Ethical, Equity, and Practical Considerations

This section prompts critical reflection on how study design, implementation, and dissemination intersect with broader ethical, social, and logistical concerns. Update this document as issues arise or are resolved.

Inclusion and Exclusion Biases

• Who is represented in the data, and who is left out?
  • Examples: Predominantly European ancestry populations in genomic data; exclusion of pediatric or elderly cohorts.
• What are the potential impacts of these biases on generalizability or fairness?
• Could decisions about inclusion criteria affect downstream interpretation or model behavior?

Open Science and Data Sovereignty

• Will code, data, and findings be openly shared? What are the licensing and access models?
• Are there any constraints due to IRB, data use agreements, or tribal/sovereign rights?
  • Examples: Respecting Indigenous Data Sovereignty principles (e.g., CARE vs. FAIR).
• How will authorship, credit, and contribution be transparently tracked?

Computational Constraints

• Are there limits in hardware, software, or pipeline reproducibility?
  • Examples: GPU access, reliance on proprietary tools, memory-intensive models.
• Are backup or low-resource versions of the analysis feasible?
• Could practical limitations introduce bias (e.g., truncating sample size, skipping QC steps)?

Potential Harms and Mitigations

• Could this work be misused or misinterpreted (e.g., in clinical settings)?
• Are there safeguards or contextual notes included in outputs?
• How will error, uncertainty, and scope limitations be communicated to end users?