This presentation shows how systems biology can bring a more robust approach to understand host response to pathogen invasion. A computational systems biology method was employed to create in silico models derived from chickens in response to Campylobacter jejuni (C. jejuni) innoculation in the ceca. The comparative pathogenicity modeling profiles the cecal gene expression in response to wild-type and mutant strains of C. jejuni between two genetic lines of chickens (resistant line A and susceptible line B) employing a chicken whole genome microarray for a pair-comparison between inoculated (I) and non-inoculated (N) chickens within each line and between lines at multiple time points post infection. Computational methods based on Dynamic Bayesian Network (DBN) machine learning were employed to conduct pathogenicity analysis and to create the models. For the chicken host, 141 signaling and metabolic pathways and 3291 Gene Ontology (GO) categories were profiled which defined the host’s immune response biosignatures. Through this DBN computational approach, the method identified significantly perturbed pathways and GO category groups of genes and used as building blocks for constructing the system-level models. New algorithms for learning host response gene networks (>800 genes) from prior biological knowledge and trained with expression data produced the models. These models become a tool that can be visualized, interrogated, and employed in simulation to help understand and discover novel molecular mechanisms governing the host response.