The Phylogenetic LASSO and the Microbiome: Metagenomic modeling in fecal microbiota transplantation
Year: 2017
Type: dissertation
Abstract: Scientific investigations that incorporate next generation sequencing involve analyses of high-dimensional data where the need to organize, collate and interpret the outcomes are pressingly important. Data of the microbiome can currently be collected, leading to possible advances in personalized medicine. In this thesis, we lay down a statistical framework for incorporating metagenomic information in predictive modeling with a view toward synthesis of products tailored to individual patients. In particular, we develop the phylogenetic LASSO ($\\Phi$-LASSO), a form of model regularization which incorporates known relationships between predictors for the purpose of model selection. We apply the $\\Phi$-LASSO to a pilot metagenomic study on the efficacy of fecal microbiota transplantation in the treatment of {\\it Clostridium difficile} infections. Although the thesis applies the technique to data for a particular infectious disease, the methodology is sufficiently rich to be expanded to other problems in medicine, especially those in which coincident `-omics' covariates and clinical responses are simultaneously captured.
Author: Stephen Rush
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2608568636