After Helicobacter pylori was implicated in the development of peptic ulcers, other common diseases - such as asthma and atherosclerosis - were screened for known and unknown pathogens, but the available detection methods failed to reveal causative microorganisms. In the January 14 online issue of Nature Genetics, Griffin Weber and colleagues from Dana-Farber Cancer Institute, Boston, Massachusetts, USA describe a novel method to detect microbial causes behind infectious diseases by sequence transcript filtering against the human genome.

Weber et al. developed an in silico approach that employed the draft sequence of the human genome to identify nonhuman DNA sequences in expressed-sequence tag (EST) libraries of human origin. Using this method (called computational subtraction) they identified known viral pathogens such as HPV-18 in cervical carcinoma tissues (Nat Genet 2002, DOI: 10.1038/ng818).

"We therefore propose to generate, sequence and filter cDNA libraries from tissues of diseases such as systemic lupus erythematosus and extraintestinal Crohn disease, which are candidates for infectious etiology. [...] Further experiments could then distinguish sequences of benign commensal organisms from pathogen sequences by assessing the strength of their association with disease", concluded the authors.