Mapping Best Crop Varieties with Environmental Data Analysis

Jim Crocker
14th March, 2024

Mapping Best Crop Varieties with Environmental Data Analysis

Image Source: Natural Science News, 2024

Key Findings

  • Researchers in Brazil developed a new plant breeding method, GIS-FA, to predict plant traits in untested environments
  • GIS-FA combines mapping, statistical models, and enviromics to identify the best plant varieties for specific locations
  • The method helps breeders choose plants with high yield and stability, improving food security
In the realm of plant breeding, one of the most complex challenges is predicting how different plant varieties, or cultivars, will perform across various environments. This is crucial for selecting the best-performing plants that can thrive in specific conditions, which is becoming increasingly important as the climate changes and food security concerns grow. A recent study by researchers at the Federal University of Viçosa[1] has introduced an innovative approach that could transform how breeders make these predictions. This new method, known as the GIS-FA approach, combines geographic information systems (GIS), factor-analytic (FA) models, partial least squares (PLS) regression, and enviromics—a term referring to the study of environmental characteristics and their interaction with an organism's genetic makeup. The approach is designed to predict the phenotypic performance, which is the observable characteristics of a plant, such as height, yield, and disease resistance, in environments where the plant has not been tested. The study showcases the GIS-FA method using datasets from rice and soybean trials conducted across tropical areas. The method stands out for its ability to predict how genotypes, or the genetic constitution of the plants, interact with the environment to influence their performance. This capability is particularly significant given the complexity of genotype-by-environment interactions (GEI), which can make or break a plant's success in a given location. Earlier research emphasized the need for efficient prediction tools in plant breeding, especially those capable of handling large numbers of variables and high correlations among them[2]. The PLS method, which has been shown to be effective in predicting potato traits, is a part of the toolkit used in the GIS-FA approach. This method is adept at dealing with large datasets where the number of predictor variables far exceeds the number of observations, a common scenario in breeding programs. Furthermore, the GIS-FA method builds upon the concept of multi-environment trials (MET), where plant performance is evaluated across different locations and conditions. By integrating GIS techniques, the method allows for the creation of thematic maps. These maps are valuable tools for decision-makers, providing visual representations of where specific genotypes are predicted to perform best. The importance of selecting cultivars not only for high yield but also for yield stability has been underlined in previous studies, such as the one focusing on canola breeding programs[3]. The GIS-FA approach addresses this by using FA models to select the best-ranking genotypes based on their overall performance and stability. This is a step forward in achieving higher rates of genetic gain, a measure of how quickly beneficial traits are incorporated into plant populations over time. Another study[4] highlighted the advantages of the PLS method over the Bayesian Genomic Best Linear Unbiased Predictor (GBLUP) method in predicting plant performance in new environments. The current study extends this by incorporating PLS into a broader framework that includes enviromics and GIS, offering a more comprehensive solution to the challenge of predicting performance in untested environments. The GIS-FA approach is not just a theoretical advancement; it has practical implications for breeding programs worldwide. By identifying groups of environments where genotypes are likely to excel, breeders can make more informed decisions about where to deploy specific cultivars. Additionally, the thematic maps produced by this method can guide strategic planning and resource allocation, ultimately contributing to the goal of food and nutrition security on a global scale. In essence, the GIS-FA method developed by the Federal University of Viçosa represents a significant leap in the field of predictive breeding. By integrating various tools and disciplines, it offers a more nuanced understanding of how plants will perform in the face of diverse environmental challenges. For breeders, this means an enhanced ability to select and recommend cultivars that are not only high-yielding but also stable and adaptable to the changing conditions of our world.

BiotechPlant ScienceAgriculture

References

Main Study

1) GIS-FA: an approach to integrating thematic maps, factor-analytic, and envirotyping for cultivar targeting.

Published 12th March, 2024

https://doi.org/10.1007/s00122-024-04579-z


Related Studies

2) Partial least squares enhance multi-trait genomic prediction of potato cultivars in new environments.

https://doi.org/10.1038/s41598-023-37169-y


3) Optimal Contribution Selection Improves the Rate of Genetic Gain in Grain Yield and Yield Stability in Spring Canola in Australia and Canada.

https://doi.org/10.3390/plants12020383


4) Partial Least Squares Enhances Genomic Prediction of New Environments.

https://doi.org/10.3389/fgene.2022.920689



Related Articles

An unhandled error has occurred. Reload 🗙