Spyros Lytras is one of four young scientists selected in 2025 to lead new research groups at the Institut Pasteur. He has headed the “Antigen Evolution & Design laboratory” 5-year Group (G5) since June 2026.
This article is part of a series presenting the research projects of the scientists selected in 2025 by the Institut Pasteur to launch their 5-year research group (G5). You can find the other articles here:
Sarah Merkling - Mosquitoes: unlocking the black box of virus transmission
What my research is about
I’m trying to understand how the surface proteins of viruses (the ones that attach to our cells and are recognized by our immune system) change to enable variants, maintaining virus circulation. By examining the diversity of these proteins among related viruses, we can detect patterns and figure out which parts to target to ultimately make better vaccines.
How I intend to find out
I use AI models to predict which elements are likely to change or not change. It’s the same logic as LLMs (large language models): a protein is like a sentence, made of connected building blocks that follow a pattern. LLMs trained on millions of data points try to “guess” the next most likely word. Similarly, our models analyze extensive datasets of proteins taken from virus relatives to “guess” the next amino acid in a sequence and whether it’s likely to differ from what we already know. If we take flu for instance, we can train our AI on the full diversity of influenza viruses to predict which amino acid is most likely to change and give rise to new variants.
What it could be used for
It could help design more effective vaccines. Traditionally, to create a vaccine we sequence the current, most worrisome strain and design a vaccine against it. But with a predictive model, we could design vaccines for strains that do not yet exist or create a sort of chimera grouping the strains most likely to appear, to broaden immunity. We wouldn’t have to wait for a threat to emerge to be ready to defend against it. We could extend this method to viruses for which we don’t have much data, like Ebola, or to pathogens that may increase their circulation in unexpected places because of climate change.
Why do it at the Institut Pasteur ?
The potential for “next-door” collaboration. The institute is already at the forefront of vaccine research, and with the opening of the new Center for Vaccinology and Immunotherapy, it will be even easier to initiate collaborations and to naturally meet people whose work is highly relevant to my research.
A piece of advice for prospective young researchers
Try to develop leadership skills. It’s not something typically included in scientific curriculums, but I went very quickly from “doing interesting research” to managing a team, figuring out who’s doing what, and nothing prepared me for that.
