Description du projet
Une plateforme de données intégrée pour accélérer le développement de médicaments et de produits chimiques plus performants et plus sûrs
Il est essentiel de caractériser la toxicité potentielle des produits chimiques et pharmaceutiques pour assurer la sécurité humaine et environnementale ainsi que la compétitivité industrielle. Les réglementations de plus en plus strictes et la volonté de réduire les expérimentations animales ont débouché sur d’énormes progrès au niveau des méthodes in vitro et in silico. Les modèles de voies d’effets indésirables, les tests moléculaires et la toxicogénomique font partie des nombreux nouveaux outils à la disposition des chercheurs. Le projet ARCHIMEDES, financé par l’UE, se focalise sur l’intégration de ces méthodes encore très hétérogènes au sein d’une plateforme de données unique. Grâce à l’IA et à la science des mégadonnées, la plateforme Toxicology Knowledge Graph du projet soutiendra le développement de médicaments et de produits chimiques plus performants, plus sûrs et plus durables.
Objectif
Traditional in vivo tesTraditional in vivo tests are hampering the development of new, safe and effective chemicals and drugs. If on one hand we need to ensure that dangerous chemicals do not emerge, on the other, we also need to promote rapid and sustainable innovation to successfully overcome the modern challenges of humankind. Toxicogenomics aims at clarifying the mechanism of action (MOA) of chemicals by using omics assays. The Adverse Outcome Pathways (AOP) concept is also emerging to contextualise toxicogenomics-derived MOA. Efforts are ongoing to anchor AOPs to molecular assays, but systematic embedding of AOP-derived in vitro tests and Integrated Approaches to Testing and Assessment (IATA) are still unestablished. At the same time, toxicogenomics-based evidence still struggles to gain regulatory acceptance. I aim to implement an integrated strategy based on state-of-the-art big data science, artificial intelligence (AI), toxicogenomics, molecular assays and cell technology via a novel Knowledge Graph approach. I will do so by developing the Toxicology Knowledge Graph (TKG), an innovative data platform where the currently fragmented knowledge in the field is going to be curated and integrated. The TKG will serve as a learning platform for artificial intelligence (AI) algorithms, which will be used to: 1) find new characteristics of chemicals/drugs; 2) infer associations between exposures and diseases; 3) select the most relevant cell lines to study specific phenotypes/chemical classes; 4) find the best genes to be used as reporters for specific AOPs; 5) define the applicability domain of computational, experimental and IATA models. I will also establish and validate regulatory-relevant high-throughput molecular assays to investigate the point of departure (PoD) of exposures. The ARCHIMEDES project will shift the paradigm of chemical and drug development, facilitating the emergence of new, smarter, greener, and more sustainable chemicals, drugs and materials.
Champ scientifique
- medical and health sciencesbasic medicinepharmacology and pharmacydrug discovery
- natural sciencescomputer and information sciencesartificial intelligence
- natural sciencescomputer and information sciencesknowledge engineering
- medical and health sciencesbasic medicinetoxicology
- medical and health sciencesmedical biotechnologycells technologies
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Régime de financement
HORIZON-AG - HORIZON Action Grant Budget-BasedInstitution d’accueil
33100 Tampere
Finlande