Association Studies in Eucalyptus

Abstract

Marker-assisted selection is attractive for tree breeding due to their long reproductive cycles and the time to expression of mature traits. In this chapter, we discuss different approaches used for developing markers in forest trees. Identification of useful markers for application in tree breeding is mainly based on two approaches, quantitative trait locus (QTL) mapping and association genetic studies. Though QTL mapping helped to determine the complexity of the genetic architecture underlying a phenotypic trait, the mode of action of specific genes is unclear as the genomic regions identified are usually large. Association studies have high resolution compared to QTL studies. While several association studies have identified significant markers, effect of the individual markers is low making it difficult to utilize them in breeding programs. Recently, genomic selection (GS) was proposed for overcoming some of these difficulties. In GS, high density markers are used for predicting phenotypes from genotypes. Currently small effective populations with high LD are being tested for GS in tree breeding. For wider application, GS needs to be applied in low LD populations which are found in many tree breeding programs. Significant markers from association studies may be used for developing prediction models in low LD populations using the same methods as in GS. GS based on large numbers of random markers or small numbers of associated markers is poised to make marker-assisted selection a reality in forest tree breeding.