DSM

DSM (Distributed String Mining) is a software framework for content-based exploration and retrieval of whole-metagenome sequencing samples. It efficiently extracts informative sequence k-mers from a pool of metagenomic samples and uses them to measure the dissimilarity between two samples. The method is unsupervised but achieves significant enrichment for diseased gut samples in query results and high accuracy in discriminating between different body sites.