France being the largest consumer of oysters in Europe, oyster farming is deeply rooted in French heritage. The Japanese oyster (Crassostrea gigas) is the oyster species the most exploited in France, and in the world. Due to filter-feeding, these bivalves are sensitive to toxic algal blooms. Although not always lethal, toxic algae can affect oyster physiology and make it unfit for human consumption. Phytoplankton toxins can be classified in several groups: amnesic, neurotoxic, diarrhetic and paralytic. For the latter group, saxitoxins are synthesized by the microalgae of genus Alexandrium that can accumulate bivalve tissues. In recent years, number of studies have been performed on the interactions between C. gigas and saxitoxins. In order to better understand these interactions, mathematical models have been developed but they did not allowed to describe accurately the kinetics of accumulation and detoxification of paralyzing toxins (PSTs). Models on DEB theory (Dynamic Energy Budget) (Kooijman, 2000) have been widely applied to the study of bivalve bio-energetics. This type of model already allowed to quantify growth and reproduction of C. gigas under different environmental forcing. These models have also been applied to study host-pathogen interactions (Flye-Sainte-Marie et al., 2009) and kinetics of accumulation and detoxification of contaminants (Bodiguel et al., 2009; Echinger et al., 2010). The aim of this PhD work is to develop a model based on DEB theory, that describes the interactions between PSTs and oysters. Indeed, different effects of PSTs contamination on oyster physiology have been shown. For example, paralytic toxins alter the immune response (overproduction and phagocytosis of hemocytes), behavior (modification of valve rhythms, production of pseudo-faeces) or organ integrity (myoatrophy, inflamed gills). In this PhD project, we will focus first on describing the accumulation and detoxification kinetics of PSTs in C. gigas and second on quantifying the effects of PSTs on the physiology of C. gigas.