Course Overview
The Fundamentals of Genomic Predictions and Data-Driven Crop Breeding is tailored for individuals engaged in crop sciences, such as students, researchers, and practitioners. It aims to integrate knowledge in quantitative genetics, predictive breeding, and innovative breeding methods, including usefulness criteria and optimal contributions in applied crop breeding. Participants will be educated on utilizing R software for fundamental and sophisticated data analysis. The course will delve deeply into the essential principles of quantitative genetics and the application of genomic selection in crop breeding.
We will explore a comprehensive understanding of basic and advanced genomic prediction models. Detailed discussions will focus on analyzing genotype-by-environment (G x E) interactions through relationship matrices and the design of sparse testing. The course will also use real-world examples and interactive exercises to illustrate the practical application of key concepts such as usefulness criterion and optimal contributions in selecting progenitors for crossing and maintaining genetic diversity in breeding programs.
Throughout this course, actual data sets will be employed to showcase these genomic predictions/selections, examine G x E interactions, and perform other pertinent analyses. The course is organized into five distinct modules spread across five days. Each day will feature lectures on critical subjects that blend theoretical knowledge with practical exercises, enabling participants to apply these skills to actual challenges in crop breeding. Assignments will be given for home study and reviewed collectively on the following day.
Course Objectives
This course intends to blend quantitative genetics and statistical genomics with modern breeding approaches and data sets to perform genomic selections and make data-driven breeding decisions in crop breeding. Overall, the course aims to enhance the understanding of quantitative genetics and statistical models used in performing genomic predictions/selections, in breeding programs. Newer approaches to selecting parents and designing the crossing block will be covered using real datasets.
By the end of the training course, the participants should be able to:
- Master the R language to perform the visualizations and analysis.
- Gain the fundamental knowledge and understanding of quantitative genetic and statistical genomics applicable in crop breeding.;
- Understand the concept of genomic selection (3W’s of Genomic selection: Why, When, and Where to apply it in the breeding program);
- Understand the different relationship matrices and their use in genomic predictions;
- Understand the linear and mixed models to perform the genomic predictions/selections;
- Understand and apply the basic and advanced statistical models to perform the genomic predictions;
- Understand the G x E interactions and design sparse testing in dissecting G x E interactions and
- Understand and explore the usefulness criterion and optimal contributions to design the crossing block and predict the performance of cross combinations.
Target Learners
The course is intended for individuals with a basic understanding of crop breeding and genetics. While a basic understanding of statistics and proficiency in using the R language would be advantageous, the course is structured to accommodate participants with diverse levels of knowledge and expertise in crop breeding, statistics, and R programming.
Key Modules:
- Module 1: Introduction and Learning R and R Markdown
- Module 2: Fundamentals of Quantitative Genetics
- Module 3: Statistical Modelling in Predictive Genetics and Breeding
- Module 4: Implementation of Genomic Selection in Plant Breeding
- Module 5: Dissecting G x E Interactions and Crossing Strategies
Learning Modality
The course will be delivered face-to-face with a mixed modality of synchronous and asynchronous discussions involving theoretical concepts, practical and hands-on exercises, and self-paced e-learning activities.
Pre-requisites
The participant highly encourages a basic understanding of breeding and statistics. Experience in using and running R is highly encouraged, but it is optional.
