Methodological Assumptions

Clearly articulating methodological assumptions helps clarify the limits of inference and flags potential sources of bias or error. This section should be updated as new methods are added or revised.

Core Assumptions

• What assumptions are embedded in your statistical models, machine learning algorithms, or inference frameworks?
  • Examples: Linearity, independence, stationarity, uniform prior distribution, normality of residuals.

Domain-Specific Assumptions

• What biological or clinical assumptions are built into the design or interpretation?
  • Examples: Gene expression reflects protein activity, mutation burden correlates with phenotype severity.

Assumption Validation

• How will each assumption be tested, relaxed, or examined?
  • Examples: Residual diagnostics, simulation studies, cross-validation, external datasets.

Known Limitations

• Which assumptions are unlikely to hold and what consequences might they have for results?
• Are any assumptions unverifiable but necessary for tractability?

Revisions or Exceptions

• Log any changes to assumptions over time, along with justifications or supporting evidence.