Hypothesis Testing and Power Calculations for Taxonomic-Based Human Microbiome Data

By Patricio S. La Rosa, J. Paul Brooks, E. Deych, Edward L. Boone, David J. Edwards, Qin Wang, Erica Sodergren, George M. Weinstock, William D. Shannon, Ethan P. White. PLoS ONE, 2012

315 citations recorded by the source corpus.

This paper presents new biostatistical methods for the analysis of microbiome data based on a fully parametric approach using all the data. The Dirichlet-multinomial distribution allows the analyst to calculate power and sample sizes for experimental design, perform tests of hypotheses (e.g., compare microbiomes across groups), and to estimate parameters describing microbiome properties. The use of a fully parametric model for these data has the benefit over alternative non-parametric approaches such as bootstrapping and permutation testing, in that this model is able to retain more information contained in the data. This paper details the statistical approaches for several tests of hypothesis and power/sample size calculations, and applies them for illustration to taxonomic abundance distribution and rank abundance distribution data using HMP Jumpstart data on 24 subjects for saliva, subgingival, and supragingival samples. Software for running these analyses is available.

Subfields: Molecular Biology, Physiology

View Paper: https://www.semanticscholar.org/paper/475a19f5869411b51abb8b89ad08aec586431c6e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

In Physiology

Citations in the corpus

Marine Vertebrate Biodiversity and Distribution Within the Central California Current Using Environmental DNA (eDNA) Metabarcoding and Ecosystem Surveys (2019, methodology): “Future efforts to develop such power analysis techniques could build off of those completed for PERMANOVA (Kelly et al., 2015) or Dirichlet Multinomial (La Rosa et al., 2012) methods (Knight et al., 2018).”
ANOVA-Like Differential Expression (ALDEx) Analysis for Mixed Population RNA-Seq (2013, background): “Others have also suggested that this distribution is appropriate for these types of data[33–35].”
Sparse and Compositionally Robust Inference of Microbial Ecological Networks (2014, background): “from HMP or APG), the zero-inflated Negative Binomial (ziNB) distribution is a good choice, as it accounts for both overdispersion [1,57] and the preponderance of zero-count data points in microbial count datasets.”
Analysis of composition of microbiomes: a novel method for studying microbial composition (2015, unknown): “Recent publications (12, 13) have advocated modeling OTU counts using variants of the multinomial distribution, including the Dirichlet-multinomial distribution.”
Normalization and microbial differential abundance strategies depend upon data characteristics (2017, methodology): “Dirichlet-multinomial was a better fit for the Global Patterns data, as determined through the “HMP” package [67] and the C(α)-optimal test-statistics [79].”