Computer simulations are revolutionizing scientific discovery, providing unprecedented insights into systems ranging from molecular interactions to complex materials and biological processes. While Density Functional Theory (DFT) and Molecular Dynamics (MD) remain foundational methods in computational molecular and materials modeling, they often struggle to capture the full complexity of real-world systems. To overcome these limitations, cutting-edge techniques have emerged, including high-performance algorithms accelerating DFT simulations, data-driven and machine-learning approaches enhancing the accuracy of MD, and advanced embedding strategies, which enable simulations of chemical transformations at unprecedented scales. At their core, these methods share a common goal: seamlessly bridging time and length scales while maintaining quantum mechanical accuracy, unlocking new possibilities for multiscale modeling.
The School for Quantum-Mechanical, Data-Driven, and Multi-Scale Simulations (QDMS) equips participants with the latest methodologies to bridge time and length scales with quantum accuracy. Through hands-on training and expert-led lectures, attendees will explore cutting-edge techniques — from high-performance quantum simulations to machine-learning-enhanced modeling — empowering them to tackle the most challenging problems in computational science.
The 2025 QDMS school is made possible by the support of the National Science Foundation (NSF) with the objective of bringing state-of-the-art methods and software for quantum-mechanical, data-driven and multi-scale simulations from molecular systems to materials.
Topics
• Ab initio simulations based on density functional theory (DFT)
• Advanced subsystem DFT simulations of materials (sDFT)
• Many-body molecular dynamics simulations of solutions and interfaces (MB-MD)
• Materials simulation in environment using ab initio continuum models (QM/C)
• Polarizable embedding of molecules using data-driven many-body potentials (QM/MB-MM)
Participation
The QDMS 2025 School targets early career researchers, including graduate students and postdoctoral researchers. It will take place at UC San Diego, at the San Diego Supercomputer Center. Registration includes lodging at UC San Diego housing near the venue, with meals included.
APPLY HERE (deadline April 15)
School format
The school will focus on live lectures, tutorials, and student-driven projects centered around scientific computing, leveraging open-source software developed by the collaborating groups.
School software stack:
• LAMMPS
• QUICK
• eQEpy
• Environ
Local organizers
Francesco Paesani (University of California, San Diego)
Etienne Palos (University of California, San Diego)
Ethan Bull-Vulpe (University of California, San Diego)
Henry Agnew (University of California, San Diego)
Xuanyu Zhu (University of California, San Diego)
Co-organizers
Michele Pavanello (Rutgers University)
Oliviero Andreussi (Boise State University)
Revati Kumar (Lousiana State University)