Trophic interactions, eg predation and herbivory, are one of the main biotic interactions driving the composition and functioning of terrestrial and aquatic food webs as they sustain energy and matter flows among organisms. The relationship between a resource availability and its consumption rate by a predator or herbivore, the functional response, is complex since often non-linear in shape and likely to depend upon the availability of other resources. Trophic interactions also nest the transmission of a large variety of multi-hosts parasites (ie trophically-transmitted parasites, TTP) within food webs. Moreover, the capacity to acquire, develop and transmit a parasite, called competence, vary among host species. The use of mathematical approaches, such as mechanistic dynamic models, to explore the transmission dynamics of TTP have often ignore the complexity of functional response and the variation in host competence. However, a prerequisite for understanding and modeling the ecology of TTP is to identify hosts species that are involved in transmission, ie the consumption of which ensures the parasite to step from the larval to the adult stage. This translates into the simple question “who is eating whom?”. Traditional techniques for investigating feces or stomach content based on macro- or microscopic identification of food remains fail to achieve fine taxonomic resolution identification of ingested items. Recently, DNA-based techniques such as metabarcoding have arisen as a powerful approach for the simultaneous taxonomic identification of several food items in feces or stomach content with high taxonomic resolution. We built a “one predator-two prey” eco-epidemiological model using a system of ordinary differential equations which allows for complex predation patterns. Using parameters related to Echinococcus multilocularis transmission in wildlife in Eastern France, we analyzed to what extent variations in prey diversity, total density and competence to transmit the parasite impact the basic reproductive number R0, which is a measure of disease outbreak risk. In parallel, we developed a molecular metabarcoding approach to investigate the diet of dog and red fox in rural settings of Eastern France, Gansu and Kyrgyzstan where E. multilocularis is endemic. Preliminary results, advantages and limitations, as well as perspectives of these recent developments will be presented.