Abstract:
Motivation: The innate immune response system is the first line of host defense against infections in most multi-cellular organisms. This system employs a number of different types of cells which in turn activate different sets of genes required for the response. Microarray studies in human and mouse cells identified hundreds of genes that are activated or repressed in response to various pathogens. However, combining datasets from different species, different types of cells and different types of pathogens to identify common and unique response patterns remained a challenge.
Results: We developed a computational method that combines expression experiments across species, cells and pathogens to identify the core immune response genes. Our method utilizes probabilistic graphical models allowing homologous genes in different species to be analyzed concurrently. This allows us to overcome problems related to noise in microarray experiments and orthology assignments. Using our method we identified 57 homologous gene pairs that are always activated in response to infection in both species. We have also identified genes that are activated in macrophages in both human and mouse but not in dendritic cells, and vice versa. Our lists shed light on the mechanisms employed in innate immune response and raise new hypotheses for the roles of several genes in fighting infecting bacteria.
