The Gini coefficient as a surrogate for the regulability or homeostasis of metabolite concentrations: focus on ergothioneine

Abstract The Gini coefficient is a simple non-parametric statistic that reflects the degree of variation of a property between individuals or samples, taking a value between zero and one. We here develop the idea that if the metabolism of a metabolite is tightly regulated it should have a low Gini coefficient whereas if for instance it is exogenous and present only in a subset of samples (e.g. from individuals that have ingested the substance and/or metabolised it to various degrees) the Gini coefficient will be much greater. Analysis of published datasets shows this to be true, allowing one to suggest that the Gini coefficient might assist in assessing both regulability and the origin of a particular metabolite. Plasma amino acid levels are especially tightly regulated (some have a Gini coefficient less than 0.1, a value lower than that for any transcript), while exogenous drugs and their metabolites have the highest Gini coefficients (some over 0.8). Vitamins are essential and exogenous (but regulated) and have intermediate values (ca. 0.2–0.4). Ergothioneine is a valuable nutraceutical that is not synthesised by humans and it is also of some interest to assess the variation in its distribution. A number of studies find it normally to have a Gini coefficient in the range 0.3–0.4, and with lower levels of the Gini coefficient (as in a pilot study summarised here) suggesting a relatively weak exogenous exposure. Overall, we consider that the Gini coefficient provides a convenient means of describing the extent to which a metabolite concentration varies over a metabolomic dataset, and hence the extent to which it may be regulated.