To measure the certainty of scCODA’s credible effects, we calculate high-density intervals36 for each effect parameter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\beta }_{m,k}$$\end{document}βm,k. Due to the spike-and-slab prior formulation, posterior samples of β are naturally zero-inflated, with the extent depending on each effect’s inclusion probability.
