The identification of suitable antigens is crucial to successful vaccine development based on subunit approaches. While many methods exist for the identification of vaccine candidates which are surface-exposed or secreted, immunogenic and conserved, contain B and T cell epitopes, most of these have a major drawback: they do not yield any information on whether the antigen is indeed expressed by the pathogen during infection. However, DNA microarrays offer a novel tool for the investigation of the transcriptional activity of all genes of a pathogenic microorganism under in vivo conditions. Employing whole genome DNA microarrays, we have analyzed the transcriptome of Neisseria meningitidis serogroup B bacteria during different stages of infection, i.e. exposure to human serum and the interaction with human epithelial and endothelial cells. Combined with data derived from genome-based approaches (such as reverse vaccinology) and immunogenicity studies, this novel transcriptome-based antigen identification should reveal ideal vaccine candidates against serogroup B meningococcal infection.