Salinity stress tolerance prediction for biomass-related traits in maize (Zea mays L.) using genome-wide markers

Authors: SINGH, VISHAL - Utah State University; KRAUS, MARGRET - Utah State University; Sandhu, Devinder; SEKHON, RAJANDEEP - Clemson University; KAUNDAL, AMITA - Utah State University

Submitted to: The Plant Genome
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/14/2023
Publication Date: 9/4/2023
Citation: Singh, V., Kraus, M., Sandhu, D., Sekhon, R.S., Kaundal, A. 2023. Salinity stress tolerance prediction for biomass-related traits in maize (Zea mays L.) using genome-wide markers. The Plant Genome. 16(4),e20385. https://doi.org/10.1002/tpg2.20385.
DOI: https://doi.org/10.1002/tpg2.20385

Interpretive Summary: Maize is a vital crop used for food, feed, and fodder and is moderately sensitive to salinity. However, the growth, development, and production potential of maize are significantly impacted by salinity. Saline soils or irrigation water with high salt concentrations pose a significant threat to global maize production, particularly in arid and semi-arid regions. This research attempted to predict salt tolerance in maize by estimating breeding values for four traits related to plant biomass: shoot length, shoot weight, root length, and root weight. These traits were measured under both salt and control conditions. The study used a five-fold cross-validation method to select the best model from a set of options and looked at the effect of different marker densities on prediction accuracy. The results showed that a set of low-density single nucleotide polymorphisms (SNPs) had the best prediction accuracy for all traits and that GBLUP, rrBLUP, and Bayesian models had similar levels of prediction accuracy. These findings can help us understand how to use genomic selection tools to select salt-tolerant genotypes for a maize breeding program. These findings will be valuable to breeders and geneticists in developing new, salt-tolerant commercial cultivars that can thrive in saline environments.

Technical Abstract: Maize is one of the most important cereal crops after rice and wheat. Maize plant biomass is an important fodder-related trait affected by salinity stress. Selecting salt-tolerant genotypes is a cumbersome and time-consuming process that requires precise phenotyping. We predicted salt tolerance in maize by estimating breeding values for four biomass-related traits – shoot length, shoot weight, root length, root weight – measured under both salt and control conditions. A five-fold cross-validation method was used to select the best model among genomic best linear unbiased prediction (GBLUP), ridge-regression BLUP (rrBLUP), extended GBLUP (EGBLUP), Bayesian Lasso, Bayesian Ridge Regression, Bayes A, Bayes B & Bayes C. The effect of different marker densities on prediction accuracies was examined using a range of marker densities in our analysis. A set of low-density SNPs obtained through ANOVA and LD-based filtering showed the best prediction accuracies for all the traits. Average prediction accuracies in cross-validations ranged from 0.46 to 0.77 across the four derived traits. GBLUP, rrBLUP, and all Bayesian models except Bayes B demonstrated similar levels of prediction accuracy that was superior to the other modeling approaches. GBLUP prediction accuracies for test set validation ranged from 0.24 to 0.93 for different training population (TRP) - testing population (TSP) combinations across the traits. The findings of this research will help understand the applicability and optimization of genomic selection tools in selecting salt-tolerant genotypes for a maize breeding program.