Empowering data-driven breeding: Advanced biometrics, quantitative genetics, and digital tools training for WCA NARES partners

The Biometrics, Quantitative Genetics, and Data Management Training workshop for the Breeding Informatics Working Group A was held in Senegal from September 2-6, 2024, aiming to strengthen the data-driven capabilities of breeders and researchers. This training was led by ISRA (CERAAS/CNRA) together with the CIMMYT Drylands Crops Program, ADCIN Network, BRS-Digital Solution along with the financial support of ACCEL-GI/CapSha, CGIAR, ILCI-CIWA, and IAVAO-ABEE projects, with a mission to enable data-driven decision-making in breeding.

The workshop was planned and executed by Dr. Aissatou Sambou from CERAAS and Dr. Abhishek Rathore from CIMMYT, with significant contributions from resource persons including Dr. Roma Rani Das, Dr. Mohan Kumar Chejerla, Dr. Subhrajit Satpathy, Dr. Jack Faye, and Dr. Hermann Some. These experts shared their knowledge and guided participants in effectively applying advanced data analytical tools.

The workshop centered on several thematic areas essential to modern breeding programs, such as data-driven decision-making, advanced experimental designs, fixed and mixed effect models, genomic data utilization, quality assurance in marker-assisted selection (MAS), and the introduction of Bioflow, a transformative tool designed to streamline breeding data analytics. These themes collectively empowered participants with the skills to make data-informed breeding decisions, optimize genetic gains, and enhance the efficiency of breeding workflows.

Generous funding from projects ACCEL-GI/CapSha, BRS-Digital Solution, ILCI-CIWA and IAVAO-ABEE enabled the successful implementation of this impactful workshop, providing crucial resources and ensuring long-lasting benefits for WCA NARES partners.

Participants were researchers from various West African Countries, primarily breeders working on different crops mainly groundnut, sorghum and pearl millet. Twenty-eight (28) breeders from different NARS breeding programs (ISRA, ITRA, INERA, INRAN, NARI, IRAG, CSIR-SARI, IER and FOFIFA) among which eight (08) females participated in the training.

The workshop was structured around several thematic areas essential for advancing modern breeding programs. Each section provided participants with practical insights, hands-on exercises, and advanced analytical methods to elevate their breeding practices. By focusing on data-driven strategies, innovative experimental designs, and cutting-edge digital tools, the workshop equipped breeders with the skills needed to implement evidence-based improvements across their programs.

Quantitative Genetics Principles for Data-Driven Decision-Making in Breeding Programs

Data-driven decision-making was central to the workshop, with a focus on setting breeding program objectives and adopting digital breeding philosophy. Participants explored the role of quantitative genetic principles, such as the Breeder's Equation, in maximizing genetic gains and refining trait selection criteria. Integrating data management systems into breeding practices supports greater accuracy, shorter breeding cycles, and optimized selection processes, driving modern breeding toward a digital, evidence-based framework that enhances program efficiency.

Advanced Experimental Designs and Practical Field Applications

A variety of experimental designs crucial for efficient field trials were covered, including complete and incomplete block designs, augmented block designs, p-rep designs, and sparse designs. Through practical exercises, participants gained hands-on experience with these designs, learning how design decisions directly impact data quality, heritability, and trait selection accuracy. This theme strengthened participants' ability to make informed design choices that support robust, repeatable outcomes in breeding research.

Fixed and Mixed Effect Models and Genotype-by-Environment (GxE) Interactions

Training sessions on fixed and mixed models emphasized their application in multi-location trials (MLTs) and in understanding genotype-by-environment (GxE) interactions. Participants learned how environmental and genetic factors interact to influence trait performance, using mixed models to capture spatial variations across diverse environments. These skills are critical for breeding decisions that rely on stable, reliable phenotypic expressions in various environments.

