DMAP question guidance:
Getting Started section

The project will generate and use the following data types:

• Genomics data: High-throughput sequencing data (e.g., SNPs, whole genome sequencing) stored in VCF format for analysis and shared with the European Bioinformatics Institute (EBI) under their data access and usage policies.
• Phenotyping data: Field observations of growth stages, yield, and stress responses, collected using electronic field notebooks and stored in CSV format. – Proteomics data: Protein expression profiles associated with stress tolerance, stored in FASTA and HDF5 formats.
• Environmental data: Collected using sensors for soil moisture, temperature, and precipitation, stored in NetCDF format.
• Historical data: Sourced from national research institutions and global repositories (e.g., CIMMYT databases), integrated into analyses under appropriate data-sharing agreements.
• Metadata: AGROVOC-aligned metadata terms will accompany datasets to enhance interoperability and reuse. These datasets will adhere to FAIR principles and will be formatted and documented for long-term use.

The project will generate and use various data types, collected through diverse methodologies, data collections will be:

• Crop productivity measurements (e.g. yield data, plant growth stages) will be gathered through direct field observations by technicians.
• Environmental data (e.g. soil moisture, temperature, precipitation) will be collected using field sensors positioned in key crop areas.
• Germplasm information (e.g. variety, genetic traits) and experimental design data (e.g. plot layout, treatment applications) will be documented in standard data formats.
• Farmer input data on crop practices, planting, and harvesting dates will be collected through structured surveys and interviews with local farmers, both in person and digitally.
• Historical soil and weather data will be sourced from national meteorological and agricultural research institutions and integrated with project data to enhance analysis, serving as third-party sources that support data contextualisation.
• New crop classification reference terms will be developed, along with standardised metadata, to enhance interoperability across datasets.

These data types will be collected, formatted, and integrated according to established standards to support reproducibility and data reuse.

The ADGG project will collect and generate the following types of data:

• Genetic data on dairy cattle breeds, including SNP (Single Nucleotide Polymorphism) markers, for performance tracking and breed improvement.
• Milk production metrics (e.g., daily milk yield, quality parameters) from on-farm records and automated sensors.
• Animal health and reproductive data (e.g., vaccination records, fertility rates, calving intervals).
• Environmental observations (e.g., weather data, grazing conditions) from both in-field measurements and regional climate datasets.
• Farmer survey data on herd management practices and milk marketing (third-party data from local partners).
• Economic and market data on dairy production costs, revenues, and local milk pricing trends (third-party data from government agencies).

Data curation will ensure datasets are well-organised, interoperable, and reusable:

• Genomics and Proteomics Formats: Genomic datasets will use VCF and FASTA formats; proteomics data will use HDF5, both non-proprietary and widely supported.
• Phenotypic and Environmental Data: Collected in CSV and NetCDF formats for tabular and geospatial data compatibility.
• Metadata Documentation: Templates aligned with CGIAR and EBI standards will capture data types, units, and methodologies. Metadata will integrate AGROVOC terms to ensure interoperability.
• Curation Tools: Key tools include CGIAR DataScribe for metadata management, OpenRefine for data cleaning, and R/Python libraries for transformation.
• Preservation: Datasets will be securely stored in a CGIAR Datapool and registered in GARDIAN, with DOIs and descriptive metadata for long-term discoverability.

The project will implement robust data curation and quality assurance processes to ensure data consistency, reliability, and interoperability, covering formats, standards, metadata, and tools.

Data Formats: Multiple data formats will be used to ensure compatibility and preserve data quality:

• CSV for tabular data (e.g. household survey results, crop productivity, environmental observations) due to its non-proprietary nature and ease of use.
• JSON for structured survey data, facilitating interoperability across platforms.
• NetCDF for soil and weather data, supporting geospatial and temporal analysis.
• TIFF for image data to maintain quality and enable compatibility with various analysis tools.

Metadata and Documentation: Data will be documented using CGIAR metadata templates, providing comprehensive details on data type, collection methods, units, processing steps, and standards. Documentation will be supplied in Markdown and PDF formats, ensuring accessibility and alignment with domain standards like AGROVOC for crop terminology. Clear descriptions will be included to facilitate data reuse and understanding.

Interoperability Tools: A suite of tools will support data interoperability and quality:

• CGIAR DataScribe for cataloguing and metadata management, ensuring consistent documentation.
• GARDIAN for data publication and discovery, promoting data visibility.
• OpenRefine for data cleaning and transformation, enhancing data quality.
• R and Python libraries for analysis, enabling reproducibility and data sharing.
• GeoJSON tools for spatial data, ensuring compatibility across GIS platforms.

