Identifying the sources of structural sensitivity in partially specified biological models

Area of Science:

• Mathematical Biology
• Systems Biology
• Ecological Modeling

Background:

• Biological systems possess inherent uncertainty and complexity, challenging precise mathematical description.
• Models can be structurally sensitive to the exact formulation of their functions, a property requiring advanced analysis beyond parameter sensitivity.

Purpose of the Study:

• To identify major sources of model uncertainty and less influential processes in biological models.
• To develop methods for quantifying structural sensitivity and reducing model complexity.

Main Methods:

• Introduced the gradient of structural sensitivity to quantify errors in specifying unknown functions.
• Defined the partial degree of sensitivity as a global measure of uncertainty from individual function variations.
• Proposed an iterative framework of experiments and analysis for heuristic reduction of structural sensitivity.

Main Results:

• Demonstrated the framework's application in a tritrophic food chain model.
• Successfully identified and quantified sources of structural sensitivity within the model.
• Provided a method to distinguish influential versus less influential model components.

Conclusions:

• The developed framework effectively identifies and quantifies structural sensitivity in biological models.
• This approach aids in understanding model behavior and reducing uncertainty by focusing on critical functions.
• The findings offer a pathway to more robust and reliable biological modeling.