Skip to main content
ARS Home » Midwest Area » Ames, Iowa » Plant Introduction Research » Research » Research Project #445206

Research Project: Germplasm Adaptation and Genetic Enhancement of Maize for Climate Resiliency in the United States

Location: Plant Introduction Research

Project Number: 5030-21000-074-000-D
Project Type: In-House Appropriated

Start Date: Feb 5, 2023
End Date: Feb 4, 2028

Objective 1: Develop and release new semi-exotic exotic maize lines with enhanced grain yield, improved disease resistance, and high stress tolerance traits (such as to heat and drought) for use by Germplasm Enhancement of Maize cooperators and other researchers. Sub-objective 1.A: Manage GEM field nurseries, germplasm exchange, and seed inventories to ensure that new sources of germplasm and information reach stakeholders annually. Sub-objective 1.B: Coordinate and manage in-kind support for evaluation, development, and genetic enhancement of GEM germplasm. Objective 2: Develop enhanced maize breeding strategies utilizing genomic evaluation and prediction to enhance exotic germplasm introgression of genes for higher grain yield, enhanced disease resistance, and improved environmental adaptation. Sub-objective 2.A: With public and private-sector cooperators, develop genomic prediction models for GEM line development populations and determine effectiveness of prediction models in comparison to conventional phenotypic selection methods for improved grain yield, disease resistance, and environmental adaptation. Sub-objective 2.B: Evaluate and compare genomic selection to phenotypic selection for successful preservation of exotic genome contributions and identification of unique alleles.

Objective 1: Four types of S0 breeding populations are used in the GEM Project, (1) 25% exotic donor populations, (2) crosses between two released GEM lines, (3) crosses between a GEM line and an ex-PVP, and (4) 12.5% exotic donor populations. Each year, 100-150 developed breeding crosses will be observed for flowering time, plant and ear height, standability, and overall appearance prior to advancing further into breeding protocol. Approximately 20-30 breeding cross populations with acceptable flowering time, plant and ear heights, standability, and overall appearance are selected and advanced to the S1 generation. Selection is practiced until the S2:3 generation when approximately 2,500 S2:3 topcrosses are evaluated for grain yield each year in a single replication across six Iowa locations. Top performing S2:3 entries are selected from field trials and advanced to second stage yield field tests. While undergoing yield testing, S2:3 lines are advanced to the S2:5 generation. The S2:3 lines with the best performance over two years of observations and testing will be coded for release as GEM lines, with the S2:5 seed used for distribution. GEM cooperators screen the S2:5 lines in natural and artificially inoculated trials conducted in areas favorable for disease development, high insect pressure, and extreme abiotic stresses. Data from in-kind support is returned to the GEM coordinator and prepared for distribution to other cooperators. Objective 2: Using the existing breeding protocol for GEM, ~1,500 of the available S2:3 and parental lines of the S0 founder population lines will be sent to a GEM cooperator to be genotyped using a proprietary genotyping system. The S2:3 lines will continue yield testing as outlined in Objective 1. These lines will serve as the training population to train the genomic prediction model. This approach will still greatly reduce the resources allocated to yield testing by reducing the number of S2:3 lines tested yearly from 2,500 to less than 1,000. To evaluate the impact of genomic selection on the contributions of exotic alleles, two populations, each comprised of eight released GEM lines representing the SS and NSS heterotic pools, will be formed. The double-double cross population will undergo the double haploid process and resulting lines will be genotyped. Doubled haploid lines will undergo yield testing and yield will serve as the trait for genomic prediction. With yield data, cross validation will be used to obtain prediction accuracies and predicted values. Selected lines will form two subgroups and the genomic regions of these groups will be compared.