These structured processes and tools will collectively ensure that the project’s data meets high standards of quality, consistency, and interoperability for effective analysis and reuse.

To ensure data consistency, reliability, and interoperability, the project will employ comprehensive data curation and quality assurance processes covering formats, standards, metadata, and essential software tools:

• Data Formats: The project will utilise non-proprietary formats tailored to each data type for maximum compatibility and ease of sharing:
• VCF (Variant Call Format) for genetic data (e.g SNP data) and CSV for genetic summary tables.
• CSV and JSON for milk production data, with tabular data in CSV and structured sensor data in JSON.
• CSV for health and reproductive data, enabling integration with other farm records.
• NetCDF for environmental data, supporting robust geospatial analysis.
• CSV for survey and economic data, facilitating accessibility and compatibility across platforms.

Metadata and Documentation: Metadata for each dataset will adhere to CGIAR standards, ensuring consistent and clear documentation. Metadata templates will capture data collection methods, units, variable definitions, and processing steps. Documentation will be accessible in Markdown and PDF formats, with updates incorporated periodically to reflect changes in data collection protocols. AGROVOC terms will be applied where relevant to standardise terminology, enhancing interoperability across CGIAR datasets

Interoperability and Quality Assurance Tools: The project will employ a suite of tools designed to standardise data handling and maintain high-quality data practices:

• CGIAR DataScribe for structured metadata management, ensuring clear data cataloguing.
• GARDIAN for dataset publication and discovery, increasing dataset visibility within CGIAR.
• OpenRefine for initial data cleaning and standardisation of formats, promoting data quality.
• R and Python libraries for data cleaning, transformation, and reproducible analysis.
• QGIS for mapping and analysis of environmental and geospatial data.

Together, these structured processes and tools will ensure that data generated by the project meets rigorous standards of quality, consistency, and interoperability, aligning with CGIAR requirements and supporting data reuse across platforms.

All datasets will be registered on the GARDIAN platform, tagged with AGROVOC terms and assigned DOIs for discoverability. Genomic data will also be shared with the European Bioinformatics Institute (EBI) under their data-sharing policies, ensuring international visibility. Sensitive farmer data will be anonymised before sharing. All datasets will be licensed under Creative Commons Attribution 4.0 (CC BY 4.0), permitting reuse with proper attribution. Metadata will include detailed licensing information and reuse guidelines.

To ensure long-term discoverability, accessibility, and reuse, the project’s data will be meticulously documented, organised, and preserved, following CGIAR standards and protocols:

• Data Hosting and Accessibility: Data will be hosted on Dataverse and registered on the GARDIAN platform. Upon project completion, all datasets will be published openly with Digital Object Identifiers (DOIs) to guarantee persistent access. Data will be publicly accessible, with the exception of sensitive data (e.g. personal information from farmer surveys), which will be anonymised prior to sharing. If any data is restricted from open access due to privacy or sensitivity concerns, clear explanations will be included in metadata records to guide users on access limitations.
• Metadata Standards and Discoverability: Comprehensive metadata will be applied to all datasets, aligned with CGIAR standards to ensure clarity and consistency. Metadata records will incorporate AGROVOC terms and relevant keywords to enhance discoverability, while structured metadata fields will document data type, collection methods, units, and processing steps. Metadata will be periodically updated to reflect any project or protocol changes, ensuring it remains relevant and informative.
• Data Sharing Agreements and Compliance: The Foundation Data Access Clause will be included in all Data Sharing Agreements, providing a foundation for data governance and access that aligns with CGIAR’s FAIR principles. This clause ensures that data shared through the project remains accessible and usable by all relevant stakeholders, in accordance with institutional standards.

Through these structured approaches to documentation, metadata application, licensing, and compliance, the project’s data will be securely preserved, accessible, and primed for broad reuse across agricultural research platforms.

