Human sarcomas are a heterogeneous group of over 50 different malignant tumors for which very few diagnostic markers currently exist. MicroRNA (miRNA) transcript levels have been proposed for use in the diagnosis, classification and prognosis of tumors. Over 700 miRNAs are identified in humans and miRNA are considered attractive candidates for developing novel biomarkers in sarcomas. However, miRNA expression patterns found in sarcomas are poorly understood and no central resource exists to contain this information. To systematically address the gap in both biological knowledge and bioinformatics infrastructure, we generated miRNA expression profiles for over 300 tumor tissue samples representing 22 different sarcoma types and developed a web-accessible database to enable facile access to the data. Sarcoma microRNA Expression Database (S-MED) is a repository that describes the patterns of miRNA expression in various human sarcoma types. S-MED provides both basic and advanced data search options for exploration of the data in heat map and text/numerical formats. The database also provides statistical details such as fold changes and P-values for differentially expressed miRNAs in each sarcoma type and corresponding normal tissue. Further, we have experimentally validated differentially expressed miRNAs in angiosarcoma and other sarcoma types. This comprehensive database is the first of its kind specifically devoted to miRNA expression patterns in sarcomas is available through the URL link http://www.oncomir.umn.edu/.