2026 IEEE NTC TC10 Modeling & Simulation September Webinar

Title: Property-guided Diffusion Modeling for Efficient Exploration of Chemical Spaces

Date/Time:  23 September 2026, 17:00 Rome/Central European Time (16:00 UK, 11:00 New York, 08:00 California)

Speaker: Dr. Leonardo Medrano Sandonas, TUD Dresden University of Technology, Dresden, Germany

Register below to receive meeting link.

Abstract:

The rational in silico design of chemical compounds requires a deep understanding of both structure-property and property-property relationships across chemical compound space, as well as efficient methodologies for defining inverse property-to-structure mappings. In this presentation, I will discuss our recent efforts to leverage the “freedom of design” concept [Chem. Sci. 14, 10702-10717, (2023)] in the chemical space of drug-like molecules to develop QALPA (“Quantum-Aware Learning for Property-space Augmentation”), an efficient generative AI framework capable of designing molecular compounds with targeted quantum-mechanical (QM) properties. To this end, we have implemented a property-guided active learning approach that optimizes the performance of equivariant diffusion models within each property space. Generated molecules and their associated QM properties are validated through exhaustive DFT calculations at the PBE0+MBD level using the FHI-aims code as well as the recently developed EquiDTB framework [Phys. Chem. Chem. Phys. 28, 3949-3962, (2026)]. As a proof of concept, QALPA is applied to augment alloQM, a new QM dataset of allosteric drug molecules, illustrating its ability to efficiently populate sparsely sampled regions of molecular property space. More broadly, our results show that combining generative AI with efficient ML/QM methods provides a practical strategy for sustainable exploration of chemical space and accelerated molecular discovery.

Speaker Bio:

Dr. Medrano Sandonas is currently a research associate at Technische Universität Dresden (TU Dresden), Germany. Born in Lima, Peru, he obtained his Bachelor and Master degrees in Physics from National University of San Marcos. In 2018, he completed his doctorate at TU Dresden as a fellow of the International Max Planck Research School and DAAD. He previously undertook a postdoctoral training at the University of Luxembourg (2019-2024). His research focuses on developing quantum-informed AI methods to accelerate the exploration of chemical spaces and advance the understanding of (bio)molecular systems. In addition to his theoretical work, he actively contributes to multidisciplinary projects with academic and industrial partners addressing challenges in physics, biochemistry, and materials design. He was recently recognized by Royal Society of Chemistry Digital Discovery as a 2026 Emerging Investigator in machine learning and selected by the committee of UNESCO International Year of Quantum Science and Technology 2025 as a member of the prestigious group QUANTUM 100. He has authored numerous scientific publications (more than 50 peer-reviewed publications), serves as a reviewer for several international journals, and has delivered guest lectures at institutions worldwide.