Topological Data Analysis of Clostridioides difficile Infection and Fecal Microbiota Transplantation

Pavel Petrov,Stephen T Rush,Zhichun Zhai,Christine H Lee,Peter T Kim,Giseon Heo

DOI: https://doi.org/10.48550/arXiv.1707.08774

2017-07-31

Abstract:Computational topologists recently developed a method, called persistent homology to analyze data presented in terms of similarity or dissimilarity. Indeed, persistent homology studies the evolution of topological features in terms of a single index, and is able to capture higher order features beyond the usual clustering techniques. There are three descriptive statistics of persistent homology, namely barcode, persistence diagram and more recently, persistence landscape. Persistence landscape is useful for statistical inference as it belongs to a space of $p-$integrable functions, a separable Banach space. We apply tools in both computational topology and statistics to DNA sequences taken from Clostridioides difficile infected patients treated with an experimental fecal microbiota transplantation. Our statistical and topological data analysis are able to detect interesting patterns among patients and donors. It also provides visualization of DNA sequences in the form of clusters and loops.

Quantitative Methods,Applications