Aitchison's Compositional Data Analysis 40 Years On: A Reappraisal

Abstract

The development of John Aitchison's approach to compositional data analysis is followed since his paper read to the Royal Statistical Society in 1982. Aitchison's logratio approach, which was proposed to solve the problematic aspects of working with data with a fixed sum constraint, is summarized and reappraised. It is maintained that the principles on which this approach was originally built, the main one being subcompositional coherence, are not required to be satisfied exactly -quasi-coherence is sufficient in practice. This opens up the field to using simpler data transformations with easier interpretations and also for variable selection to be possible to make results parsimonious. The additional principle of exact isometry, which was subsequently introduced and not in Aitchison's original conception, imposed the use of isometric logratio transformations, but these have been shown to be problematic to interpret. If this principle is regarded as important, it can be relaxed by showing that simpler transformations are quasi-isometric. It is concluded that the isometric and related logratio transformations such as pivot logratios are not a prerequisite for good practice, and this conclusion is fully supported by a case study in geochemistry provided as an appendix.

Ratios in Disguise, Truths Arise: Glycomics Meets Compositional Data Analysis

“…Factoring in this scale difference led to much stronger findings (Fig. 5b, Supplementary Table 11) that were, on average, indicating an increase in most oligo- and pauci-mannose N -glycans in cancer, including the abovementioned structure ( d combined = 1.31, p adj = 6.2 x 10 - 12 ). Hence, analyzing the data without scale would have erroneously resulted in the conclusion that some of these structures decrease in absolute abundance between conditions, whereas the opposite seems to be true.…”

“…It is a mathematical imposition, not a biological phenomenon, and necessitates a tailored analytical approach to focus the findings on biological effects instead. Compositional data analysis (CoDA) 12 provides such a framework, respecting the relative scale of the data and avoiding the misapprehensions that traditional methods incur.…”

A flexible Bayesian tool for CoDa mixed models: logistic-normal distribution with Dirichlet covariance

“…where D is the reference category. In Greenacre et al [37], the authors depicted some criteria to select the reference category. They recommended choosing the one whose logarithm has low variance as a reference, and avoiding taking a reference with low relative abundances across samples.…”

“…They can be constructed in different ways, including centered log-ratio, isometric log-ratio or additive log-ratio, among others [36]. In this work, we focus on the well-known additive log-ratio transformation because of its straightforward interpretation [37], and due to its being a one-to-one mapping from S D to R D−1 . It is defined as:…”

Power Transformations of Relative Count Data as a Shrinkage Problem

“…When using finite, "small enough" values of the power parameter β, the subcompositional coherence of LRA remains approximately satisfied while there is no need for zero imputation (as CA does not involve logarithms). One can obtain an optimal value of the power parameter in the sense that it maximizes the Procrustes correlation between the log-ratio transformed data (using zero imputation) and the coordinates from the power-transformed CA (keeping the zeros) [28].…”