Chemical Data, Discovery and Design

Canvas site

Hello! This book will hopefully act as the main resource for content for the more experimentally-focused, fundamental (!) part of CHEM502.

It is very much a work in progress. Please let me (Sam) know if you spot any errors, cannot access any of the expected content, have suggestions for additional content or resources to include (I will do my best on the latter). If you hover over the Octocat icon (the GitHub cat) at the top right of any page, there is an option to “Open issue” and you can write a message and I should get a notification. Or just drop me an email.

You can access the GitHub repository containing the markdown files and notebooks here or by hovering over the Octocat button and clicking on “Repository”.

The code is available in the book directory and is broken down into topics.

There is a readme and environment yaml/requirements files in the root directory of the repository which should hopefully enable you create a conda environment in which you can run the notebooks.

The repo is also hosted on Binder, but Binder seems to be having some issues. Theoretically, once it is restored, you should be able to work interactively with the notebooks by hovering over the rocket launch button at the top on a notebook page and clicking on “Binder” or the “Live Code” button.

Chemical data

• Molecular descriptors and similarity
• Notebook exercise - working with spectral data (or patterns)

Data fundamentals

• Exploratory Data Analysis (EDA)
• Notebook - Visualisation in EDA
• Workshop - Exploratory Data Analysis

Intro to ML

• Introduction to Machine Learning
• Notebook - ML for thermochemical data prediction

Data in the digital chemistry lab

• Workshop - Structured data and digital chemistry

Warning

The book is under active development. Content will be added throughout the semester.