Winning Projects

– Memory and/or Peace

Armelle GOSSELIN-GORAND

CMAP : Chaire Mémoire et avenir de la paix

The project examines and analyses the conditions of peace and peacekeeping with regard to supporting victims and justice considerations, drawing on work relating to individual and collective memory;  It promotes multidisciplinary research (law, history and neurosciences) in an international context in connection with territories like Normandy, Senegal, Armenia, Lebanon and Madagascar, that are deeply affected by conflicts. It allows to educate people about human rights guarantees and the key to understanding the elements that make it possible to build a lasting peace.

Peggy QUINETTE

MEMOIRE(S) :  Mémoriser l’Histoire : La Libération de la Normandie comme composante de l’identité individuelle et collective

This project explores how childhood and adolescent experiences during the Battle of Normandy influence individual narratives through specific psychological and social mechanisms. Focusing on the unique Norman context, it examines the evolution of personal stories in relation to societal and media shifts. Additionally, it compares these accounts with narratives from other occupied and non-occupied regions to assess how geographical context shapes the meaning-making process.

Thomas HINAULT

VRTIME : Virtual reality modulations of the perception and memory of time

The temporal dimension of our perceptions and memory is central in the adaptation to a constantly accelerating world, with its everincreasing demands. The aim of this project is to bring together a wide range of scientific and technical expertise in the fields of temporal processing and Virtual Reality (VR) at the heart of the Normandy research ecosystem. VRTIME proposes to initiate an innovative research topic combining fundamental explorations and clinical applications: the use of virtual reality to study the perception and memory of time, a mechanism at the heart of normal and pathological cognitive functioning, yet traditionally understudied because of difficulties in evaluation that immersive technology can overcome. VR offers unique possibilities for creating modified environments that modulates our relationship with time, while retaining both an ‘ecological’ dimension, i.e. taking into account the interaction between an individual and his or her natural environment, and total experimental control.

– Biodiversity and the climate risks

Bastien LEMAIRE

SEPIASpeak : Synthesizing Ethology and Programming for an Integrative Analysis of Sepia officinalis communication 

SEPIA Speak aims to investigate the visual communication system of the common cuttlefish. To this end, we will first identify the visual signals used during social interactions. We will then develop a biomimetic robot capable of realistically reproducing these signals to experimentally test the complexity of cuttlefish communication. This approach, combining ethology, robotics and computer science, could provide new insights into the evolution of animal communication.

Bernadette TESSIER

ESTUARIX : Estuaries under climatic risks. Lessons from the recent past

The objective of the Estuarix project is to investigate the impacts of these pressures on the morpho-sedimentary dynamics of estuarine systems, and on associated anthropogenic landscape transformations, through the lens of the transition to the “Anthropocene”, a shift from natural climatic regimes to those driven by global change. This transition, initiated in the 19th century, corresponds both to the end of the Little Ice Age, a cold and stormy natural climatic episode, and the onset of major coastal engineering works, followed in the early 20th century by the emergence of anthropogenic climate change.

– Artificial intelligence

Frédéric JURIE 

CAI4Science – AI for Science at Caen

CAI4Science (AI for Science in Caen) is an ambitious research project that aims to harness the transformative potential of artificial intelligence (AI) to accelerate scientific discovery across various fields. The project brings together the expertise of four research laboratories at the University of Caen Normandie (CIMAP, CRISMAT, LPC, and GREYC) to develop and deploy advanced AI/machine learning technologies tailored to the specific needs of the domain sciences. The main research thrusts include: the integration of physical models and machine learning, the use of foundation models for scientific discovery, the development of surrogate and generative models for scientific simulations, as well as fundamental advancements in core AI areas such as optimization, interpretability, and robustness. The project also aims to train and disseminate these innovations to the scientific community through summer schools, workshops, and doctoral programs.

Gaël DIAS

MENTAL.AI@CaeSAR : Artificial Intelligence for Mental and Brain Health

The MENTAL.AI@CaeSAR project aims to develop artificial intelligence tools to improve the early detection, monitoring, and diagnosis of psychiatric and neurodegenerative disorders. It relies on the analysis of multimodal data, including patient-therapist interviews, language and speech signals, and neuroimaging data, to identify digital and biological markers of these conditions. The project combines expertise in artificial intelligence, neuroscience, linguistics, and medicine to design advanced clinical decision-support tools. It also explores the automatic generation of clinical notes to reduce the administrative workload of healthcare professionals. Led by an international and interdisciplinary consortium, the project seeks to contribute to more personalized and predictive mental and brain healthcare.

