Analysis of host-pathogen interactions: A bioinformatics approach

Over the past decade, there has been an explosion of information related to the protein-protein interactions within specific organisms. However, our understanding of the interactions between a pathogen and its host has lagged behind. This research uses ordinary differential equations and interaction data to model the progression of H. pylori infection, a major risk factor for gastric cancer. This model allows the study of proteins in the MAP Kinase and Apoptosis pathways and their change under bacterial stimulation. Microarray experiments performed on epithelial cells confirmed the time-course of the infection progression. These techniques were incorporated into the study of the HIV-1 viral infection that has a larger number of interactions, a handful of major interactions in H. pylori versus hundreds in HIV-1 infection. Interaction information gathered from human sequence motifs co-opted by HIV-1 for infection formed the basis of a predictor for the progression of the infection. Since HIV-1 easily develops resistance to multiple anti-retrovirals and high viral loads lead to increased mortality and morbidity, it is important to pick a drug that will inhibit viral replication. A logistic regression model trained with these sequence motifs as predictors has a high accuracy and specificity in choosing the correct drug when predicting historical samples. In an effort to expand these historical samples, which sequence only a small fragment of the HIV-1 genome, a viral linkage map was developed. This linkage map, similar to the human HAPMAP project, describes the evolutionary linkage across the HIV-1 genome. This allows researchers to gather the likely mutations on unsequenced regions of the virus and is the first step in a whole viral genome view of HIV-1 infection.