Manage experiment data, end‑to‑end

A set of tools for (meta)data handling in databases.

Explore tools Quick start Quick start (developers)

Find Your Path

Get a personalized list of resources based on your role and goals

  • Research Student (PhD/Postdoc/Masters)
  • Data Scientist
  • Data Engineer
  • Project Manager
  • Principal Investigator
  • Lab Administrator
  • DevOps/SysAdmin (VM deployment)
  • Upload experiments to a database
  • Upload simulations to a database
  • Upload data from lab instruments
  • Browse and analyze existing experiments
  • Write a Data Management Plan (DMP)
  • Deploy Altar on lab servers / cloud
  • Provide (secured) online access to a database
  • Backup / migrate a database (eg. local to server)
  • Manage users and permissions
  • Apps and web interfaces
  • Writing Python scripts
  • Detailed explanations
  • Quick explanations

Why FAIR?

Altar helps you implement FAIR data principles (Findable, Accessible, Interoperable, Reusable) for your experimental research:

  • Findable: Search and filter experiments by any parameter, making data discovery effortless
  • Accessible: Web interfaces and APIs provide multiple access points to your data
  • Interoperable: Standard formats (Sacred, MongoDB, MinIO) ensure compatibility with other tools
  • Reusable: Complete metadata and provenance tracking enables reproducibility and reanalysis

By centralizing your data in structured databases with proper metadata, Altar makes your research more transparent, reproducible, and valuable to the scientific community.

What you can do

Upload experimental data
Store large raw data
Store metadata and results
Visualize experimental data
Filter, Visualize and download datasets

Tools overview

AltarSender

Application to send (meta)data stored in local folders to a database where they can be organized and visualized.

Learn more →

AltarViewer

Desktop app to connect to MongoDB, select a database, and launch Omniboard in Docker for experiment visualization.

Learn more →

AltarExtractor

Application to explore the database, filter experiments, visualise data and export datasets.

Learn more →

Quick start

1

Install Docker Desktop

Download and install Docker Desktop (free for personal use). Open it and wait until it shows 'Engine running'.

2

Download and set up MongoDB

Follow the MongoDB community instructions to download MongoDB and launch it. By default, it runs on port 27017. Check MongoDB content with MongoDB Compass app.

3

Send your experiments

Download AltarSender from the Altar Releases (or run from source via the guide). Map your folders and upload experiments to MongoDB and artifacts to MinIO.

4

View your data

Download AltarViewer from the Altar Releases (or run from source via the guide). Enter the port or MongoDB URI and click Connect; then Launch Omniboard.

5

Explore and export data (optional)

Follow the AltarExtractor guide to run it with Docker. Open localhost:8050 to filter experiments, visualize data, and export as CSV.

Quick start for developers

1

Install Docker Desktop

Download and install Docker Desktop (free for personal use). Make sure it's running before proceeding.

2

Clone the repository

Download the ZIP or use git clone https://github.com/DreamRepo/Altar.git

3

Run the local stack

Go to AltarDocker, edit .env, then docker compose up -d. This sets up MongoDB (database) and MinIO (storage for large files).

4

Send experiments

Use AltarSender to map folders and push runs to Sacred and files to MinIO.

5

Visualize database content

Browse and filter experiments with Omniboard using AltarViewer

6

Visualise data and export datasets

Use AltarExtractor to interact with data using Pygwalker and download datasets as CSV.