Ilia Igashov

I am a third year PhD student at École Polytechnique Fédérale de Lausanne co-advised by Prof. Bruno Correia and Prof. Michael Bronstein, and grateful to be a fellow of the EPFLglobaLeaders program and to be awarded Marie Skłodowska-Curie doctoral fellowship. My research interests lie in geometric learning and its application in scientific domains such as biology and chemistry. I am especially focused on using geometric deep learning for protein-protein interactions and drug discovery.

I earned my BS in Applied Mathematics and Physics, and my MS in Applied Mathematics and Computer Science from the Moscow Institute of Physics and Technology. Additionally, I completed the M2 program in Science in Industrial and Applied Mathematics (MSIAM) at Université Grenoble Alpes. Throughout my Master's studies, I had an opportunity to intern at Inria Rhône-Alpes and Laboratoire Jean Kuntzmann, where I was working with Dr. Sergei Grudinin.

Curriculum Vitae

Publications

RetroBridge: Modeling Retrosynthesis with Markov Bridges
Ilia Igashov^*, Arne Schneuing^*, Marwin Segler, Michael Bronstein, Bruno Correia
ICLR 2024 Spotlight
Paper Code

Equivariant 3D-Conditional Diffusion Models for Molecular Linker Design
Ilia Igashov, Hannes Stärk, Clément Vignac, Victor Garcia Satorras, Pascal Frossard, Max Welling, Michael Bronstein, Bruno Correia
Accepted in Nature Machine Intelligence (2024)
Paper Code Hugging Face

Structure-based Drug Design with Equivariant Diffusion Models
Arne Schneuing, Yuanqi Du, Charles Harris, Arian Jamasb, Ilia Igashov, Weitao Du, Tom Blundell, Pietro Lió, Carla Gomes, Max Welling, Michael Bronstein, Bruno Correia
Under review, 2023
Paper Code Google Colab

Decoding Surface Fingerprints for Protein-Ligand Interactions
Ilia Igashov, Arian R. Jamasb, Ahmed Sadek, Freyr Sverrisson, Arne Schneuing, Pietro Liò, Tom L. Blundell, Michael Bronstein, Bruno Correia
Machine Learning for Drug Discovery Workshop, ICLR, 2022
Paper

6DCNN with Roto-Translational Convolution Filters for Volumetric Data Processing
Dmitrii Zhemchuzhnikov, Ilia Igashov, Sergei Grudinin
36th AAAI Conference on Artificial Intelligence, 2022
Paper

VoroCNN: Deep Convolutional Neural Network Built on 3D Voronoi Tessellation of Protein Structures
Ilia Igashov, Kliment Olechnovič, Maria Kadukova, Česlovas Venclovas, Sergei Grudinin
Bioinformatics, 2021
Paper Code Project Page

Spherical Convolutions on Molecular Graphs for Protein Model Quality Assessment
Ilia Igashov, Nikita Pavlichenko, Sergei Grudinin
Machine Learning: Science and Technology, 2021
Paper Code Project Page

Invited Talks

Symposium on Biochemical Machine Learning, IOCB Prague, July 2023. Fragment-Based Drug Discovery Using Protein Surface Fingerprints
Interplay between AI and mathematical modelling in the post-structural genomics era, CIRM, Marseille, March 2023. Equivariant 3D-Conditional Diffusion Models for Molecular Linker Design
M2D2 Seminar Series, November 2022. Equivariant 3D-Conditional Diffusion Models for Molecular Linker Design
CIS SV Retreat 2022 AI in life sciences: challenges and opportunities, EPFL Lausanne, October 2022. Fragment-Based Drug Discovery Using Protein Surface Fingerprints

Other Activities

Co-organizer of Learning on Graphs Conference (LoG)
Co-organizer of AI & the Molecular World Track @ AMLD EPFL 2022
Reviewer @ ICLR 2024
Reviewer @ NeurIPS 2023
Reviewer @ ICML 2023 Workshop Frontiers4LCD Reviewers
Reviewed for Bioinformatics and Cell