Amy Willis
Assistant Professor
Department of Biostatistics
F-657, Box 357232
Health Sciences Building, 1959 NE Pacific St
University of Washington, Seattle, WA 98195
(206) 543 8027
Email: adwillis(at)uw(dot)edu (My CV)
News 07/25/2018: (Post-doc opportunity!) Thea Whitman and I are recruiting a post-doc to help us develop tools to improve reproducibility in microbiome science by integrating uncertainty in bioinformatics and sequencing into statistical analysis. The project, Addressing misclassification in the microbiome: A data-scientific approach to propagating uncertainty in microbial community composition, was selected for funding through the University of Wisconsin-Madison Data Science Initiative. Come join our team! Please forward your application materials to Thea and I by August 20 (details in the position description, and here).
I am a tenure-track Assistant Professor in the Department of Biostatistics at UW. I develop statistical methods for the analysis of ecological data obtained from high throughput sequencing, with a particular emphasis on microbiome data. Microbial communities are incredibly diverse, responsive, and critical to ecosystem function, and new microbiome data is being generated every day by researchers in biology, ecology, medicine, environmental health, agriculture, and many other fields. I develop rigorously grounded statistical and data-scientific tools for the analysis microbiome data that apply across multiple scientific disciplines. I actively develop and maintain my code on github, and engage with my users on Twitter.
From a statistical standpoint, I love quirky and non-standard statistical problems. Non-Euclidean metric spaces, boundary value problems, heavy-tailed estimators, and other situations that fall outside common regularity conditions fascinate me. Microbial data is high-dimensional, networked, often multi-scale, and suffers from differing levels of data quality, different resolution, and missingness -- that's why I think it's so fun to work with!
I am interested in problems such as utilizing microbial networks to improve inference and prediction, correcting for differential data quality in analysis, estimating the number of missing microbes and adjusting for (or imputing) them, and propagating phylogenetic uncertainty through downstream analyses. If you are a PhD student in Biostatistics or Statistics at UW and are interested in any of these topics (or another!), please drop by my office to introduce yourself!