Analysis of genetic isolation by distance (IBD) is of prime importance for the study of processes responsible for spatial population genetic structure and is thus frequently used in case studies. However, the identification of a significant IBD pattern does not necessarily imply the absence of sharp discontinuities in gene frequencies. Therefore, identifying barriers to gene flow and/or secondary contact between differentiated entities remains a major challenge in population biology. Geographical genetic structure of 41 populations (1080 individuals) of an alpine insect species, Carabus solieri, was studied using 10 microsatellite loci. All populations were significantly differentiated and spatially structured according to IBD over the entire range. However, clustering analyses clearly identified three main clusters of populations, which correspond to geographical entities. Whereas IBD also occurs within each cluster, population structure was different according to which group of populations was considered. The southernmost cluster corresponds to the most fragmented part of the range. Consistently, it was characterized by relatively high levels of differentiation associated with low genetic diversity, and the slope of the regression of genetic differentiation against geographical distances was threefold those of the two other clusters. Comparisons of within-cluster and between-cluster IBD patterns revealed barriers to gene flow. A comparison of the two approaches, IBD and clustering analyses, provided us with valuable information with which to infer the phylogeography of the species, and in particular to propose postglacial colonization routes from two potential refugia located in Italy and in southeastern France. Our study highlights strongly the possible confounding contribution of barriers to gene flow to IBD pattern and emphasizes the utility of the model-based clustering analysis to identify such barriers.

Genes of the Major Histocompatibility Complex (Mhc) play a fundamental role during the immune response because MHC molecules expressed on cell surface allow the recognition and presentation of antigenic peptides to T-lymphocytes. Although Mhc alleles have been found to correlate with pathogen resistance in several host-parasite systems, several studies have also reported associations between Mhc alleles and an accrued infection risk or an accelerated disease progression. The existence of these susceptibility alleles is puzzling, as the cost generated by the infection should rapidly eliminate them from the population. Here, we show that susceptibility alleles may be maintained in a population of house sparrows (Passer domesticus) if they have antagonistic effects on different malaria parasites. We found that one Mhc class I allele was associated with a 2.5-fold increase in the risk to be infected with a Plasmodium strain, but with a 6.4-fold reduction in the risk to harbour a Haemoproteus strain. We suggest that this antagonistic effect might arise because Mhc genes can alter the competitive interactions between malaria parasites within the host.