Hi there! I'm Michael Zietz. I'm a data scientist and genetics researcher, focused on developing new statistical and machine learnings methods with applications in healthcare and biomedicine. ๐งฌ I'm currently a Research Data Scientist at Cedars-Sinai Computational Biomedicine. Previously, I did my PhD at Columbia University DBMI and studied Physics at Penn, where I did research on heterogeneous networks in the Greene Lab. ๐ธ๏ธ I'm interested in methods development, reproducible research, and accelerating the pace of scientific progress on complex diseases. ๐ฅ
WebGWAS: Instant, free, online genome-wide association studies (GWAS)
- Novel statistical method: Indirect GWAS gives an efficient approximation (code, analysis, preprint)
- Backend web server: Computes GWAS in ~3 seconds, Rust Axum backend (code)
- Frontend application: Interactive cohort builders and data visualization, React, Typescript (code)
COVID Blood type: Study of the relationship between ABO type and COVID-19 (analysis, paper, New York Times)
- sumher_rs: Efficiently estimating a genetic covariance matrix using SumHer (code)
- mdav: Data anonymization tool implementing maximum distance to the average vector (MDAV) anonymization (code)
- pymbend: Bending matrices to be positive semi-definite (code)
- Micromanubot: A user-friendly build tool for academic preprints in LaTeX (code)
- OnsidesDB.org: Interactive exploration of adverse drug events extracted using large language models (code)