Multi-Resolution Sensitivity Analysis of Model of Immune Response to Helicobacter pylori Infection via Spatio-Temporal Metamodeling

Vida Abedi,Meghna Verma,Xi Chen,Guangrui Xie,Raquel Hontecillas,Josep Bassaganya-Riera,Wenjing Wang,Andrew Leber

doi:10.3389/fams.2019.00004

Abstract

Computational immunology studies the interactions between the components of the immune system that includes the interplay between regulatory and inflammatory elements. It provides a solid framework that aids the conversion of pre-clinical and clinical data into mathematical equations to enable modeling and in silico experimentation. The modeling-driven insights shed lights on some of the most pressing immunological questions and aid the design of fruitful validation experiments. A typical system of equations, mapping the interaction among various immunological entities and a pathogen, consists of a high-dimensional input parameter space that could drive the stochastic system outputs in unpredictable directions. In this paper, we perform spatio-temporal metamodel-based sensitivity analysis of immune response to Helicobacter pylori infection using the computational model developed by the ENteric Immune SImulator. We propose a two-stage procedure to obtain the estimates of the Sobol’ total and first-order indices for each input parameter, for quantifying their time-varying impacts on each output of interest. In particular, we fully reuse and exploit information from an existing simulated dataset, develop a novel sampling design for constructing the two-stage metamodels, and perform metamodel-based sensitivity analysis. The proposed procedure is scalable, easily interpretable, and adaptable to any multi-input multi-output complex systems of equations with a high-dimensional input parameter space.

Highlights

Computational immunology studies the interactions between various immunological elements, including proinflammatory and regulatory components in addition to the pathogen of interest
We develop a twostage metamodel-based Sensitivity analysis (SA) approach to quantify the temporal significance of each individual model parameter of large-scale agent-based modeling (ABM) and apply it to analyze the model of immune response to Helicobacter pylori infection
In our previous reported study we noted that the peak of importance is after week 2, using partial rank correlation coefficient (PRCC) we find that the peak of importance is toward the end of week 2 while using metamodeling approach, we note that the peak of importance ranges from end of week 2 to week 3, slowly decreasing after week 3

Summary

Introduction

Computational immunology studies the interactions between various immunological elements, including proinflammatory and regulatory components in addition to the pathogen of interest Understanding how these interactions affect the behavior of the complex stochastic system of interest can shed lights on some of the most fundamental questions in the field. Used metamodeling methods include splines, radial basis functions, support vector machines, neural networks, and Gaussian process regression (GPR) models, to name a few (see, e.g., Chapter 6 of [9]). The primary reason for GPR models’ popularity is that they unite sophisticated and consistent theoretical investigations with computational tractability These models enjoy desirable properties such as being highly flexible to capture various features exhibited by the data at hand and providing an uncertainty measure for the resulting prediction

Objectives

Results

Conclusion