Description: SparseDOSSA introduces a hierarchical model of microbial ecological population structure. It is capable of simulating realistic metagenomic data with known correlation structures and thus provides a gold standard to enable benchmarking of statistical metagenomics methods. SparseDOSSA’s model captures the marginal distribution of each microbial feature as a truncated, zero-inflated log-normal distribution, with parameters derived in turn from a parent log-normal distribution. The model can be effectively fit to reference microbial datasets in order to parameterize their microbes and communities, or to simulate synthetic datasets of similar population structure. Most importantly, it allows users to include both known feature-feature and feature-metadata correlation structures.