Breeding Management System (BMS) and Enterprise Breeding System (EBS)

In-depth training on the Breeding Management System (BMS) and Enterprise Breeding System (EBS) provided participants with tools for managing breeding workflows effectively. These systems support germplasm and study management, genotyping, and data quality control (QC), enabling organized, traceable data management across breeding programs. The EBS demonstration highlighted its potential for future integration in breeding operations, underscoring its scalability and utility in streamlining workflows and centralizing data.

Genomic Data Utilization and Quality Assurance in Marker-Assisted Selection (MAS)

The workshop also covered the application of molecular markers for quality assurance (QA), quality control (QC), and marker-assisted selection (MAS). Sessions focused on data curation, marker validation, and strategies for effective plant tissue sampling. A detailed discussion was organized on interpreting results generated through QA/QC analysis. These practices enabled breeders to understand and implement markers within breeding pipelines, facilitating data-driven decisions.

R Programming for Advanced Data Quality Control and Visualization

Hands-on sessions in R programming introduced participants to advanced data quality control and visualization techniques. They practiced data summarization, tabulation, and visualizations using R's ggplot2 package, alongside model-based outlier detection methods. These skills enable participants to ensure data integrity and accuracy, supporting strategic decision-making through effective data monitoring and presentation.

Bioflow: The CGIAR Breeding Analytical Pipeline

The workshop dedicated significant time to training participants on Bioflow, also known as the CGIAR Breeding Analytical Pipeline, a central tool in breeding data analysis and decision-making. Bioflow's user-friendly interface and integrated data analysis capabilities make it an invaluable asset for modern breeding programs, supporting seamless data handling and visualization for quick, accurate insights. Bioflow simplifies workflows and provides real-time insights that enable breeders to make strategic decisions efficiently, accelerating the breeding cycle and enhancing breeding program outcomes. Its scalability and accessibility across breeding programs support broader data needs, promoting a data-driven approach adaptable to various research and breeding contexts.

Workshop Outcomes and Takeaways

The workshop provided participants with a strong foundation in breeding informatics, data management, and advanced analytical methods essential for achieving genetic gains and enhancing breeding efficiency. Participants gained hands-on experience with tools like Bioflow (CGIAR Breeding Analytical Pipeline), BMS, and EBS, fostering a collaborative environment that encouraged knowledge-sharing and practical applications. This thematic training empowered breeders with data-driven decision-making skills, contributing to improved breeding outcomes across Africa's NARES networks and setting the stage for continued advancements in digital breeding practices.

Comments from participants

Dr. Aissatou Sambou:"Organizing this workshop with Dr. Abhishek's team was incredibly fulfilling. Seeing breeders embrace data-driven methods with such enthusiasm was inspiring. Watching participants engage deeply with modern analytics tools showed the potential of these systems to reshape breeding programs. Streamlined data management and better decision-making were clear benefits. It was especially rewarding to see participants grow confident in using R programming for data quality control and visualization. The collaborative environment we fostered was invaluable. Everyone gained new knowledge and shared insights that will strengthen breeding efforts across WCA for years to come."

Dr. Mariam Coulibaly:"This workshop was transformative! The hands-on sessions with Bioflow and R programming made a big difference in how I approach data-driven breeding. Learning to use next generation data management system like EBS will surely streamline our data and work, and we feel confident applying these tools for our research trials. I feel equipped to make more data-driven decisions, which I know will have a direct impact on the success of our breeding programs. The collaborative environment also helped us all learn from each other, which was an invaluable experience."

Dr. Cyril Diatta:"This training was incredibly insightful, especially in terms of understanding experimental designs and applying genomic data for quality assurance in marker-assisted selection. The sessions on fixed and mixed effect models helped me see the value of advanced statistical methods in accounting for environmental variability, which is a common challenge in our region. I am excited to implement what I learned about Bioflow to support real-time decision-making. The trainers were highly knowledgeable and approachable, and I truly feel this experience has set the foundation for impactful, data-driven breeding in my program."

***

We also express our gratitude to the CGIAR research funders for their invaluable contributions to the CGIAR Trust Fund. Photo credits: CIMMYT.

Public link

Please use the above public link if you want to share this noodl on another website.