Fig. 2.

Boxplots of ARIs for 10 clustering methods across 100 simulations. a Investigating how different heterogeneities among multiple individuals (measured by mean ${\sigma }_{ik}^2$ values) affect clustering results. The simulated dataset consists of 10 individuals with 400 cells for each. b Investigating how different numbers of individuals affect clustering results. We set the level of heterogeneity (mean of ${\sigma }_{ik}^2$) among individuals as 0.1. In the boxplots, the box spans from the first to third quartile (depicting median as a line in the middle), the whiskers extend to 1.5× IQR (interquartile range)