TeachOpenCADD 2021 release is out!
2021.12.09 · By Dominique Sydow
Have you heard of TeachOpenCADD?
-
Yes? Perfect, we have good news! We are back with more content (12 fresh new notebooks!), a new website, simplified installation options, and much, much more.
-
No, you haven’t? Well, we are happy you found your way here!
We will show you what TeachOpenCADD has to offer and tell you more about the exciting details on the new release!
How did it start?
In 2019, we launched TeachOpenCADD, a teaching platform for cheminformatics and structural bioinformatics that show-cases how to perform central tasks in computer-aided drug design (CADD).
- We are using only open source Python packages and data resources. This setup makes the material accessible to everyone.
- Each topic is covered in a Jupyter notebook always following the same aim-theory-code-discussion-quiz scheme. This all-in-one approach makes the topics comprehensible to users from all backgrounds. Thanks to this setup, these tutorials are also suitable for oral presentations, hence we call them talktorials (talk + tutorial).
- TeachOpenCADD is used by students and scientists who seek to learn or teach these topics, or who need a workflow template for their own research.
- For users who prefer GUI-based pipelines, we offer TeachOpenCADD-KNIME workflows for some of the topics on the KNIME Hub.
TeachOpenCADD’s 2021 release
These are the top 3 major updates in this release:
- New website for read-only browsing of our content!
- More topics — now in total 22 — covering online queries, cheminformatics, and structural bioinformatics tasks!
- Easy local installation via our
conda
package for users who want to execute and modify the talktorials!
You have heard enough? Great, enjoy our website! — You are interested in more details? Wonderful, please read on.
New website and content
Since 2019, we have been working a lot on extending and improving our material. Let’s start with the feature that will make all our material much easier and quicker to access - our new website!
The website hosts a static view of all our Jupyter notebooks, which makes browsing through the TeachOpenCADD content convenient, fast, and searchable. We extended our cheminformatics-focused 2019 release with a lot of new topics from structural bioinformatics, and we demonstrate in detail how to query different online resources from within a Python pipeline.
Querying online APIs/servers from Python
T011
Querying online API webservicesT001
Data acquisition from ChEMBLT008
Data acquisition from PDBT012
Data acquisition from KLIFST013
Data acquisition from PubChem
Cheminformatics
Mostly powered by the RDKit!
T002
ADME and lead-likeness criteriaT003
Unwanted substructuresT004
Compound similarityT005
Compound clusteringT006
Maximum common substructureT007
Ligand-based screening: Machine learningT009
Ligand-based pharmacophoresT021
One-hot encodingT022
Ligand-based screening: Neural networks
Structural bioinformatics
T010
Binding site similarity and off-target predictionT014
Binding site detection (Proteins.Plus)T015
Protein ligand docking (smina)T016
Protein-ligand interactions (PLIP)T017
Advanced NGLview usage (NGLview)T018
Automated pipeline for lead optimizationT019
Molecular dynamics simulation (OpenMM)T020
Analyzing molecular dynamics simulations (MDAnalysis)
TeachOpenCADD and FAIR principles
Since the platform has been growing and will continue to grow, we set up a continuous integration (GitHub Actions) that tests all notebooks — thanks to pytest
and nbval
— on a regular basis to ensure functional, reusable, and reproducible pipelines.
We love open research. We comply with the FAIR principles for data and software as follows:
- findable — our code is registered on GitHub and
conda-forge
- accessible — our code and all its dependencies are free to download
- interoperable — we support Windows, Linux, and MacOS for Python >= 3.7; dependencies are defined and managed within the
conda
ecosystem - reusable — re-use is easy thanks to our
conda
package (CY BB 4.0 license); maintenance is monitored by our continuous integration
Contributors
This new TeachOpenCADD release was possible thanks to huge contributions from Dominique Sydow, Jaime Rodríguez-Guerra, Talia B. Kimber, David Schaller, Corey J. Taylor, Yonghui Chen, Mareike Leja, Sakshi Misra, Michele Wichmann, Armin Ariamajd, and Andrea Volkamer. It was a pleasure working with you all on this next milestone!
Furthermore, we thank Piedro Gerletti, Ahmed Atta, Melanie Vogel, Abishek Laxmanan Ravi Shankar, and Maria Trofimova for their work on initial drafts for new talktorials; and we thank Jeffrey R. Wagner, Richard Gowers, and Floriane Montanari for their support on improving code and testing of TeachOpenCADD.
TeachOpenCADD relies on external resources; regarding the newly released talktorials, we especially thank Patrick Kunzmann (biotite), Albert Kooistra (KLIFS), and Hai Nguyen (NGLview) for their help with questions and issues.