Infectious diseases transmitted by mosquitoes, such as dengue or chikungunya, are of worldwide concern. Within a mosquito population, the viral infection rate may be rather low and highly variable, and the causes of this variation remain unknown. All genomes contain genomic parasites called transposable elements (TEs), which resemble viruses, and also display variable amounts and activities. TEs are controlled by small RNAs of the piRNA class, and viruses are fought against by small RNAs of the siRNA class. Considering these similarities in structure and control pathways between TEs and viruses, we propose to use a Drosophila model to study the reciprocal impacts of TE control and antiviral immunity. Our preliminary results indicate that, upon infection, TE transcript amounts are modulated, as well as their controlling small RNAs. In this project, we will 1) identify the effects of viral infection on TE activity, and reciprocally, 2) identify the underlying mechanisms, and 3) assess the extent of the phenomenon in the frame of natural genetic variability. This will allow us to better understand variation in viral load, notably regarding arboviruses, and propose the use of TE activity as a marker for vectorial competence. In addition, if viral infections are associated with TE activation, we will have to consider viral epidemics as a new source of genetic diversification.