Drosophila melanogaster is a genetics star, but the phylogenetic relationships within the family Drosophilidae (~5.000 species) has long been nebulous. Molecular phylogenies challenged classification schemes proposed in the 1990s based on morphological data. The genus Drosophila, many of its species groups, as well as the family itself are not monophyletic natural clades. However, early morphological studies had the advantage of being taxonomically inclusive, which is difficult for molecular studies since DNA cannot be obtained from all taxa. Whereas phylogenomic studies are currently increasingly filling this taxonomic gap, proper understanding of morphological evolution remains essential to propose correct phylogenetic scheme. Phylogenomics, on the other hand, is the main source to place drosophilid taxa whose highly-specialized life has led to extreme morphological changes, such as the bee louse Braula coeca, an inquiline parasite in honey bee hives. Part of our work investigates the role of morphology in phylogenetic classification, whereas other parts focus on sources of incongruences (e.g., selection, introgression) in phylogenomic studies. 

In most sexually reproducing animals, genitalia are often the most rapidly evolving traits. This has led to the proposition of different sexual selection hypotheses explaining this general evolutionary trend. These hypotheses usually differ in their predictions concerning the degree of morphological coevolution between the sexes. For a very long time, it has been presumed that no morphological coevolution exists in Drosophila. We have shown, through thorough anatomical dissections, that this presumption was false, at least among the nine species of the D. melanogaster species subgroup. Male-specific grasping extensions usually co-evolved with female-specific seemingly anti-grasping grooves, and male-specific spines coevolved with female-specific shields. To better understand the genetic basis and evolutionary consequences of these trends, we have participated in multiple international efforts to unify genital terms between different biological disciplines and investigate through evo-devo research the homology relationships between rapidly evolving substructures. Currently, we are investigating the evolution of allometric relationship between genital and body sizes across multiple species, and its consequences in driving speciation through mechanical isolation. 

In most species of the subgenus Sophophora, a sexual dimorphism of abdominal coloration exists. Males have the last abdominal segments either monomorphically dark or light, whereas a Mendelian polymorphism has evolved in females of D. erecta (melanogaster group) and multiple species of the montium group. In these species females have two morphs with one morph usually mimicking males. We have shown that this sexual mimicry has distinct genetic basis, involving an enhancer of the melanin-synthesis gene t in D. erecta and a locus involving the transcription factor Pdm3 in montium species. Part of our current work investigates the impact of these genetic polymorphisms on behavioral variation and the potential role of pigmentation in driving reproductive isolation, whereas other parts studies how selective processes drive the evolution of the genetic architectures underlying the discrete polymorphisms. 

The breadth of the trophic niche evolves rapidly between specialist and generalist species and can play a major role in ecological speciation as well as in the emergence of invasive species. We have conducted several drosophilid collection missions in Africa, the cradle of the melanogaster species subgroup, in particular in the islands of the Western Indian Ocean. On the island of Mayotte, we have discovered a subspecies of the widespread and generalist D. yakuba which exclusively breed in the toxic fruit of noni (Morinda citrifolia). This case is reminiscent to the species D. sechellia, another noni-specialist in the Seychelles archipelago. We are currently conducting comparative population genomics studies to investigate the genetic basis of the behavioral and physiological parallelism between the two taxa towards noni toxins, and to date the time of their encounter with the noni fruit on their respective islands. We have previously investigated two tropical drosophilids of the genus Zaprionus, namely Z. indianus and Z. tuberculatus which have recently invaded temperate regions in Egypt. Whereas Z. indianus became a notorious pest of figs and has invaded the Americas during the last two decades, Z. tuberculatus is currently expanding its range in southern Europe (including France) and in South America. We are therefore currently investigating signals of parallelism in the genome of these two species and their behavioral adaptations.