Fig 1.
Two-dimensional plot of the first two PCs derived from PCA on animals’ genomic relationships.
A total number of four clusters were created, one with genotypes from Lemnos (green) and Lesvos (brown) islands, a second with genotypes from Skyros island (red), a third with genotypes from Peloponnesus (dark grey) and Oreino breed of Epirus (blue) and finally a fourth with genotypes of Katsika breed of Epirus (blue). An unrooted phylogenetic tree depicting pairwise Reynold’s genetic distances between regions was also constructed (top). Individual assignment probabilities generated with ADMIXTURE (K = 2) are shown at the bottom of the graph. Each color represents a cluster, and the ratio of colored bars is proportional to the assignment probability of an individual to each cluster. Blue and red colors represent island and mainland membership, respectively. A possible partition of populations into mainland-island origin is indicated by the vertical blue colored line in the graph’s center.
Fig 2.
Plots of the mean nearest neighbor distances (A) and Gaussian kernel smoothed intensity (B) of animals’ genomic relationships along the geographical landscape. The plots indicate that animals sampled from the same location have higher genetic similarity than distantly placed or randomly selected samples.
Fig 3.
Box plots of LD scores for 500 kb and 1000 kb windows for mainland- island sheep populations.
Fig 4.
Manhattan plots of–log10(p-values) of SNPs (top), raw FST estimates (middle) and FLK statistic (bottom) across autosomes. Grey line on the upper Manhattan plot denote threshold (-log10(p-value) = 5.893) for genome wide significance, in the middle the 99.999th percentile bound (0.3554) and the threshold for the FLK statistic (bottom). The id of significant SNP is also presented. FLK plot was constructed with the CMplot package (https://github.com/YinLiLin/R-CMplot) in R (http://www.r-project.org/).
Table 1.
Significant SNPs obtained by allele frequency-based FST, FLK analysis and GWAS.
Fig 5.
(A) Loess curves of FST (orange line), within diversity (purple line) and between diversity (green line) for 1 Mb genomic distances around the FST outlier marker (s25289.1). Smoothed lines were obtained via LOESS procedure in SAS (ver.9.4) using a smoothing parameter equal to 0.21 for the three variables and 27 points (markers) (B) Positional candidate genes within 500 kb down- and up-stream the significant marker (s25289.1) are also shown here.
Table 2.
Positional candidate genes around the s25289.1 significant marker on chromosome 4.
Fig 6.
Three network representations of the total number of significantly enriched GO BP terms (nodes with khaki, light green and light blue color) and their respective member genes (red labels).
Figure was constructed via ClueGO V2.5.7 and CluePedia V1.5.7 plugins in Cytoscape.