To ensure the long-term discoverability, accessibility, and reuse of project data, we will implement robust documentation, organisation, and preservation practices, aligned with CGIAR standards:

• Data Hosting and Access: Data will be openly accessible through a Public Data Marketplace and registered on the GARDIAN platform, where it will be tagged with AGROVOC terms to enhance searchability and ensure interoperability across agricultural research. Upon project completion, Digital Object Identifiers (DOIs) will be assigned to datasets to support reliable citation and long-term discoverability.
• Metadata Standards and Documentation: Metadata records will be enriched with relevant keywords, AGROVOC terms, and linked to associated publications to facilitate comprehensive discoverability. In cases where data cannot be made fully open (e.g., sensitive farmer data), metadata will clearly specify access restrictions and provide instructions for requesting access, ensuring transparency and controlled sharing where necessary.
• Licensing and Reuse: All accessible data will be licensed under Creative Commons Attribution 4.0 (CC BY 4.0), allowing reuse with appropriate attribution. Detailed licensing information and reuse guidelines will be provided within metadata and on each dataset’s GARDIAN page, guiding users on proper use and attribution.

By implementing these structured documentation and accessibility measures, the project will support the FAIR data principles, ensuring that data remains accessible, discoverable, and reusable by the agricultural research community and beyond.

• Informed consent will be obtained from participating farmers, detailing data use, confidentiality, and rights. Consent records will be securely stored, and all data collection will comply with CGIAR ethical guidelines.
• Genomic and proteomic datasets will be deposited with EBI under their usage entitlements.
• Data-sharing agreements with institutions such as CIMMYT and national agricultural research systems (NARS) will ensure compliance with FAIR principles and local regulations.
• An 18-month embargo period will apply to genomic datasets for publication and intellectual property protection, with public release after the embargo unless an extension is justified and approved.
• The Foundation Data Access Clause will be incorporated into all agreements, ensuring responsible data governance.

To uphold ethical standards, ensure compliance, and protect intellectual property while maximising research impact, the project will implement structured data access and sharing protocols:

• Embargo and Public Release: Certain genetic data will be subject to an 18-month embargo period to allow time for publication and intellectual property protection. Following this period, the data will be automatically released for public access to align with CGIAR’s FAIR principles and maximise research impact. Any extensions beyond 18 months will require formal justification, and each request will undergo a review process to ensure a valid reason for temporary data withholding.
• Ethical Considerations and Informed Consent: Written informed consent will be obtained from all participating farmers, clearly detailing data use, confidentiality, and participant rights. Consent records will be securely stored, and data collection processes will strictly adhere to ethical guidelines to protect participant privacy and maintain confidentiality.
• Data Sharing Agreements and Compliance: A data sharing agreement will be established with collaborating national agricultural research institutions and local universities. This agreement will restrict data use to research purposes, prohibit unauthorised redistribution, and ensure compliance with both CGIAR standards and local data policies, safeguarding data security and farmer privacy. The Foundation Data Access Clause will be incorporated into all data sharing agreements, guaranteeing that data access remains in line with CGIAR’s principles and supports the responsible and ethical dissemination of research data.

Together, these measures will enable the project to balance open data principles with necessary protections for research integrity, intellectual property, and participant privacy, ensuring compliance with ethical standards and CGIAR policies.

To ensure responsible data sharing while maximising future reuse, several factors have been carefully considered, including licensing, embargo periods, privacy constraints, and data-sharing agreements:

• Embargo Period and Public Release: An 18-month embargo period will apply to certain genetic data to allow for publication and protect intellectual property. This sunset clause ensures that after 18 months, data will be automatically released for public access, supporting FAIR principles and maximising research impact. Extensions beyond this period may be granted but will require formal justification and be subject to review, ensuring only valid reasons justify temporary data withholding.
• Informed Consent and Privacy Protections: To uphold ethical standards and privacy, informed consent will be obtained from all participating farmers. Consent forms will detail data use, confidentiality, and participants’ rights. Consent records will be securely stored, and all data collection will adhere to strict ethical guidelines to safeguard participant privacy. Sensitive data, such as personal information, will be anonymised before public sharing to protect participants’ identities while still supporting data reuse.
• Data Sharing Agreements and Compliance: A data-sharing agreement will be established with collaborating national agricultural research institutions and local universities. This agreement will limit data use to research purposes, prohibit unauthorised redistribution, and ensure alignment with CGIAR and local data policies, especially around data security and farmer privacy. Furthermore, the Foundation Data Access Clause will be included in all data-sharing agreements to guarantee responsible and ethical dissemination, safeguarding compliance with institutional standards.
• Licensing: Data will be released under the Creative Commons Attribution 4.0 (CC BY 4.0) licence, allowing broad reuse with attribution. This licensing approach promotes accessibility and collaboration while preserving proper recognition for data originators.

By addressing these factors, the project supports the reuse of data by collaborators, third parties, and other research programmes in a manner that balances openness with necessary protections for intellectual property, participant privacy, and compliance with CGIAR’s ethical standards.

