Ouail Zakary
Ph.D. · Postdoctoral Researcher · NMR Research Unit, University of Oulu, Finland
My research focuses on developing machine learning-driven approaches for the atomistic modeling of molecular and solid-state systems under real physicochemical conditions. By integrating quantum mechanics at the relativistic level with classical and path integral molecular dynamics, I aim to deliver highly predictive models for structural, dynamic, and spectroscopic observables.
A significant part of my work involves training equivariant graph neural networks and kernel regression models to develop ML interatomic potentials and ML NMR models that enable long-timescale simulations of complex materials.
MLIPs
Equivariant GNNs
¹²⁹Xe NMR
Porous Materials
MACE
Path Integral MD
HPC
Latest news
Preprint
March 2026
RuP preprint published on ChemRxiv
Manuscript uncovering a two-stage phase transition in RuP via fine-tuned MACE-MH-1 now available as a preprint.
Read more
Conference
April 2026
Oral presentation at EUROMAR 2026
Presenting work on ML-accelerated MD and noble gas NMR in porous materials at EUROMAR 2026, Lodge on Loch Lomond, Scotland.
Read more
Award
December 2025
JPCL paper selected as Supplementary Journal Cover
Our paper on equivariant neural networks for ¹²⁹Xe NMR in porous liquids featured as a Supplementary Cover of J. Phys. Chem. Lett.
Read more
Publication
January 2026
New paper in J. Phys. Chem. A
Collaborative paper on kernel and GNN-based NMR models for multi-element prediction published in JPCA with Laurila, Jacklin, and Lantto.
Read more
Grant
November 2025
Academy Research Fellowship application submitted
Submitted a Research Council of Finland Academy Research Fellowship application with collaboration letters from DKFZ and Imperial College London.
Read more
Publication
October 2025
C₆₀ manuscript submitted to Nature Communications
Submitted work on ML-accelerated path integral MD and ¹³C NMR of C₆₀ combining quantum nuclear effects with equivariant potentials.
Read more