for |
|---|---|
| Name | Human-readable label, e.g., "TCA flux pilot" |
| Description | One-line summary shown on the dashboard |
| Members (optional) | Lab colleagues to invite — they get the default project role |
Click Create. You're now inside the project page.
Step 2: Create an experiment
Click New experiment. MINT auto-assigns a unique code (EXP-001, EXP-002, …) — you don't pick it.
| Field | What it's for |
|---|---|
| Title | Human label |
| Type | Pick an experiment type registered by an installed design plugin (e.g., LCMS sequence, drug-response panel). Determines the design fields below. |
| Status | Starts at planned |
| Collaborators (optional) | Per-experiment overrides on top of project members |
Fill in the design fields exposed by the experiment type, then Save. The experiment is now in planned status. See Experiments for the status flow.
[Screenshot: experiment-detail page in planned status]
Step 3: Move to ongoing and upload data
Switch the status to ongoing. Most plugins gate result writes on ongoing or completed. Then upload the relevant artifacts (RAW files, sequence sheets, design plates) — the file picker shows the experiment-type's expected attachments.
[Screenshot: experiment file uploader with attachments queued]
Step 4: Run an analysis plugin
Open the Analyze tab on the experiment. Pick an installed analysis plugin from the dropdown, fill in its parameters, and click Run.
The plugin runs in its own isolated subprocess; MINT proxies its UI back into the page. Progress is tracked in the Jobs panel (top-right).
[Screenshot: analysis-plugin sidebar with parameters and Run button]
Approximate runtimes depend on the plugin and dataset size. The Jobs panel shows live status: queued → running → done (or failed).
Step 5: See your results
When the plugin finishes, the experiment's Results tab populates with whatever the plugin wrote (charts, tables, downloadable artifacts). Multiple analysis runs accumulate over time — every run is timestamped and attributed to the user who launched it.
[Screenshot: results tab showing accumulated analysis runs]
Step 6: Wrap up
Switch the experiment status to completed to lock it. Completed experiments are read-only by default; only owners and admins can re-open them.
Further steps
- Invite collaborators — see Members & roles
- Install another plugin — see Marketplace
- Use the CLI — see
mintoverview for scripted experiment + project access - Build your own plugin — see Plugin Development and the First analysis plugin tutorial