Logging RDKit Molecular Data

RDKit is a popular open source toolkit for cheminformatics. In version 0.12.7 of the wandb client library, we added wandb.Molecule support for rdkit data formats. In particular, you can now initialize wandb.Molecule from SMILES strings, rdkit.Chem.rdchem.Mol objects, and files in rdkit-supported formats, such as .mol.

This Colab showcases how you can log rdkit molecular data in Weights & Biases and visualize it both in 3D and 2D.

###Click here to view and interact with a live W&B Dashboard built with this notebook.

Overview

In this example, we're using Google Colab as a convenient hosted environment, but you can run your own training scripts from anywhere and visualize metrics and data with W&B's experiment tracking tool.

As an example, we will initialize wandb.Molecule objects from different rdkit formats and log them to a wandb.Table for visualization.

Let us save a .mol file:

2D Views of a Molecule

First, we'll log 2D views of molecule using the wandb.Image data type.

3D Representations of Molecules

Now, let us log 3D representations of a few sample molecules using a wandb.Table.

Log Molecular Data to W&B

This will produce the following visualization:

More about Weights & Biases

We're always free for academics and open source projects. Email carey@wandb.com with any questions or feature suggestions. Here are some more resources:

1. Documentation - Python docs
2. Gallery - example reports in W&B
3. Articles - blog posts and tutorials
4. Community - join our Slack community forum

Sign up or login to W&B to see and interact with your experiments in the browser.