Practical Cheminformatics With Open Source Software

A set of Jupyter notebooks for learning Cheminformatics. The links below will open the tutorials on Google Colab. This way you can run the notebooks without having to install software on your computer. Of course, you can also clone the repo and run these notebooks on your own computer.

Fundamentals

1. A Whirlwind Introduction to the RDKit for Cheminformatics
2. A Brief Introduction to Pandas for Cheminformatics
3. SMILES Tutorial
4. SMARTS Tutorial
5. Recursive SMARTS Tutorial
6. Reaction Enumeration, the basics
7. Reaction Enumeration, advanced
8. Enumerating Stereoisomers and Tautomers

Using datamol and molfeat to Streamline Cheminformatics Workflows

9. Data Manipulation, Descriptors and Clustering

Clustering

10. K-Means Clustering
11. Taylor-Butina Clustering
12. Self-Organizing Maps

ChEMBL and BindingDB

13. ChEMBL Ring System Analysis
14. Working With Drug Data from the ChEMBL Database
15. Analyzing Patent Data from BindingDB
16. Searching ChEMBL for Interesting SAR
17. Using Data from BindingDB to Build a Machine Learning Model

Misc Cheminformatics Analysis

18. Interactive Visualization of Chemical Space

SAR Analysis

19. Identifying Scaffolds
20. R-group Analysis
21. Positional Analogue Scanning
22. Free-Wilson Analysis
23. Matched Molecular Pairs
24. Matched Molecular Series

Machine Learning Models

25. Build and Test a QSAR Model in 8 Lines of Python
26. Building a Classification Model
27. Comparing Classification Models
28. Building a Regression Model
29. Comparing Regression Models
30. Performing Cross-Valiation

Biotite

31. Introduction to Biotite for Structural Bioinformatics
32. Evaluating Co-folding Results with Biotite

Active Learning

33. Active Classification
34. Active Regression
35. Active Shape Search

Neural Network Potentials

36. Tautomer Energies with Auto3D

Generative Models

37. Generative Molecular Design with SMILES-RNN

Bayesian Optimization

37. Using Bayesian Methods to Optimize Reaction Conditions

Tricks and Tips

38. RDKit Tricks and Tips

Additional topics will be added. Please reach out with any requests.

A Bit of History

These notebooks began as part of a two-day workshop I did at the University of Bonn in 2019. Since then, they have evolved, and additional notebooks have been added. Many of the notebooks started as posts on my blog, Practical Cheminformatics. I originally had everything running on Binder, but I found some capacity limits with larger groups and moved everything to Google Colab. I plan to continue adding to and revising these as long as people are interested.

Suggestions and Corrections Welcome

This is a work in progress. It's more than possible that these tutorials contain mistakes and/or typos. If you find something that should be corrected, please submit an issue or a PR. In addition, I'm always looking to improve the text. Please let me know if aspects could be explained more clearly. I'd also like to hear more about topics you'd like to see covered. If you find these tutorials useful, please consider starring this repo.

Acknowledgements

These tutorials wouldn't be possible without the work of Greg Landrum, Brian Kelley, and the RDKit team, as well as Cédric Bouysset and his work on mols2grid. Thanks to Noel O'Boyle, Rafael Lameiro, and Jennifer HY Lin for helping to keep the notebooks up to date as the underlying software libraries evolve.