were solely induced by the inherent negative correlations of the cell-type proportions (Fig. 1b). Yet, common statistical approaches currently applied in compositional cell-type analysis ignore this effect. For example, Haber et al.6 applied a univariate test based on Poisson regression, Hashimoto et al.3 a Wilcoxon rank-sum test, and Cao et al.7 proposed a method based on a generalized linear regression framework with a Poisson likelihood, all thus not addressing the issue of compositionality.
