Research Projects

Below are my current and past research projects, which focus on applying machine learning and experimental approaches to better understand microbiome-driven mechanisms of human health and disease.

NCBI-AI: Foundation Model for Microbiome-Driven Mechanistic Discovery

This project focuses on developing a foundation model for microbiome data that integrates genomic, metabolic, and clinical information to enable mechanistic discovery. I work with large-scale sequencing datasets from the NCBI Sequence Read Archive to build deep learning models that move beyond prediction and toward generating and testing biological hypotheses. The broader goal is to create a closed-loop system that links data-driven modeling with experimental validation to identify disease-relevant microbial pathways.

Gut Microbiome and Cardiovascular Disease (HFpEF Mouse Model)

This project investigates how the gut microbiome contributes to cardiovascular disease using a heart failure mouse model. I designed and implemented the computational pipeline, from sample collection and sequencing to bioinformatics and statistical analysis. My work focuses on identifying microbial and functional shifts associated with disease progression and integrating these findings with computational models to better understand host–microbiome interactions.

Microwave-Based Sensing of Microbial Metabolites

This project explores a novel, non-invasive approach to measuring microbial metabolites using microwave sensing technology. I contributed to the computational analysis and applied machine learning methods to classify short-chain fatty acids based on their electrical and dielectric signatures. This work supports the development of real-time tools for monitoring gut microbial activity and has potential applications in clinical diagnostics.

Microbiome Dynamics in Wildlife Systems (Bat Model)

In this project, I study how microbiomes change over time in wildlife systems, with a focus on nocturnal bat species. Using metagenomic and metatranscriptomic data, I analyze temporal patterns in microbial composition and function. My work includes reconstructing microbial genomes to better understand how environmental and biological factors shape microbiome dynamics in natural systems.

MEEMA Study: Maternal–Fetal Microbiome Transfer

This project examines how the microbiome is transferred and shaped during pregnancy and early life. I analyze longitudinal microbiome data using bioinformatics pipelines to characterize changes across maternal and infant samples. The study also explores how interventions such as exercise and probiotic supplementation influence microbiome development and maternal–infant health outcomes.

Teaching as Research (TAR) Study Microbiology Education

This project applies a research-based approach to improving microbiology education. I designed a randomized controlled study to evaluate how different e-learning strategies impact student learning in laboratory settings. The goal is to develop more effective and equitable teaching methods for building practical scientific skills.