The Block Bootstrap Method for Longitudinal Microbiome Data

摘要: Microbial ecology serves as a foundation for wide range of scientific and biomedical studies. Rapidly-evolving high-throughput sequencing technology enables the comprehensive search microbial biomarkers using longitudinal experiments. Such experiments consist repeated biological observations from each subject over time are essential in accounting high between-subject within-subject variability. Unfortunately, many statistical tests based on parametric models rely correctly specifying temporal dependence structure which is unavailable most microbiome data. In this paper, we propose an extension nonparametric bootstrap method that inference these types The proposed moving block (MBB) accounts dependency by overlapping blocks within to draw valid inferences approximately pivotal statistics. Our simulation studies show increase power compared merge-by-subject (MBS) strategies. We also presume independent samples (PIS), our reduces false biomarker discovery rates. illustrated MBB three different pregnancy data oral provide open-source R package https URL make accessible study paper reproducible.