Efficiency of multi-trait, indirect, and trait-assisted genomic selection for improvement of biomass sorghum

• DOI: 10.1007/s00122-017-3033-y
• Publication Date: 2017-12-07T16:46:49Z
• Authors: Samuel B. Fernandes; Kaio O. G. Dias; Daniel F. Ferreira; Patrick J. Brown
• Keywords: Trait; Plant biochemistry; Genomic Selection; Sweet sorghum

We compare genomic selection methods that use correlated traits to help predict biomass yield in sorghum, and find trait-assisted performs best. Genomic (GS) is usually performed on a single trait, but can also focal trait through indirect or multi-trait GS. In this study, we pre-breeding population of sorghum strategies improve prediction yield, the trait. Correlated include moisture, plant height measured at monthly intervals between planting harvest, area under growth progress curve. addition single- direct GS, test new strategy called which are used along with marker data validation Single-trait GS for had accuracy 0.40. Indirect best using curve 0.37, did not differ from moisture information. Multi-trait single-trait yielded similar results, indicating standard scenario. However, increased by up [Formula: see text] when both training populations population. Coincidence selected genotypes phenotypic was highest Overall, these results suggest be an efficient obtained earlier more inexpensively than