Skip to content

Accessing Third-party Data

Simmate provides a set of Data Apps that simplify the process of downloading and querying data from external repositories. Instead of manually querying external APIs every time you run a script, these apps allow you to store the data locally in your Simmate database for lightning-fast access.

Supported Data Apps

For a full list of supported third-party datasets, please refer to the Apps Overview page. This includes:

  • Crystal Structures: Materials Project, OQMD, JARVIS, COD, and AFLOW.
  • Molecular Data: BCPC, CAS Registry, ChEMBL, and more.

Tip

The individual documentation for each app provides details on the data source, provider, and citation information.

Downloading Data (Remote Archives)

Downloading thousands (or millions) of entries through a REST API is often slow and inefficient. To solve this, Simmate uses Remote Archives — pre-packaged snapshots of these databases that can be downloaded and loaded in one step.

From the Command Line

To download the archive for a specific app (e.g., Materials Project): 

simmate database download materials_project

In Python

You can also load the archive from within your script. This is especially useful for managing specific tables:

from simmate.database import connect
from simmate.apps.materials_project.models import MatprojStructure

# This will download the zip file, unpack it, and save all structures to your DB.
# WARNING: This can take >1 hour depending on the dataset size.
MatprojStructure.load_remote_archive()

Tip

Once loaded, this data stays in your local database permanently. You only need to run this command once!

Updating Stability and Energy

Some data apps provide uncorrected energies. If you're using the Thermodynamics mix-in features (like energy_above_hull), you can update the stability information after loading the archive:

# Update stability for all structures in the table
MatprojStructure.update_all_stabilities()

# Update stability for a specific chemical system
MatprojStructure.update_chemical_system_stabilities("Y-C-F")

Why use Simmate Data Apps?

Unlike alternatives that query external APIs (like MPRester or matminer), Simmate Data Apps offer:

  1. Speed: Loading data from your local database is hundreds of times faster than making web requests.
  2. Stability: Your source data won't change unexpectedly in the middle of a study.
  3. Rich Querying: Use the full power of the Django ORM to filter data by spacegroup, elements, density, and more.
  4. Integration: Data is already in the Simmate format, ready to be used in workflows or with the Simmate Toolkit.