Microbiome Datasets Are Compositional: And This Is Not Optional
By G. Gloor, Jean M. Macklaim, Vera Pawlowsky-Glahn, Juan José Egozcue. Frontiers in Microbiology, 2017
2,474 citations recorded by the source corpus.
Datasets collected by high-throughput sequencing (HTS) of 16S rRNA gene amplimers, metagenomes or metatranscriptomes are commonplace and being used to study human disease states, ecological differences between sites, and the built environment. There is increasing awareness that microbiome datasets generated by HTS are compositional because they have an arbitrary total imposed by the instrument. However, many investigators are either unaware of this or assume specific properties of the compositional data. The purpose of this review is to alert investigators to the dangers inherent in ignoring the compositional nature of the data, and point out that HTS datasets derived from microbiome studies can and should be treated as compositions at all stages of analysis. We briefly introduce compositional data, illustrate the pathologies that occur when compositional data are analyzed inappropriately, and finally give guidance and point to resources and examples for the analysis of microbiome datasets using compositional data analysis.
Subfields: Artificial Intelligence, Molecular Biology, Periodontics
View Paper: https://www.semanticscholar.org/paper/7ec205c81289ec1aeed8ef643f81e49465e2f1c0
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Eric C. Grunsky
- John Bacon-Shone
- Ionas Erb
- Thomas P. Quinn
- Michael J. Greenacre
- Karel Hron
- Michael R. McLaren
- Levi Waldron
In Molecular Biology
- Curtis Huttenhower
- Malte D. Luecken
- Eric A. Franzosa
- Fabian J Theis
- Levi Waldron
- George Weingart
- Maren Büttner
- Timothy L. Tickle
In Periodontics
Citations in the corpus
- What drives study‐dependent differences in distance–decay relationships of microbial communities? (2021, background): “Alternatively, metrics such as the Aitchison distance perform well when appropriate (centred log- ratio) transformations are applied to counts (Gloor et al., 2017), or recently developed metrics, such as the rank bias overlap index, show promise for analysing similarity between communities based on…”
- scCODA is a Bayesian model for compositional single-cell data analysis (2021, methodology): “To account for the inherent bias present in cell-type compositions, we drew inspiration from methods for compositional analysis of microbiome data 8,9 and propose a Bayesian approach for cell-type composition differential abundance analysis to further address the low replicate issue.”
- Network analysis methods for studying microbial communities: A mini review (2021, background): “Firstly, microbiome data are compositional [11]; i.e. microbial counts represent proportions instead of absolute abundances.”
- Current best practices in single‐cell RNA‐seq analysis: a tutorial (2019, background): “Future developments in this field will likely borrow from the mass cytometry (e.g. Tibshirani et al , 2002; Arvaniti & Claassen, 2017; Lun et al , 2017; Weber et al , 2018) or the microbiome literature (Gloor et al , 2017), where compositional data analysis has received more attention.”
- NetCoMi: network construction and comparison for microbiome data in R (2020, neutral): “Rather, proportions are always compositional, even if the original data are not [11].”
- Applications of Machine Learning in Human Microbiome Studies: A Review on Feature Selection, Biomarker Identification, Disease Prediction and Treatment (2021, methodology): “…number of publications motivating and using the log-ratio methodology of compositional data for statistical processing of microbiome (e.g., Gloor et al., 2017; Silverman et al., 2017; Quinn et al., 2018; Randolph et al., 2018; Rivera-Pinto et al., 2018; Jiang et al., 2019; Quinn and Erb,…”
- Aitchison’s Compositional Data Analysis 40 Years on: A Reappraisal (2022, background): “In recent years, the approach has been taken up by biologists working with “wide” datasets, for example hundreds of bacterial species or thousands of genes [26, 70, 68], where the total count in each sample has no substantive meaning, making the data strictly compositional.”
- microViz: an R package for microbiome data visualization and statistics (2021, background): “The former provides a balanced intermediate between unweighted and weighted UniFrac (Chen et al., 2012), and the latter is a distance measure designed for use with compositional data, such as sequencing read counts (Gloor et al., 2017).”
- Multivariable association discovery in population-scale meta-omics studies (2021, result): “In sharp contrast to previous claims, however, compositionality-corrected methods such as ANCOM [14,25] as well as specialized normalization and transformation methods such as CLR [52] did not improve performance over non-compositional approaches [8,53], consistent with recent findings that composit”
- Consistent and correctable bias in metagenomic sequencing experiments (2019, methodology): “The left column shows taxon relative abundances as in Figure 4B and the right column shows the first two principal components from a compositional principle-components analysis (Gloor et al., 2017).”