Pierre LARRIVEE

The BREAKTHROUGH project contributes to the quality of large language models in terms of (varieties of) languages, as well as the quality of their automatic an-notation, in order to solve the crucial problems of what might be called noisy data – the narrowness of the practices represented, and the still approximate nature of enrichments.

The advances represented by large language models, as symbolized by ChatGPT, depend on the possibility of megadata-based training. Such megadata are only available for a subset of languages. For historical French, there are no models available for the period between 1500 and 1900; for the linguistic varie-ties of Northern and even Southern France, some models are being developed, but on a very limited empirical basis; for contemporary French, it is mainly for-mal writing that has been analyzed, rather than spontaneous speech. Gaps are often filled by recourse to existing models for nearby languages, for which the analyzed varieties constitute noisy environments: the distance between the ana-lyzed varieties and existing models for other languages thus requires considera-ble investments of time and energy on the part of human users to correct the syntactic annotation. Yet advanced annotation of linguistic data is essential not only for language study and computer sciences, it also has a considerable poten-tial industrial impact in the fields of user identification, opinion and argumenta-tion mining, and idiomatic translation, including for varieties which only have a limited place in the digital world.

Romain HERAULT

Elie BESSERER-OFFROY

Concerto : Collaborative Network for Cancer Radionuclide Therapy Development

Personalized medicine aims to tailor cancer treatment to each patient’s tumor profile, improving efficacy and reducing side effects. A leading approach in this field is radiotheranostics, which combines imaging and radionuclide therapy for precise, targeted cancer care. Our international consortium develops peptide-based probes targeting overexpressed G-coupled receptors (GPCRs) such as APJ, NTSR1, and GHSR in aggressive cancers like ovarian cancer and glioblastoma. These probes can both detect and damage cancer cells making them valuable tools, adaptable for diagnostic and radionuclide therapy. By integrating chemistry, molecular biology, and imaging, the project seeks to advance precision oncology and promote public engagement through a summer school dedicated to scientific communication.

Samuel VALABLE

TargetedNanoOnco : Functionalized Multimodal therapeutic nanoparticles to synergize radiotherapy and immunotherapy in Glioblastom

Nanoparticles hold great promise for the treatment of glioblastoma by targeting hypoxia and enhancing immune responses. However, following intravenous administration, the amount of nanoparticles reaching the tumor remains limited. Here, we propose grafting moieties that specifically recognize proteins overexpressed in the glioblastoma vasculature and on the surface of immunosuppressive macrophages.

Anne Sophie VOISIN-CHIRET

CAR-DD : Cancer Advanced therapy through Rational Degrader Design

The CAR-DD project aims to develop a new generation of anticancer drugs known as degraders (PROteolysis Targeting Chimeras, or PROTACs). These innovative molecules do not merely block a target; they induce its complete destruction through the cell’s natural recycling system, called the proteasome. In this project, the target is Mcl-1, a protein that protects cancer cells from cell death. Overexpression of Mcl-1 is frequently observed in several types of cancer, including ovarian, lung, lymphoma, and pancreatic cancers, and constitutes a major mechanism of resistance to conventional therapies. Current Mcl-1 inhibitors have shown important limitations, particularly cardiac toxicity. Degraders offer a promising alternative to selectively eliminate this protein in tumor cells while minimizing side effects. The CAR-DD project, led by the Centre d’Etudes et de Recherche sur le Médicament de Normandie (CERMN) within the University of Caen Normandy, builds on very encouraging results obtained with a first generation of patented degraders that demonstrated activity in chemoresistant ovarian cancer cell models. CAR-DD aims to go further by designing degraders that are even more effective, selective, and safe. This will be achieved by combining molecular modelling, artificial intelligence, and medicinal chemistry approaches, while exploring new cellular enzymes known as E3 ligases, capable of initiating protein degradation. The candidate drugs will first be tested on cancer cell lines and then on patient-derived tumor organoids.

