Comparing Sequence-Capture and SNP-Array Genotyping Methods for Development of Genomic Selection Prediction Models in Eucalyptus

High-throughput and low-cost methods of genotyping are critical for the successful application of genomic selection (GS) approaches. Until recently, GS studies in trees have predominantly relied on SNP arrays as the source of genotypic data. However, this technology has a high entry cost that can prevent the use of GS in many tree breeding programs. Also, tree breeding programs typically employ locally adapted species and populations, hampering the worldwide transferability of a standard genotyping platform. Advances in next-generation sequencing and methods of genome complexity reduction, such as sequence-capture (a.k.a. exome capture), allow the development of flexible genotyping platforms that can be tailored to specific species and populations with a low entry cost. However, the suitability of sequence-capture approaches to genotyping and development of GS prediction models has not been evaluated, or compared to models developed from SNP arrays. Here we evaluate the impact of sequence-capture and array-based genotyping methodologies on the development of GS prediction models for a Eucalyptus breeding population from Fibria Celulose S.A., composed of 739 trees phenotyped for 13 wood quality and growth traits. The descriptive analysis of the datasets (pattern of linkage disequilibrium, minor allele frequency and missing data content) was used to compare both methods. RR-Blup and BayesB prediction methods were developed using data from both genotypic datasets, and predictive ability estimated using cross validation was employed to evaluate the performance of GS models. Differences in linkage disequilibrium patterns, minor allele frequency, missing data and marker distribution were detected among datasets derived from sequence-capture and SNP arrays. However, models RR-Blup and BayesB resulted in similar predictive abilities and demonstrated that both genotyping methodologies are equivalents for the traits evaluated.