Skip to main content
ARS Home » Plains Area » College Station, Texas » Southern Plains Agricultural Research Center » Crop Germplasm Research » Research » Research Project #442118

Research Project: Omics-Assisted Breeding Approaches for Sorghum Hybrid Improvement

Location: Crop Germplasm Research

Project Number: 3091-21000-047-001-S
Project Type: Non-Assistance Cooperative Agreement

Start Date: Jun 30, 2022
End Date: Jun 29, 2027

The main objective of this cooperative agreement is to integrate genomic selection and haploid induction technology into traditional methods of sorghum genetic improvement. The ultimate goal of this cooperative agreement is to critically evaluate the impact of these modern tools to accelerate genetic gain over traditional breeding approaches and deciminate these omics-assisted breeding approaches to sorghum improvement programs across the U.S.

This cooperative agreement will implement rapid line derivation (through haploid induction or speed breeding) combined with genomic selection prediction models appropriate for applied sorghum breeding programs. A large number of hybrids will be generated and grain hybrid yield performance will be evaluated. Each parental line will be genotyped by sequencing (GBS) and genomic Bayesian models will be compared to classical combining ability models for predicting hybrid performance. To improve the accuracy of genomic prediction, models will be augmented with hybrid performance phenomics and enviromics datasets. Prediction model accuracy will be evaluated by a series of cross-validation schemes representing common challenges experienced by plant breeders. Once developed, omics-assisted methods will be decimated to ARS, university, and private sector sorghum improvement programs with interest but without the knowledge, to build genomic predictions models specific for their program. These efforts will coalesce in a new breeding model for sorghum that will demonstrate the efficiency in time, numbers, and cost when compared to traditional sorghum breeding methods.