This collaborative project between french and German aims at the identification and development of Transketolase (TK) variants that will tolerant specific structural modifications not (or only poorly) accessible to the wild-type enzyme. This will be achieved by a technique termed directed evolution, by which site-specifically and randomly mutated protein variants will be tested by enzymatic assays for their improved tolerance to non-natural screening probes. Among those objectives, the target defined for ketone acceptors probably is the most ambitious goal because none of the enzymes commonly studied for synthetic applications of carboligation (various aldolases, TK) is able to accept ketone electrophiles, most likely for steric hindrance. Availability of new mutant TK enzymes with novel catalytic abilities for the novel product structures as chiral building blocks will be used in the synthesis of chiral bioactive compounds and scaffolds of higher complexity.

Because half of cancerous people develop secondary bone tumors (metastases), it is of tremendous importance to design new materials which will lead both to cancer cells destruction and fast bone tissue regeneration. This project aims to combine the high bioactivity property of nanometric bioactive glass particles synthesized by sol-gel chemistry to the magnetic properties of iron oxide nanoparticles through the synthesis of heterostructures. These magnetic particles would bring not only the possibility to realize hyperthermia treatment (heat generation under alternating magnetic field) in order to selectively destroy cancer cells, but also would have a positive impact on heterostructures bioactivity and bactericidity. The relationships between their structural, morphological, magnetic and biological (cytocompatibility, bioactivity, bactericidity) properties will be evaluated.

In the last few years, machine learning has become an increasingly important part of UCA’s research programs. In the last three years, at least 10 thesis on IA have been started and several laboratories are directly involved in a project related to IA. Key investments have been made in computing capacities especially dedicated to IA within the Mésocentre (GPU cluster) (https://mesocentre.uca.fr/) thanks to CPER funding and UCA’s own resource. The purpose of UCA’s IA program is to strengthen the current IA research projects spread in many areas by creating a pool of resources and knowledge, available to all fields that will benefit from Artificial Intelligence at the service of a coordinated strategy to aim for IA excellence, both in research and training. To achieve this, UCA is working towards : - 1) increasing the human resources involved in IA research both in the understanding of machine learning and it societal implication as well as exploring new ways to apply machine learning and deep learning in the fields of expertise of UCA : especially agro-system, robotics, health and mobility and natural risk management, 4 main challenges identified by I-SITE Cap 2025. Other key areas of the university area such as medical imaging, genomics, human social sciences (Ethics section; Cognitivism section) will be impacted ; - 2) investing in more IA dedicated computational capacity of its labs and the Mesocentre ; - 3) developing IA dedicated training programs for student, and researchers. The ANR Contrat doctoraux IA Etablissement programme would increase the research power needed to develop our growing expertise into an ongoing support program in the service of the four I-site challenges and emerging IA related research.

Information flow through cortical neurons can be decomposed into 3 main steps: 1) encoding (action potential (AP) emission), 2) propagation (travel of AP into the axon) and 3) transmission (synaptic relase). This process is determined by voltage-gated ion channels, myelination and synaptic release probability. Moreover, the neurons can display two type of coding regimen: a Firing Rate-based Coding (FRC) and an AP Waveform-based Coding (AWC, coding via modulation of AP shape). Interestingly, several lines of evidence suggest that regulations of ion channels, myelination and synaptic release probability promoting FRC lead to decrease of AWC and vice-versa. Therefore, we postulated the hypothesis of neuronal coding homeostasis: neurons can undergo a modification of their coding regimen but the amount of information flow is preserved. Using a transdisciplinary approach (electrophysiological recordings, mathematical information quantification and computational modelling), we propose to study this hypothesis by observing information flow modifications following excitability regulation of layer V cortical neurons in the context of facial neuropathic pain.