05 September 2024, South Asia: In today’s rapidly evolving digital landscape, data has emerged as a cornerstone of innovation and growth, writes Akanksha Nagpal, CABI’s Project Coordinator, South Asia. With data and AI playing an increasingly pivotal role in advancing agriculture innovations, the adoption of FAIR (Findable, Accessible, Interoperable, Reusable) principles has never been more critical. Recognizing this, the Bill & Melinda Gates Foundation is spearheading a comprehensive AgDev-wide initiative to establish practical guidelines for the implementation of FAIR principles across its investments.
CABI is a part of this transformative journey, partnering with the Foundation to develop a set of detailed guidelines or “recipe books” intended to provide clear, actionable instructions for Program Officers (POs) and their grantees, ensuring responsible and effective data practices within grant-making processes. This effort will serve to enhance the FAIRness and responsible utilization of unique, high-value digital assets generated through investments.
Building a roadmap for better data management for CGIAR
The Gates Foundation’s initiative is driven by the recognition that setting clear guidelines and instructions for data management is crucial for maximizing the impact of their investments. As the Foundation approaches its next investment cycle, it aims to build on effective data practices across the CGIAR investments.
Some of the earlier successful initiatives like the “Excellence in Agronomy” (EiA) have demonstrated strong adherence to the FAIR principles, with a focus on developing robust data management processes and infrastructures that enable the creation of large agricultural databases, a key requirement for advancing agricultural science and policy. Taking learnings from such initiatives and working in collaboration with key data people of CGIAR, CABI with support from Knowmatics, is working towards actionable FAIR implementation processes for CGIAR investments.
Discovery phase: Key insights from the team
The work started with the discovery phase which involved engagement with data managers and practitioners of CGIAR across different investments encompassing crop breeding, livestock and agronomy, who play a critical role in upholding robust data practices. Our objective was to learn from their experiences, understand both the successful implementations and the challenges they face in adhering to FAIR principles within CGIAR, and use these insights to create practical, user-friendly guidelines for the community.
These insights encompass not just the technological aspects but also cultural, resourcing, support, and policy-related issues that impact the effective implementation of FAIR practices.
1)Support: The need for strong leadership and adequate funding
A recurring theme in our discussions is the insufficient support for metadata and ontology development. Inadequate funding and leadership often lead to inefficient practices, such as repetitive data cleaning, which could be avoided through early standardization. Effective implementation of FAIR principles requires strong leadership and adequate funding to drive these initiatives forward.
2) Technology: Overcoming the challenges of data integration
Technological issues also pose significant challenges. For example, flexible trial designs can lead to varying terminologies that complicate meta-analysis and require extensive data cleaning. Data often arrives in formats that are difficult to integrate into analysis pipelines, necessitating manual or scripted standardization efforts. While automated standardization is ideal, it has not yet been fully realized. Furthermore, the presence of existing datasets, some over a decade old, adds another layer of complexity to achieving consistent standardization.
3) Culture: Bridging the gaps in data management practices
Cultural challenges also play a crucial role in data management. We have observed that teams often shift blame, with data managers pointing fingers at tool designers for poor metadata, while data analysts alter metadata to suit their needs, further exacerbating the issues. The lack of integration among teams handling experimental design, data collection, and data analysis can undermine the overall effectiveness of data management efforts. Moreover, problems that arise during the initial data collection phase can compromise the FAIRness of the data, highlighting a cultural challenge in prioritizing data quality from the start.
4) Policy: Aligning data management practices with practical needs
Policy-related challenges also need to be addressed. Policies across different CGIAR institutes for data management may not always align with the practical needs or be well-informed by the researchers and data practitioners responsible for achieving investment outcomes within their specific contexts. There is a clear need for policies that are not only well-defined but also flexible enough to adapt to the varying needs of different research environments.
5) Resourcing: Integrating data perspectives in proposal development
Finally, resourcing is a critical issue. Adequate resource planning from a data perspective should be integrated during the proposal development stages. For instance, limited skill sets for metadata and ontology development often lead to inefficient practices. Ensuring that sufficient resources are allocated to data management from the outset is crucial to achieving FAIR principles.
A shared commitment to transforming data management
CABI is working towards providing practical guidelines through well-defined “recipes” tailored to the CGIAR context. These guidelines will help funders and grantees navigate the complexities of implementing FAIR principles in their projects. We aim to address the key challenges of data management, including technological, cultural, resourcing, support, and policy-related issues. Our shared goal is to enhance the impact of CGIAR investments, particularly in the GenAI era, where AI-ready FAIR data is essential to unlocking its full potential.
As we continued in the work after the discovery phase, we remained mindful of the words shared by one of the Program Officers: “There is a technical solution, but the human side of it is important to understand for the proper uptake of the technologies.” By balancing both the technical and human dimensions of data management, we believe we can unlock the full potential of FAIR principles and create an actionable approach for CGIAR investments with respect to data management.
The figure below illustrates the framework created for implementing FAIR data practices within grant investments, from initial investment to execution. It begins with a FAIR Potential Assessment to evaluate the data’s potential to meet FAIR standards, followed by the creation of a Data Management & Access Plan (DMAP) that guides target setting and strategy development for data management.
Throughout this process, continuous monitoring via a FAIR Data Assessment ensures adherence to FAIR principles. The entire framework is guided by the FAIR Process Framework, ensuring a structured approach to data management that maximizes the value and usability of data generated from grants. The framework emphasizes both informed decision-making and ongoing monitoring to achieve effective FAIR data practices.
Project page
Find out more about how CABI is working to address constraints in realizing the value of data as part of the project ‘Enabling FAIR data sharing and responsible data use.’
(For Latest Agriculture News & Updates, follow Krishak Jagat on Google News) (+80 Million Farming Audience Visits Krishak Jagat’s Hindi Website – Click Here for Website)
