The tumour hacker

Vincent Zoete is developing computer tools to fight cancer. The chemist, who heads two groups at the University of Lausanne and the Swiss Institute of Bioinformatics, predicts the effects of new drugs.

In front of his first computer at age 12, Vincent discovered the joys of coding. The boy soon realised that he had more fun programming the machine himself rather than simply playing games developed by others. He did take a few computer courses during his time at engineering school in Lille, but the young man wanted to pursue more serious studies. “I didn’t think IT would become such an important thing,” he smiles some thirty years later. So he put his passion to the side and chose chemistry after discovering his talent for the subject.

He continued his studies with a PhD in Drug Design, a field that aims to understand the atomic structures of active ingredients in order to anticipate their effects. Because “everything is a bit chemical,” explains Vincent Zoete – from medicines to the proteins that form the building blocks of our cells. Knowing their structures on an atomic scale allows us to predict their interactions. And we can, for example, design a potential new drug so that its shape fits perfectly into a pocket on a protein, thereby modifying its activity.

Doing chemistry with a Computer

During his PhD, the researcher met several people who changed his life. On the personal side, one of his colleagues became his life partner. On the professional side, his discovery of computer modelling also resulted in new encounters. He was so fascinated by this emerging technology that he moved to another laboratory to complete his PhD with a pioneer in the field, Professor Martin Karplus, who received the Nobel Prize in Chemistry in 2013 for his calculations and simulations that enable us to understand and predict chemical processes. Vincent Zoete was immediately intrigued by modelling, which made it possible for him to “bring IT into chemistry.” The premise is to feed the computer with chemical and physical knowledge so as to study the interactions between drugs and proteins without necessarily having to synthesise them. “We’re not at the stage where we can predict 100% that the potential drug will bind, act, not be toxic and be distributed and metabolised correctly throughout the body,” clarifies Zoete. But it is possible to create functional molecules rationally, and to anticipate certain toxic effects or a lack of efficacy. This allows laboratories to focus on molecules offering the greatest chance of success, and to considerably reduce the number of failures at the stages of animal and, later, human testing.

While learning about modelling from one of the pioneers in the field, Vincent Zoete soon began programming for oncologists. In 2004, this collaboration led him to the Swiss Institute of Bioinformatics (SIB) in Lausanne, where he became the head of research for Professor Olivier Michielin’s group, whom he had met a few years earlier through Professor Karplus. The Swiss Institute of Bioinformatics brings together a network of specialists in fields ranging from genetic data analysis to molecular modelling. Vincent Zoete is delighted with this interdisciplinary ecosystem: “The multiplicity of scientific cultures means that concepts specific to one area of research can be transferred to other areas where they are new and useful.”

Predicting which cells will combat tumours most effectively

Since 2017, he is a professor in the Oncology Department of the Faculty of Biology and Medicine at the University of Lausanne and the Lausanne University Hospital (UNIL CHUV), and part of the Lausanne Branch of the Ludwig Institute for Cancer Research. He uses his various areas of expertise to help develop new cancer therapies. He is interested in optimising cellular immunotherapies. During these treatments, patients’ immune cells that are potentially capable of attacking the tumour – T lymphocytes – are harvested, uniformly amplified and reinjected. This homogeneous amplification favours quantity, but not always effectiveness since not all T lymphocytes have the same capability to fight tumours. Plus, tumours often mutate, which can lead to resistance.

In one of his most recent projects funded by the Swiss National Science Foundation (SNSF), Vincent Zoete and his team developed an algorithm for classifying T lymphocytes into families. This classification helps to predict which cells are most likely to bind to a tumour and destroy it. Scientists can then select and amplify the cells that will fight a particular cancer most effectively. This targeted amplification of T lymphocytes increases the effectiveness of immunotherapies while reducing the risk of resistance.

Making drug development child’s Play

The development of this tool is just one of the researcher’s many projects. He is also working with his teams on algorithms that should make it possible to produce targeted chemotherapies. And he uses modelling to illustrate how drugs work or the impact of mutations identified in patients. Coding behind his screen, the bioinformatician is able to attack tumours from all sides. But the tools he has developed don’t stop there. As a group leader at the SIB, he provides the scientific community with numerous websites and databases, contributing to other drug development projects. “We create knowledge in order to operate tools that allow us to create more knowledge,” explains the researcher. He adds enthusiastically: “We can see that people are using our algorithms and databases from the number of citations and requests for calculations and support that we receive.”

Vincent Zoete’s schedule is full with the management of his two groups. However, he is not content with just doing research: sharing it is essential to him. In addition to the six university courses he teaches, he is involved in communication projects aimed at the public. With his teams, he is preparing a board game on drug design. There will be two levels, one more advanced for students and the other for the general public. Will he have time to play it too? The programming enthusiast laughs: “I have to say that I don’t have much time for anything other than being a father of three and a researcher. To relax, I often cook. It’s not that different from chemistry!” But while he no longer has time to code for pleasure, his enjoyment of coding remains: “In the midst of running laboratories, it’s essential for me to still do research with my own hands, to do some programming myself. It allows me to test the feasibility of projects before bringing my teams on board, and above all to stay in touch with the material I enjoy.”

The text of this press release, a download image and further information are available on the website of the Swiss National Science Foundation.

Wissenschaftliche Ansprechpartner:

Vincent Zoete
University of Lausanne
Faculty of Biology and Medicine
Oncology Department UNIL CHUV
Ludwig Institute for Cancer Research, Lausanne Branch
Phone: +41 (0) 21 692 5907
Email: vincent.zoete@unil.ch

Weitere Informationen:

https://www.snf.ch/en/qRJ7QGzpLwSLiQFh/news/the-tumour-hacker