Sexual conflict is broadly defined as a conflict between the evolutionary interests of the two sexes. Depending on the genetic architecture of the traits involved, it can occur at the level of male-female interactions or take the form of selection acting to change the mean of a shared trait against the sign of its genetic correlation. The aim of my thesis was to use genome-wide expression profiles in the model organism Drosophila melanogaster to provide novel insights in the study of sexual conflict. First, we studied the female post-mating response to partition transcriptional changes associated with reproduction from male-induced effects, which are known to be harmful to females. We found substantial changes in expression of metabolic pathways associated with the activation of reproduction, while male-specific effects were dominated by the onset of an immune response. Changes in female response under different mating strategies was studied using experimental evolution: we found that monogamous females suffered decreased fecundity and their gene expression profiles suggested an overall weaker response to mating. To identify sexually antagonistic genes, we used hemiclonal lines and associated their sex-specific fitness with genome-wide transcript abundance. We confirmed the presence of a negative covariance for fitness and identified a group of candidate genes experiencing sexually antagonistic selection. We then focused on mitochondria, which can enable the accumulation of deleterious mutations with sex-specific effects due to their maternal inheritance, and found few effects on nuclear gene expression in females but major effects in males, predominantly in male-specific tissues. Finally, we used published data to compare intraspecific and interspecific genetic variation for a set of transcripts, to test whether speciation occurs along lines of maximum genetic variance. In conclusion, gene expression techniques can generate useful results in the study of sexual conflict, particularly in association with phenotypic data or when integrated with published datasets.