Bootcamp for Data Chemistry 2025

Welcome to Bootcamp for Data Chemistry

22 Jul 2025

Overview

Dates: September 21 – 25, 2025
Location: University of Notre Dame

LOGO

Introduction

In today’s research landscape, chemistry is evolving from a discipline driven mainly by experimental intuition to one powered by data-centric discovery. Advances in machine learning, high-throughput simulation, and automated lab instrumentation are generating vast datasets that can reveal patterns no human could spot unaided. Harnessing these data effectively—whether to predict molecular properties, design greener reactions, or accelerate materials discovery—demands a new skill set at the intersection of chemistry and artificial intelligence. Our Data-Chemistry Bootcamp is designed to equip chemistry students with exactly this toolkit: foundational AI concepts, hands-on coding experience, and domain-specific case studies that show how algorithms translate into real scientific breakthroughs. By bridging chemical insight with modern data methods, participants will be prepared to drive the next wave of innovation in both academia and industry.

Please register in advance so we can reserve seating and meals.

  • Complete the form here
  • Registration deadline: 1 September 2025 (or until all places are filled)

We look forward to welcoming you to Notre Dame for a week of data-driven chemistry!

Invited Presentations

  • 5 external faculty (Nicholas Jackson – UIUC; Paul Zimmerman – U Mich; Dylan Anstine – MSU; Eunjae Shim – U Mich; Pushkar Ghanekar - Eli Lilly)
  • Industry scientists (AbbVie, Dow, Lilly, 3M, P&G) – attend selected sessions

Agenda at a Glance

Sunday, Sept 21

18:00 – 20:00

  • Optional: Welcome Dinner and Registration
  • Workshop overview and logistics
  • Brief participant introductions
  • Networking
    Location: Ivy Court Inn Dining Room

Monday, Sept 22 – Representing Chemistry to ML Algorithms

08:30 – 09:30

  • Registration & Beverage Service

09:30 – 10:30

  • Seminar 1: Evolution of Molecular Representations: From SMILES to Graphs (Olaf Wiest)

10:30 – 11:00

  • Break

11:00 – 12:00

  • Software Tutorial: RDKit Fundamentals
    • Loading and manipulating molecules
    • Computing molecular descriptors
    • Generating fingerprints
    • Substructure searching

12:00 – 13:00

  • Lunch & Networking

13:00 – 14:00

  • Seminar 2: Graph Neural Networks for Chemistry: Theory and Practice (Xiangliang Zhang)

14:00 – 14:30

  • Break

14:30 – 16:00

  • Project Tutorial: Data Scarce Molecular Property Prediction
    • Data loading and splitting
    • Feature engineering with RDKit
    • Implementing prototypical networks
    • Model evaluation and interpretation

Dinner on your own

Tuesday, Sept 23 – Design of Experiments & Active Learning

08:30 – 09:30

  • Coffee & Conversation

09:30 – 10:30

  • Seminar 1: Active Learning for Efficient Chemical Space Exploration (Paul Zimmerman)

10:30 – 11:00

  • Break

11:00 – 12:00

  • Software Tutorial: scikit-learn for Chemistry
    • Regression methods overview
    • Cross-validation strategies
    • Feature importance analysis
    • Gaussian processes basics

12:00 – 13:00

  • Lunch

13:00 – 14:00

  • Seminar 2: LLMs and Chemistry (Title TBD, Meng Jiang)

14:00 – 14:30

  • Break

14:30 – 16:00

  • Research Tutorial: Building an Active Learning Pipeline for Reaction Optimization
    • Setting up Ax/BoTorch for chemistry problems
    • Defining parameter spaces and constraints
    • Implementing acquisition functions
    • Integrating with experimental/computational workflows
    • Case study: optimizing catalytic reaction conditions

Dinner on your own

Wednesday, Sept 24 – Machine Learning Potentials

08:30 – 09:30

  • Coffee & Conversation

09:30 – 10:30

  • Seminar 1: Coarse-Graining and ML: Bridging Scales in Molecular Simulation (Nick Jackson)

10:30 – 11:00

  • Break

11:00 – 12:00

  • Software Tutorial: PyTorch
    • Library overview
    • Building custom model architectures
    • Training demonstrations
    • Debugging in training scenarios

12:00 – 13:00

  • Lunch

13:00 – 14:00

  • Seminar 2: Neural Network Potentials: Replacing Quantum Chemistry with ML (Dylan Anstine)

14:00 – 14:30

  • Break and Group Photo

14:30 – 16:00

  • Research Tutorial: Deploying ML Potentials as Quantum Chemistry Surrogates
    • Installing and configuring MACE/NequIP/Orb/AIMnet/UAM
    • Training a potential on custom data
    • Running MD simulations with ML potentials
    • Comparing accuracy/speed with DFT
    • Case study: reaction pathway optimization

Dinner on your own

Thursday, Sept 25 – AI Agents for Chemistry

08:30 – 09:30

  • Coffee & Conversation

09:30 – 10:30

  • Seminar 1: Title TBD (Pushkar Ghanekar)

10:30 – 11:00

  • Break

11:00 – 12:00

  • Software Tutorial: PyTorch Geometric Deep Dive
    • Building custom GNN architectures
    • Message passing frameworks
    • Attention mechanisms for molecules
    • Integration with molecular datasets

12:00 – 13:00

  • Lunch

13:00 – 14:00

  • Seminar 2: Title TBD (Margaret Hilton)

14:00 – 14:30

  • Break

14:30 – 16:00

  • Research Tutorial: Creating an AI Research Assistant for Chemistry
    • Using LLM APIs for chemical reasoning
    • Building a retrieval-augmented generation (RAG) system
    • Integrating PubChem/ChEMBL queries
    • Automated report generation
    • Demo: Literature review automation

16:00

  • Closing Remarks

Acknowledgement

We gratefully acknowledge the generous support of the ND SAI and C-CAS. Their commitment to fostering interdisciplinary collaboration and advancing data-driven chemistry has been instrumental in making this bootcamp possible, enabling us to bring together students, researchers, and industry partners for a productive week of learning and discovery.