Matt Koslovsky (PI) received NSF grant “Bayesian Sparse Dirichlet-Multinomial Models for Discovering Latent Structure in High-Dimensional Compositional Count Data” funded by the Division of Mathematical Sciences program “Computational and Data-Enabled Science and Engineering in Mathematical and Statistical Sciences.” Motivated by data generated by high-throughput sequencing technology in omics research, this research project will develop a novel sparse Dirichlet-multinomial (sDM) model that simultaneously accommodates potential zero inflation in multivariate compositional count data while estimating compositional probabilities. Extensions of the sDM modeling framework will be investigated for high-dimensional variable selection and clustering problems, and user-friendly software will be made publicly available. Graduate students will receive educational and professional training and mentoring.

For more details, click here.