Global sensitivity analysis used to interpret biological experimental results

Area of Science:

• Computational Biology
• Immunology
• Infectious Disease Modeling

Background:

• Host/pathogen interactions are crucial for understanding immune defects and vaccine efficacy.
• Complex models of these interactions require extensive parameter exploration.
• Simplified models are needed to make these investigations tractable.

Purpose of the Study:

• To employ global sensitivity analysis to reduce the parameter space of a host/pathogen interaction model.
• To identify key parameters governing biofilm infections in mice.
• To create a more computationally efficient yet accurate model for studying osteomyelitis.

Main Methods:

• Global sensitivity analysis was applied to parameters of a mouse biofilm infection model.
• Insignificant parameters were identified and 'frozen' to create a reduced model.
• The reduced model's accuracy was validated against the full model.

Main Results:

• Sensitivity analysis successfully identified parameters that could be fixed without compromising model accuracy.
• A reduced model was developed, utilizing approximately half the original parameter space.
• The reduced model accurately replicated the full model's predictions for biofilm infections.

Conclusions:

• Global sensitivity analysis is an effective technique for simplifying complex biological models.
• The reduced model provides a more efficient tool for investigating immune responses in osteomyelitis.
• This approach can be generalized to other host/pathogen modeling scenarios.