Biostatistics & Computational Biology Branch
- Matt Wheeler, Ph.D.
- Tel 984-287-3704
- P.O. Box 12233Mail Drop A3-03Durham, N.C. 27709
Matt Wheeler, Ph.D., is a staff scientist that develops statistical models for dose-response analysis and novel Bayesian non-parametric methodologies. His research includes methodological, applied and translational components. As a biostatistician, he develops novel methodologies from applied problems seen in a wide array environmental health data. The translational component flows from the concept of findability, accessibility, interoperability, and reusability (FAIR) in science. Here Dr. Wheeler’s researches and develops software platforms enabling other scientists’ easier access to his methods research. This also allows reproducibility of applied research problems.
- BMDS 3.x / ToxicR
This is the premiere dose-response software for dose-response analysis. Dr. Wheeler wrote the dichotomous and continuous module subroutines, which includes the Bayesian model averaging components. Currently ToxicR is under active development to include all of the functionality of BMDS in R as well as utilize common routines used by the National Toxicology Program.
This software is under active development and was used in “The COVID-19 pandemic vulnerability index (PVI) dashboard: monitoring county level vulnerability.” It builds upon the work of K Moran to approximate Gaussian Process regression within machine precision. Here all computations are near linear as opposed to the traditional cubic time complexity seen in Gaussian Process regression.
- Marvel SW, House JS, Wheeler MW, Song K, Zhou Y, Wright FA, Chiu WA, Rusyn I, Motsinger-Reif A, Reif DM. The COVID-19 pandemic vulnerability index (PVI) dashboard: monitoring county level vulnerability. Enironmental Health Perspectives, 2021.
- Wheeler MW, Westerhout J, Baumert JL, Remington BC. Bayesian stacked parametric survival with frailty components and interval censored failure times. Risk Analysis, 2020.
- Wheeler MW. Bayesian additive adaptive basis tensor product models for modeling high dimensional surfaces: An application to high-throughput toxicity testing. Biometrics, 75(1):193–201, 2019.
- Wheeler MW, Dunson D, Herring AH. Bayesian local extremum splines. Biometrika, 2017.
- Wheeler MW, Dunson DB, Pandalai SP, Baker BA, and Herring AH. Mechanistic hierarchical Gaussian processes. Journal of the American Statistical Association, 109:894–904, 2014.