Genomic selection: A method for accelerated improvement of oil palms

Oil palm is the most productive oil crop producing 37 % of total vegetable oil from just 18 million ha, or 6% of total agricultural land cultivated with oil crops (Oil World, 2016). Despite having yields greater than 12 MT/oil/ha/yr in breeding trials (Corley and Tinker, 2003), commercial yields have stagnated over the past 25 years at around 3-4 MT oil/ha/yr, on average. In contrast, dramatic yield increases have been realized commercially for soybean and corn through the use of biotechnology techniques. Sime Darby Plantation has invested in an extensive genomics research programme and has developed tools to facilitate the study of the genetic components of oil yield. The trait is complex and therefore suitably large populations are required for this study. For oil palm, this often poses a problem as only small populations are usually available, thereby restricting the trait associated information that can be gained. However, using large multi-parental populations, high-throughput SNP genotyping has enabled successful genetic association studies for both annual crops (Huang et al., 2010; Jia et al., 2013; Li et at., 2013) and oil palm (Teh et al., 2016). Genomic selection (GS), a genome-wide marker approach has been described as a superior method for genetic gain, especially with complex traits and has proved very successful in animal breeding. We have deployed genomic selection for oil palm to facilitate shortening of the breeding cycle through removing the need for progeny testing and dura phenotyping in the case of mother palm selection. Our paper describes the development of a SNP genotyping array, genome-wide association analysis (GWAS) for oil yield components, GS methodology followed by deployment for selective oil palm breeding.