In a multicenter study, we determined the expression profiles of 863 microRNAs by array analysis of 454 blood samples from human individuals with different cancers or noncancer diseases, and validated this 'miRNome' by quantitative real-time PCR. We detected consistently deregulated profiles for all tested diseases; pathway analysis confirmed disease association of the respective microRNAs. We observed significant correlations (P = 0.004) between the genomic location of disease-associated genetic variants and deregulated microRNAs.