Genomic selection (GS) is a revolutionary approach in plant breeding that leverages the power of genomics to accelerate the development of improved crop varieties. By using genome-wide marker data, breeders can predict the performance of plant varieties and make more informed selection decisions. This method contrasts with traditional breeding techniques that rely on phenotypic selection and are often time-consuming and resource-intensive. Below is a detailed step-by-step guide on the genomic selection protocol in plant breeding.

1. Population Development and Genotyping

  • Population Selection: The first step involves selecting a diverse and representative population for the breeding program. This population is usually a training population comprising individuals with known phenotypes for the trait of interest.
  • Genotyping: Once the population is selected, DNA is extracted from plant tissues, and genotyping is performed. High-density molecular markers, such as Single Nucleotide Polymorphisms (SNPs), are used to capture the genetic variation across the genome. Modern technologies like genotyping-by-sequencing (GBS) or SNP arrays are commonly employed for this purpose.
  • Marker Data Quality Control: Post-genotyping, it is crucial to ensure that the marker data is of high quality. This involves filtering markers based on criteria such as minor allele frequency, missing data, and linkage disequilibrium. Poor-quality markers are removed to ensure accurate predictions.

2. Phenotyping and Data Collection

  • Trait Measurement: Simultaneously with genotyping, the phenotypic data for the trait of interest is collected. Phenotyping should be conducted across multiple environments to capture the genetic by environment interaction (G×E) and improve the robustness of the genomic predictions.
  • Data Standardization: Phenotypic data needs to be standardized and cleaned to remove any outliers or errors. This step ensures that the data is consistent and reliable for building the prediction model.

3. Genomic Prediction Model Development

  • Model Selection: A crucial step in GS is selecting an appropriate statistical model to predict breeding values based on the marker data. Common models include Ridge Regression Best Linear Unbiased Prediction (RR-BLUP), Bayesian methods, and machine learning algorithms like Random Forest or Support Vector Machines.
  • Training the Model: The genomic prediction model is trained using the genotypic and phenotypic data from the training population. The model learns the relationship between markers and phenotypes, estimating the effect of each marker on the trait of interest.
  • Cross-Validation: To validate the model’s accuracy, cross-validation techniques are employed. This involves dividing the training population into subsets, where the model is trained on one subset and tested on the other. This process is repeated multiple times to assess the model’s predictive ability.

4. Genomic Estimated Breeding Values (GEBVs) Prediction

  • GEBV Calculation: After the model is trained and validated, it is used to calculate the Genomic Estimated Breeding Values (GEBVs) for individuals in the population. GEBVs represent the genetic merit of an individual based on the sum of the effects of all markers.
  • Selection of Candidates: Based on the GEBVs, the top-performing individuals are selected for the next breeding cycle. This selection is more accurate and faster than traditional phenotypic selection, as it does not require waiting for the plants to reach maturity.

5. Selection and Advancement

  • Marker-Assisted Selection (MAS): In some cases, specific markers associated with important traits are identified, and Marker-Assisted Selection (MAS) is used alongside GS to ensure that individuals with desirable alleles are selected.
  • Crossing and Advancement: The selected individuals are then crossed to create the next generation of plants. The progeny are genotyped and the GS process is repeated for multiple cycles until a superior variety is developed.

6. Validation and Field Testing

  • Validation of Selection: Before releasing a new variety, the selected individuals undergo extensive field testing to validate the predictions made by the GS model. These trials are conducted across multiple environments to ensure that the new variety performs well under different conditions.
  • Data Integration: The phenotypic data from these trials are integrated into the GS model to refine predictions and improve future breeding cycles.

7. Deployment and Commercialization

  • Variety Release: Once validated, the new variety is officially released for commercial cultivation. The breeding program then shifts focus to the next cycle, potentially incorporating new traits or improving existing ones.
  • Feedback Loop: After deployment, continuous monitoring and feedback from farmers and consumers are used to make further improvements. This feedback is essential for refining the breeding objectives and the GS model.

Advantages of Genomic Selection

Genomic selection offers several advantages over traditional breeding methods:

  • Increased Accuracy: By using genome-wide markers, GS captures the genetic architecture of complex traits more effectively than traditional marker-assisted selection.
  • Reduced Breeding Cycle Time: GS allows for early selection of individuals based on their GEBVs, reducing the time required to develop new varieties.
  • Cost-Effective: Although the initial setup of a GS program can be expensive, it becomes cost-effective in the long run due to the reduction in phenotyping costs and the accelerated breeding cycle.
  • Improved Genetic Gain: GS enhances the rate of genetic gain per unit time, leading to the faster development of superior varieties.

Challenges and Considerations

While GS holds great promise, it is not without challenges:

  • High Initial Costs: The setup cost for genotyping and model development can be prohibitive, especially for small breeding programs.
  • Need for Large Training Populations: The accuracy of genomic predictions depends on the size and diversity of the training population, which can be difficult to assemble.
  • Integration with Existing Breeding Programs: GS needs to be integrated with traditional breeding practices, requiring breeders to adapt to new methodologies.

Conclusion

Genomic selection represents a significant advancement in plant breeding, offering a more precise and efficient way to develop improved crop varieties. By following the steps outlined above, breeders can leverage genomic information to make better selection decisions and accelerate the breeding process. As technology continues to advance, GS will likely become an integral part of modern plant breeding, contributing to food security and agricultural sustainability.

References

  1. Meuwissen, T.H.E., Hayes, B.J., & Goddard, M.E. (2001). Prediction of Total Genetic Value Using Genome-Wide Dense Marker Maps. Genetics, 157(4), 1819-1829.
  2. Heffner, E.L., Sorrells, M.E., & Jannink, J.L. (2009). Genomic Selection for Crop Improvement. Crop Science, 49(1), 1-12.
  3. Crossa, J., Pérez, P., Hickey, J., Burgueño, J., Ornella, L., Cerón-Rojas, J., …& Babu, R. (2014). Genomic Prediction in CIMMYT Maize and Wheat Breeding Programs. Heredity, 112(1), 48-60.
  4. Bernardo, R. (2010). Breeding for Quantitative Traits in Plants. Stemma Press.
  5. Jannink, J.L., Lorenz, A.J., & Iwata, H. (2010). Genomic Selection in Plant Breeding: From Theory to Practice. Briefings in Functional Genomics, 9(2), 166-177.