The project will allocate 10% of the total budget to achieving FAIR data outcomes:

• 5% for data generation/acquisition, including genomic sequencing and phenotyping activities.
• 3% for data management, including metadata creation, documentation, and hosting on GARDIAN and support for 2 years post-delivery
• 2% for data processing and curation, including cleaning, transformation, and tool implementation.

The project will allocate 11% of the total budget to achieving FAIR data outcomes:

• Data generation / Acquisition: 5.5%
• Data processing, curation and reference data management: 2.5%
• Management including post-delivery hosting and support for 2 years: 3%

The project will allocate 10% of the total budget to achieving FAIR data outcomes:

• Data generation / Acquisition: 5%
• Data processing, curation and reference data management: 2%
• Management including post-delivery hosting and support for 2 years : 3%

Roles and responsibilities will be clearly outlined in the project documentation:

• The Data Manager will oversee data collection and storage protocols.
• Bioinformatics Specialists will process and analyse genomic and proteomic datasets.
• Field Technicians will collect phenotyping data using electronic field notebooks.
• The Project Coordinator will ensure alignment with CGIAR data policies and manage metadata creation.

Regular reviews will update roles and responsibilities, and metadata templates will ensure data is interoperable and aligned with FAIR standards.

The investment organisation has data-related roles assigned to individuals, and their responsibilities are clearly defined in documentation provided to support those roles, so that the investment delivers FAIR data outcomes for future agricultural research.

We will conduct regular reviews of the documented processes currently in place and update them to align with FAIR principles, so that we achieve our FAIR data commitments more efficiently.

We will adopt and document recommended tools for data management, data collection, and design, so that we meet FAIR commitments for associated metadata and align with organisational goals for FAIR data.

The programme will have libraries of metadata templates, supporting documents for processes and tools for data management, and documented role definitions, so that our data is interoperable with other research programmes.

Data management roles and responsibilities will be clearly outlined in project documentation:

• The Data Manager will oversee data collection protocols and compliance with FAIR principles.
• Genomic Analysts will handle SNP data processing and sharing with EBI.
• Veterinary Specialists will manage health and reproductive datasets.
• Field Technicians will collect on-farm production data using digital tools.
• The Project Coordinator will ensure metadata and data-sharing agreements align with CGIAR standards.

Regular reviews and updates to documentation will maintain clarity and adherence to FAIR principles.

Compliance will be monitored through:

• Regular audits to assess alignment with FAIR principles.
• Quarterly progress updates included in the Progress Narrative, tracking milestones and achievements.
• FAIR Data Assessment Tool to evaluate data maturity and identify improvement areas.
• Feedback from stakeholders, including NARS and EBI, to refine the DMAP. Training and updates will ensure the team aligns with DMAP requirements.

To ensure compliance with the Data Management and Access Plan (DMAP) throughout the Crop Productivity investment, we will implement the following monitoring measures:

• Regular Compliance Audits: Periodic audits will assess adherence to the DMAP, verifying that data management practices align with FAIR principles and project objectives. Audit findings will be documented, and any compliance issues will be addressed promptly.
• Progress Tracking and Reporting: Quarterly progress reports will track DMAP compliance, documenting data management activities, milestones achieved, and any challenges encountered. These updates will be included in the investment’s Progress Narrative, providing the donor with an ongoing view of data management efforts and improvements.
• FAIR Data Assessment Tool: At defined intervals, we will apply the FAIR Assessment Tool to evaluate the FAIR maturity of our data assets. This helps us identify and address areas for improvement to meet FAIR standards consistently.
• Ongoing Training and Support: Training sessions on FAIR data management practices will be held for the project team, ensuring all members are informed of and aligned with DMAP requirements. Any updates to data management policies or best practices will be communicated promptly.
• Stakeholder Feedback: We will regularly gather feedback from project stakeholders and data users, using their insights to refine our data management practices, keeping the DMAP relevant and effective throughout the investment.

These structured measures, combined with transparent updates through the Progress Narrative, will ensure strong compliance with the DMAP and that data generated meets CGIAR’s standards for accessibility, interoperability, and reusability.

Compliance will be monitored through:

• Regular audits to ensure adherence to FAIR principles and ILRI policies.
• Quarterly progress reports tracking milestones, integrated into the investment’s Progress Narrative.
• FAIR Data Assessment Tool to evaluate data maturity and improve practices.
• Stakeholder feedback, including input from farmers, NARS, and EBI collaborators, to refine the DMAP. Training and updates will ensure the team remains aligned with compliance requirements.