Partnership
Work With Us
Join us in building representative genomics resources for Africa through research collaboration, dataset co-design, and capacity building.
Audience
Who This Is For
Researchers & Method Developers
Teams working on machine learning, statistics, and bioinformatics.
- ML and statistics researchers
- Bioinformatics teams
- QC and normalization developers
- Variant calling pipeline builders
African Health Institutions
Cancer centres, universities, and public health agencies across the continent.
- Cancer centres and registries
- Universities and research groups
- National public health agencies
- Regional health organizations
Capacity Builders & Educators
Training programmes that need realistic but non-identifiable data.
- Bioinformatics courses
- Data science bootcamps
- Workshop and hackathon organizers
- Curriculum developers
Opportunities
Ways to Collaborate
Method Development & Benchmarking
Use synthetic datasets as public benchmarks and contribute results back to the community.
- Survival and risk prediction models
- Fairness-aware ML methods
- Missing data handling and imputation
- Batch correction and harmonisation
- Share methods, metrics, and code
Dataset Co-design
Help design new modules that reflect local realities and refine existing parameters.
- Population structure modeling
- Care pathways and resource constraints
- Local risk factors and treatments
- Stage distributions and survival curves
- Aggregated statistics from registries
Teaching & Capacity Building
Integrate datasets into educational programmes and share materials with others.
- Bioinformatics curricula
- Short courses and bootcamps
- Hackathons and code sprints
- Slides, notebooks, exercises
- Reusable teaching bundles
Values
Guiding Principles
African Data Sovereignty
Synthetic datasets are not a substitute for real African data governance. Wherever real data is involved, it must be under local governance, ethical oversight, and appropriate data use agreements.
Transparency
We keep configurations, assumptions, and validation checks as transparent as possible. Collaborators are encouraged to document assumptions clearly and make code available where feasible.
Mutual Capacity Building
Collaborations should support skills transfer and mutual learning. Synthetic datasets can train early-career researchers and test infrastructure before working with real data.
Examples
Collaboration Scenarios
Scenario 1
Cancer Centre Piloting Pipelines
An African cancer centre uses synthetic genomics datasets to prototype analysis pipelines without real patient data.
- Test alignment and variant calling
- Stress-test storage and compute
- Validate governance workflows
- Transition to real data with ethics approval
Scenario 2
Building Fairness-aware Models
A method developer uses integrated ML and SNP datasets to build and evaluate models across ancestries.
- Train models across populations
- Quantify performance differences
- Assess calibration by ancestry
- Contribute open-source code and results
Scenario 3
Designing a Training Course
An educator assembles a teaching bundle from multiple modules for a bioinformatics course.
- Combine methylation, expression, SNP data
- Design differential expression exercises
- Build prognostic model assignments
- Teach batch effects and missing data
Contact
Get in Touch
General Enquiries
Questions about datasets, methodology, or the project in general.
[email protected]Collaboration Proposals
Send a short concept note outlining your institution, intended use, and desired outputs.
Submit Proposal →Feedback & Issues
Report issues, inconsistencies, or suggest new validation checks and parameters.
Send Feedback →Ready to collaborate?
We're actively looking for partners across Africa and globally to help build representative genomics resources.
Start a Conversation