Whole-Genome Genotyping of Gene Copy Number Variation (CNV) in Forest Trees

Trees present a remarkable adaptation to a variety of environments, which in part might be explained by the diversity in their genome. For instance, most flowering plants have undergone one or more event of whole-genome duplication. As a consequence, in a population of plants from the same species, genes may be present in different copies in the genome, referred to as copy number variation (CNV). It is not known, however, the extent to which CNV is responsible for changing the phenotype of the plant, particularly for complex traits. We are addressing this topic by developing a high-throughput, genome-wide, genotyping pipeline for CNV based on targeted-resequencing and, subsequently, genotyping a large association population of 579 unrelated trees. Sequence capture data for Pinus taeda, Pinus elliottii, Populus trichocarpa and Populus deltoides was generated for 24, 24, 48 and 579 individuals, respectively. Depending on the species, 14,729 to 34,647 target genes were captured from the genome using complementary probes and sequenced using next-generation sequencing, with median depth of sequencing varying from 10 to 52X. A mixed-model is being used to identify genes for which some individuals are consistently sequenced with higher or lower depth, characterized by CNV or present/absence variation. Different data normalization approaches are also being studied for better discrimination of false-positive CNVs. The optimized model will be used to genotype the P. deltoides population containing 579 individuals, with the objective of associating population level CNV with phenotypic variation. Population-level patterns of CNV for the species analyzed will also be reported.