Pathogen-plant host coevolutionary interactions exert strong natural selection on both organisms, specifically on the genes coding for effectors (pathogens), as well as on those coding for effector targets and R proteins (plant hosts). Natural selection leaves behind DNA sequence signatures on such genes and on linked genomic regions. These signatures can readily be detected by studying the patterns of intraspecies polymorphisms and interspecies divergence of the DNA sequences. Recent developments in DNA sequencing technology have made whole-genome studies on patterns of DNA polymorphisms : divergence possible. This type of analysis, called 'population genomics', appears to be powerful enough to identify novel effector-effector target genes. Here, we provide an overview of the statistical tools used for population genomics and their applications. This is followed by a brief review of evolutionary studies on plant genes involved in host-pathogen interactions. Finally we provide an example from our study on Magnaporthe oryzae.