Using Bayesian and Eigen approaches to study spatial genetic structure of Moroccan and Syrian durum wheat landraces

Abstract

The Mediterranean durum wheat landraces are genetically diverse and important sources for improving resistance to abiotic and biotic stresses and developing adapted and productive durum wheat varieties in the Mediterranean region. To study the diversity two distant countries (MoroccoandSyria) durum landraces were studied. Fifty-one microsatellites were used as molecular markers tool to determine the genetic structure and spatial adaptation of these landraces. We used two spatially-explicit methods (Bayesian and Eigen) to determine the genetic diversity and structure of a population composed of Moroccan (98) and Syrian (90) durum wheat landraces. Non-spatial methods were also applied for comparison. A significant genetic difference was detected between the landraces originated from Morocco and Syria. Six subpopulations were revealed for each country using the Bayesian method and the Eigenanalysis, which generated PC1 and sPC1, showed similar structure. Eigenanalysis exhibited a significant global genetic structure for both countries landraces; and showed that neighboring landraces tend to have close genetic profile. The two first axes of PC1 and sPC1 had discriminated four out of the six subpopulations revealed by the Bayesian methodology. Also, our study detected the close relationship between the durum landraces from the coastal areas of Syria and the Moroccan landraces from the Atlantic coastal regions where the Phoenicians/Carthaginians had settled in Morocco. These results demonstrate the importance of using the spatial models in genetic analysis of durum wheat landraces; and also recommend the use of the easily usable Eigenanalysis to analyze the genetic diversity and structure.