Research

My research focuses on the development of statistical and computational methods for complex, high-dimensional data, with particular applications in genomics, bioinformatics, biostatistics, and biomedical research. A central theme of my work is the recovery of latent structure and biological heterogeneity from noisy, incomplete, or indirect measurements.

One major area of my research is Bayesian and Empirical Bayes inference for large-scale studies. In modern genomic and molecular applications, thousands or millions of features—such as genes, genetic variants, proteins, or other molecular measurements—may be analyzed simultaneously. I am interested in developing methods that can distinguish meaningful signals from noise while providing reliable measures of uncertainty. My work in this area includes posterior probability estimation, local false discovery rates, multiple testing, genomic association analysis, and statistical inference for high-dimensional data.

A second major theme is the development of mixture and latent-variable models for biological heterogeneity. Many biological samples contain mixtures of unobserved subpopulations, and the underlying composition must be inferred statistically. My research in this area includes cell-type decomposition and microbial strain mixture identification, where the goals are to determine the number of underlying components, estimate their relative abundances, and quantify uncertainty in the inferred structure.

I also work extensively with next-generation sequencing and genomic data, including problems involving variant detection, sequencing error, strain identification, and genomic variation. These applications often involve complex dependence structures, low signal-to-noise ratios, measurement error, and substantial computational demands, creating opportunities for both methodological and algorithmic development.

An important component of my research is translating statistical methodology into reproducible, scalable, and open-source computational tools. I have developed statistical software and analytical pipelines for local false discovery rate estimation, genomic variant analysis, cell composition estimation, and microbial strain mixture identification. Across these areas, my goal is to develop methods that are statistically principled, computationally practical, and useful to researchers working with real-world biological and biomedical data.