Machine Learning for Undergraduate Chemistry

Welcome to Machine Learning for Undergraduate Chemistry, a project designed to integrate the power of machine learning with foundational concepts in the undergraduate chemistry major. This initiative aims to enrich the educational experience of undergraduate students by introducing them to cutting-edge computational tools that are transforming the field of chemistry.

Project Overview

The project offers a suite of interactive lab programs packaged within the ChemAIstry Launcher, a user-friendly application that serves as a gateway to various machine learning experiments tailored for chemistry education. These programs are designed to align with the curriculum of introductory chemistry courses, providing hands-on experience with real-world applications of machine learning. There are currently two parts, each with four lab experiences.

Key Features

• Interactive Lab Programs: Access eigth main applications with the two ChemAIstry launchers each focusing on different aspects of chemistry and machine learning.

• User-Friendly Interface: The ChemAIstry Launchers presents an intuitive layout with easy navigation, making it accessible for students with varying levels of technical expertise.

• Educational Integration: Programs are developed to complement classroom learning, reinforcing theoretical concepts through practical application.

Objectives

• Enhance Learning: Integrate machine learning methodologies into chemistry education to deepen understanding and stimulate interest.

• Develop Skills: Equip students with computational skills that are increasingly valuable in scientific research and industry.

• Promote Innovation: Encourage critical thinking and problem-solving by exploring data-driven approaches to chemical phenomena.

Why Machine Learning in Chemistry?

Machine learning is revolutionizing the way chemists approach research and development, from accelerating drug discovery to optimizing materials design. By incorporating machine learning into undergraduate education, we prepare students for the evolving landscape of chemistry and empower them with the tools necessary for future success.

Get Started

Download ChemAIstry1 ExplorerPack.zip: The explorer pack has everything in blue below.

• Download the ChemAIstry1 Launcher: Access all lab programs through a single platform. Setup guide here.

• Download the ChemAIstry1 Laboratory Manual: Full instructors manual. 

• Download the ChemAIstry1 Datasets:  Datasets needed for the labs.

• Explore the Applications, Part 1:

  • Basic Residue Analyzer: Introduction to machine learning terms along with graphical and numerical measures of machine learning model quality.

  • ML Synthesizer: Visit Cobberland where matter is made up of kernels. Use machine learning to predice the EARitability, aMAIZEingness, and STOCKiness properties of kernels.

  • PEM Sorter: Use machine learning in a simulated qualtity control setting to classify and sort polymer batches based on properties.

  • Cobber Crystal Visualizer: Use machine learning of simulated cyrstallographic data from surfaces with point defects. Predice surface reactivity from the density of edge atoms.

• Access Resources: Visit the GitHub repository for source code.

Go further

Download ChemAIstry2 ExplorerPack.zip: The explorer pack has everything in blue below.

• Download the ChemAIstry2 Launcher: Access all lab programs through a single platform. Setup guide here.

• Download the ChemAIstry Laboratory Manual: Full instructors manual. 

• Explore the Applications, Part 2:

  • Titrating Robot: Introduction to reinforcement learning where we train a simulated robot "in silico" to perform an acid/base titration.

  • Chemioinformatics: Use an evolutionary algorithm to design a 1D ligand to bind to a 1D protein.

  • Hydrogen bonging: Use unsupervised machine learning to explore a simulated molecular data set and "discover" hydrogen bonding.

  • Bioplastics: Use a convolutional neural network alogrithm to explore the kinetics of loading a bioplastic with dye.

• Access Resources: Visit the GitHub repository for source code.

Join the Community

We invite educators, students, and enthusiasts to be part of this project:

• Educational Institutions: Integrate these tools into your curriculum to offer students a modern and engaging learning experience.

• Students: Enhance your understanding of chemistry and gain valuable computational skills.

• Contributors: Collaborate with us on GitHub to improve and expand the project.

Contact and Support

For more information, support, or to provide feedback:

• Website: www.darinulness.com

• Concordia College Chemistry Department: Visit the department page for additional resources and contact information.