Normalization of Microbiome Profiling Data

Abstract

Normalization is a term that is often used but rarely defined and poorly understood. The number of choices of normalization procedure is large—some are inappropriate or inadmissible—and all are narrowly relevant to a specific analysis that depends on both the nature of the data and the question being asked. This chapter describes key definitions of normalization as they apply in metagenomics, mainly for taxonomic profiling data; while also demonstrating specific, reproducible examples of normalization procedures in the context of analysis techniques for which they were intended. The analysis and graphics code is distributed as a supplemental companion to this chapter so that the motivated reader can re-use it on new data.