Projects
Doctoral research in HLA immunogenetics and transplant informatics, prior work in pulmonary immunology, and the software, AI tools, and analytical workflows developed through this research and independent projects.
Research Projects
Doctoral research · Gragert Lab, Tulane University School of Medicine · 2023 – Present
Improving Kidney Transplant Outcomes by Using Molecular-Level HLA Matching in Allocation
NIH R01 DK139240 · MPIs: Michal A. Mankowski (NYU Langone Health) & Loren Gragert (Tulane University School of Medicine)
Developed HLA Matchability, a patient-specific metric of the likelihood of finding a well-matched donor, by analyzing more than 2.1 million OPTN HLA reference genotypes with Python on HPC (SLURM). The analysis revealed racial disparities in access to well-matched donors and informed a matchability-adjusted allocation framework. With collaborators in the Center for Surgical and Transplant Applied Research (CSTAR) at NYU Langone, we evaluated the framework in national organ-allocation simulations (OASIM), which showed improvements in both HLA matching and equity of access. HLA Matchability is publicly available as a calculator deployed on AWS EC2.
Probabilistic HLA Offer Filtering for Kidney Allocation
Dissertation research · PI: Loren Gragert
Developed a nine-locus, high-resolution HLA imputation pipeline on GPU-enabled HPC to resolve ambiguous deceased-donor molecular typing. Using a population-level haplotype frequency database derived from ~10 million high-resolution genotypes, the pipeline applies expectation-maximization and linkage-disequilibrium patterns to infer each donor's most likely high-resolution typing, with a probability for each possibility. Built on these imputed typings, a Probabilistic Offer Filter screened more than 1.2 million simulated kidney offers with 99.9% sensitivity and 99.5% specificity, reducing offer refusals driven solely by typing ambiguity.
Prior research · Kolls Lab, Tulane University School of Medicine · 2021 – 2023
LIF-Fc Therapy for Multidrug-Resistant Klebsiella pneumoniae Pneumonia
Graduate research · PI: Professor Jay Kolls
Murine pulmonary infection study testing LIF-Fc as a host-directed therapy against multidrug-resistant K. pneumoniae ST258 C4, benchmarked against IL-22-Fc. LIF-Fc reduced lung infection burden and induced acute-phase proteins (SAA1, SAA2), more strongly in liver than lung.
Isolation and Functional Characterization of Pulmonary ILC3s in Murine Models
Graduate research · PI: Professor Jay Kolls
Enrichment and sorting workflows for rare lung-resident ILC3s, used to define their IL-22 and IL-17A responses to pulmonary infection. A Ccr7-Cre × Icoslfl/fl model was generated to study ICOS-dependent regulation of lung ILC3s.
Web Applications & Research Software
HLA Matchability Calculator
Public Django web application on AWS EC2 — real-time individualized HLA matching-opportunity assessment against 2.17M OPTN nine-locus HLA reference genotypes; PDF / CSV / spreadsheet export. The first U.S. individualized HLA matching-opportunity metric, analogous to CPRA on the matching-likelihood side.
Transplant Toolbox
Modernized the Gragert Lab's suite of public-facing transplant immunology tools: upgraded the Django stack and dependencies, redesigned the frontend, and deployed updates to production (AWS EC2 via Bitnami).
Originally developed by the Gragert Lab, Tulane University School of Medicine.
Clinical Typing Parsers
Three production-grade ingestion pipelines that convert heterogeneous vendor-specific histocompatibility-laboratory typing outputs into structured GL Strings consumable by the imputation pipeline: PDF-to-structured-data extractor; RT-PCR SureTyper XML-to-GL-String parser; nine-locus LinkSeq XML-to-GL-String parser. Deployed by the UPenn / Kamoun group for multi-laboratory deceased-donor typing extraction.
AI-Powered Research Tools
Domain-specific retrieval-augmented generation (RAG) systems that connect curated scientific literature to large language models via MCP — enabling precise, citation-grounded querying with minimal hallucination.
HLA-RAG
Domain-specific retrieval-augmented generation system over ~3,500 curated HLA articles; exposed to LLMs via Model Context Protocol for citation-grounded scientific querying through any LLM (ChatGPT, Claude, Gemini).
ILC3-RAG
Retrieval-augmented generation system built on ~2,600 curated Q1 journal articles focused on ILC3 biology, lung infection, and immune response. Literature processed, tokenized, and embedded into a vector database, then connected to LLMs via MCP for rapid, comprehensive, hallucination-minimized literature access.
Independent Analyses
End-to-end analyses of public datasets, conducted independently of my dissertation and built as reproducible pipelines with tested code.
Mapping the Immune Landscape of Lung Cancer with Xenium Spatial Transcriptomics
End-to-end single-cell spatial analysis of a public 10x Genomics Xenium lung cancer dataset (~162,000 cells): immune exclusion, tertiary lymphoid structures, location-dependent macrophage states and what a pathology foundation model can see in the matched H&E, with the key results re-tested for technical artefacts.
Self-directed analysis of a public 10x Genomics dataset, conducted independently of my dissertation research.
Interactive Apps & Demos
Self-contained interactive web applications built to make technical concepts tangible.
Nextflow Workflow Management — Interactive Teaching Deck
A browser-based interactive deck that explains Nextflow by reading a real production pipeline (nf-core/hlatyping) rather than a toy example. Hand-built widgets include an animated DAG, channel-operator marble diagrams, an executor switcher, a -resume cache simulator, and a live DSL2 parameter editor.
GitHub Projects
Curated repositories from my GitHub profile.
GitHub projects will be auto-fetched at build time.
Add your pinned repositories to your GitHub profile.