The human gut microbiome is a source of great genetic and metabolic diversity. Varying composition of these microbial communities is known to be associated with several diseases. Exploring the relationship between biological/environmental covariates and the taxonomic composition of the gut microbial community can shed light on key aspects of the disease. A Dirichlet-multinomial (DM) regression model allows for a probabilistic investigation of the relationship between bacterial abundance and biological/environmental covariates. The presentation will focus on an application of these DM regression models to microbiome data of neonatal infants that are susceptible to Necrotizing Enterocolitis (NEC). The talk will conclude with a discussion of some on-going and future extensions of these DM regression models.