CGIAR’s digital and breeding entities team up for smarter crop innovation

By Ram Dhulipala, Interim Director of CGIAR’s Digital Transformation Accelerator, and Peter Coaldrake, Interim Director of Breeding for Tomorrow 

CGIAR’s 2025–2030 Research Portfolio is designed to strengthen connections between science, partners, technology, and the people who rely on them. Within the Portfolio, Science Programs and Accelerators work hand in hand. An example of this integrated approach is the collaboration between the Digital Transformation Accelerator (DTA) and Breeding for Tomorrow (B4T). 

Together, these two entities harness the power of digital innovations to make crop breeding across the CGIAR and National Agricultural Research and Extension Systems (NARES) network more efficient, inclusive, and responsive. 

Why this collaboration matters

DTA supports CGIAR and its partners in leveraging the transformative technologies of our time – like artificial intelligence, big data, and machine learning – to solve agricultural challenges and advance CGIAR’s impact areas.  

B4T steers CGIAR’s work in crop breeding. It sets the standards, processes, and technologies used by breeding teams across CGIAR Centers and national breeding programs, helping ensure new varieties are developed more quickly and effectively to meet the needs of smallholder farmers. 

It’s only natural that DTA and B4T join forces. While DTA builds the digital foundation and strategy, B4T develops and implements the tools that breeders and researchers use on the ground, in the field and in the lab.  

Empowering crop breeders with data-driven solutions. Credit: CGIAR Breeding for Tomorrow.

Where breeding meets digital

B4T is already improving digital integration in breeding by standardizing how breeding data is managed and analysed with the Breeding Portal, promoting shared digital systems, like the Enterprise Breeding System (EBS), for breeding data management, Bioflow, CGIAR breeding analytics pipeline, and other cloud-based tools.  

However, challenges remain. Breeding teams need better solutions to support decision-making and integrate insights from other scientific areas – like nutrition and environmental data. At the same time, breeding schemes must be optimized on a continuous basis.  

B4T is also pushing for a more collaborative approach with national partners, to move from participatory variety selection to participatory breeding. This shift comes with the need for joint priority setting and decision-making, clearer goals and roles, better resource alignment and clear value proposition for partners.  

Digital tools are essential to making this collaborative model work.

How DTA x B4T collaboration works

Each program brings a complementary role to the table. DTA builds the enabling environment, thought infrastructure, data standards, governance and capacity building. B4T develops and rolls out tools and workflows that breeders can use day to day, within that environment.  

DTA will deliver on its role through its Data Ecosystem, Action Lab and Enabling Environments Area of Works (AOW). Through these AOWs, DTA will address issues on data governance, AI-ready datasets, technology strategy, benchmarking B4T tools and platforms to industry compliant standards and help B4T conceptualize and develop cutting edge AI based capabilities to accelerate breeding  

Meanwhile, B4T will keep building collaborative data management solutions for CGIAR and partners, promote the widespread use of these tools with an operational framework, and create a feedback loop so research and real-world application constantly inform each other.

Making it stick: training, trust, and responsible tech 

Technology alone isn’t enough. That’s why DTA will also lead capacity-building efforts to help researchers and partners adopt new tools and practices. This includes training workshops, cross-Centers hackathons and “curathon” events to clean and organize data and clear guidelines to ensure data is handled ethically and inclusively.  

CGIAR crop breeding data tool principles. Credit: B4T.

Projects to keep under your radar

DTA and B4T are already teaming up on exciting projects that aim to transform how data and digital tools support CGIAR and partners’ breeding programs. 

Fairground (previously AgPile) 

Backed by a grant from the Gates Foundation, DTA and B4T, through Breeding Resources, are building Fairgrounds, a federated, collaborative platform that brings together AI-ready, regionally relevant datasets for agricultural research.  

By making it easier to find, access and use diverse datasets across CGIAR and partner institutions, Fairgrounds saves time, efforts and money and opens the door to new applications of AI and machine learning for smallholder farming systems. 

Google Project 

The goal of this $7 million project funded by Google and the Gates Foundation is to develop a crop-agnostic, end-to-end phenotyping solution that improves data quality and offers breeding programs a scalable, cost-effective alternative to current tools. 

The project leverages Google’s powerful image machine learning models to support multi-crop phenotyping hubs in three regions. It is guided by real use cases from crop teams, who validate solutions even when implementation is outsourced to third parties, to increase adoption. 

To know more about these projects, stay tuned as the collaboration between DTA and B4T takes shape!

Resources:  

• Learn more about the Digital Transformation Accelerator and follow the Accelerator on social media 
• Learn more about Breeding for Tomorrow and follow the Science Program on social media 
• CGIAR secures a grant to develop AgPile, a federated platform for global agricultural data integration 

***

Main image: Sita Kumari, farmer, uses mobile phone apps to enhance her yields and get access to markets and labor. Credit: Photo by C. De Bode/CGIAR. Written by Julie Puech, CGIAR Breeding for Tomorrow. This work contributes to CGIAR Breeding for Tomorrow (B4T) Science Program and Digital Transformation Accelerator (DTA).