Course Description
This course is designed to equip participants with both the conceptual foundation and practical expertise required to apply Artificial Intelligence (AI) and machine learning-based approaches in intelligent plant breeding. Emphasizing hands-on learning, the training introduces participants to modern, breeder-oriented AI tools and analytical pipelines that enable data-driven decision-making across the breeding cycle.
Participants will gain extensive practical experience in robust phenotypic data analysis and visualization, linkage map construction, trait and QTL mapping, AI-guided gene prioritization, haplotype diversity assessment, and haplotype-phenotype association analysis using real breeding datasets. These modules are designed to strengthen participants’ ability to translate complex phenotypic and genomic data into biologically meaningful and breeding-relevant insights.
A key component of the course centers on machine learning–based genomic prediction for pre-breeding and early-generation selection, with a strong emphasis on elite × exotic cross prediction. Participants will learn how advanced AI and statistical learning models are applied to model additive and epistatic genetic effects, enabling accurate prediction of performance and genetic gain across diverse parental backgrounds.
The course covers haplotype-based prediction frameworks to evaluate favorable haplotype combinations and to quantify epistatic interactions influencing complex traits. Learners will gain hands-on experience in using machine learning models to identify optimal parental combinations, prioritize elite × exotic crosses with maximal expected genetic gain, and reduce linkage drag during introgression.
In addition, the course provides practical exposure to predicted haplotype combinations, supporting downstream molecular characterization, functional genomics, and data-driven breeding strategies.
Course Objectives
By the end of the training course, the participants should be able to:
- Gain knowledge in AI-enabled breeding pipelines
- Enable robust phenotypic data analysis, visualization, and quality control
- Develop skills in high-resolution trait mapping and AI-guided gene prioritization
- Apply haplotype-based diversity and association analyses for elite allele discovery
- Implement genomic prediction and optimal cross selection in breeding
- Support breeder-friendly, deployable AI enabled model solutions for precision breeding
Key Modules:
- Module 1: Introduction to foundations for intelligent breeding pipelines
- Breeding with speed of climate change
- Genomics assisted breeding
- Haplotype-based breeding in rice
- Introduction to AI-enabled Trait Mapping
- Next Gen Sequencing Platform for Genotyping (4K/8K)
- Core concepts tailored for Genomic prediction, optimal cross prediction
- Supervised, ensemble, and predictive learning approaches for breeding datasets
- Model interpretability and validation in breeding
- Module 2: Phenotypic analytics & visualization
- Robust phenotypic data curation and quality control
- Advanced visualization of trait distributions and correlations
- Multi-environment phenotypic analysis
- Module 3: Linkage mapping & trait dissection using Terratool
- High-density linkage map construction
- QTL and trait mapping
- Candidate gene prioritization integrated with genomic annotations
- Module 4. Haplotype-Based Diversity & Association Analysis
- Haplotype block construction and diversity assessment
- Haplotype–phenotype association analysis
- Identification of elite haplotypes for breeding applications
- Module 5: Genomic Selection for precision breeding
- Machine learning models for genomic prediction (GBLUP, RF, SVM, DL)
- Identification of optimal parental crosses with high expected genetic gain
Learning Modality
The course will be delivered primarily through face-to-face instruction, supported by a blended learning approach that combines synchronous and asynchronous sessions. The training will include theoretical lectures, hands-on practical exercises, interactive discussions, and self-paced e-learning activities, ensuring effective knowledge transfer and skill development across all modules.
Target Audience
This course is intended for plant breeders, geneticists, graduate students, postdoctoral fellows, research scholars, and research assistants engaged in crop improvement and breeding programs. It is particularly suited for participants interested in modern data-driven breeding approaches, including the application of artificial intelligence and machine learning models for phenotypic and genomic data analysis. The course will benefit those seeking to leverage large-scale datasets for genomic prediction, haplotype-based analysis, optimal cross selection, and implementation of genomic selection pipelines in practical breeding programs.