Siamak HAGHDOOST

PancResist : Strategies to Overcome Radioresistance in Pancreatic Cancer: A Collaborative Approach Targeting Cancer Stem Cells, Gene Pathways combined with Hadron Therapy

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancers worldwide due to its late diagnosis, early spread, and strong resistance to current treatments like chemotherapy and radiotherapy. Standard X-ray radiotherapy often fails to control PDAC because the tumor contains special cells called cancer stem cells (CSCs), which are highly resistant to radiation and drive tumor relapse. Our project aims to develop a new, more effective treatment strategy for PDAC by focusing on these resistant CSCs and using advanced radiation therapies called hadrontherapy (HT), including proton therapy and carbon ion radiotherapy (CIRT). Unlike conventional X-rays, hadrontherapy delivers highly precise, powerful radiation that causes complex DNA damage in cancer cells, making it harder for them to repair and survive, especially the resistant CSCs. CSCs survive radiation because they have strong antioxidant defenses regulated by a protein called Nrf2, and they thrive in lowoxygen (hypoxic) environments where other proteins called Hypoxia-Inducible Factors (HIFs) help maintain their stem-like and resistant nature. CSCs also possess efficient DNA repair systems that help them fix radiation damage. By targeting Nrf2, HIFs, and DNA repair pathways simultaneously, we hope to weaken these defences, increase cancer cell death, and reduce the chance of tumor relapse. Our research involves isolating CSCs from pancreatic cancer cell lines and patient-derived tumors to study their behaviour and resistance. We will test different radiation types X-rays, protons, and carbon ions to see which is most effective against CSCs. Using cutting-edge gene-editing tools like CRISPR, we will precisely inhibit Nrf2 and HIF proteins and combine this with DNA repair inhibitors to amplify the effects of radiation. Additionally, this project will explore how these treatments influence the immune system’s ability to fight cancer. PDAC often suppresses immune responses, but advanced particle radiotherapy may help stimulate anti-tumor immunity, especially when combined with inhibitors targeting tumor microenvironment factors. Finally, we will translate our laboratory findings into animal models of pancreatic cancer to test the best treatment combinations in living systems. By monitoring tumor growth, immune responses, and survival, we aim to identify new therapies that can ultimately improve patient outcomes. This interdisciplinary project brings together experts in cancer biology, radiation physics, immunology, and molecular genetics across multiple institutions. It addresses a critical need for novel approaches against PDAC’s resistance and poor prognosis by integrating innovative radiation techniques with targeted molecular therapies and immunomodulation. Through this comprehensive strategy, we aspire to pave the way for future clinical trials, offering hope for more effective and personalized treatment options for pancreatic cancer patients.

Aurélien JUSTET

SomaFibRad : Cancer Advanced therapy through Rational Degrader Design

Pulmonary Fibrosis (PF) describes a chronic lung disease in which lung tissue becomes scarred over time. This condition ultimately leads to chronic respiratory failure creating and eventually death within 2-5 years after the diagnosis in patient with Idiopathic Pulmonary Fibrosis (IPF), one of the most frequent diseases among PF. While lung fibrosis can be idiopathic or triggered by various factors, such as radiation, the interplay between genome instability, somatic mutations, and their impact on lung tissue integrity and fibrosis progression remains an underexplored research domain. Our preliminary results suggest a significantly increased rate of somatic mutations in IPF as compared to controls. We hypothesize that these mutations trigger molecular cascades that ultimately lead to the fibrotic remodeling of lung parenchyma. We propose to use radiation as a controlled tool to induce fibrosis, genome instability, and somatic mutations in lung tissues. The originality of the proposal lies in the combined use of an original ex vivo model based on Human Precision Cut Lung slices culture, a radiation-induced murine lung fibrosis model, and the application of cutting-edge technologies to assess the effect of radiation on the cellular transcriptomic signature but also the validation of this mutation signature in a real-life cohort in collaboration with internationally renowned experts in the field of lung fibrosis, transcriptomics and genetics. By incorporating genomic analysis and transcriptomic analysis we aim to establish the connection between hypoxia, aging, and radiationinduced somatic mutations and the development of lung fibrosis. This integrated approach will provide a comprehensive understanding of the genomic alterations associated with radiation-induced fibrosis, facilitating the identification of novel therapeutic targets and personalized treatment strategies for patients affected by this condition. The feasibility of this project relies on the collaboration of internationally recognized multidisciplinary experts in various fields, including cellular effects radiation, hypoxia, bioinformatics analysis, multi-omic approaches applied to pulmonary fibrosis, and genetics of these pathologies. Securing funding would provide crucial assistance in accelerating the acquisition of robust preliminary results, notably in establishing the model through the recruitment of a postdoctoral researcher to apply for an ERC Starting Grant within two years. In a medium-term perspective, this funding will also have a significant impact on the establishment of a team focused on fibrogenesis, led by Dr. Justet within the ISTCT.