Violin SuperPlots: visualizing replicate heterogeneity in large data sets
Martin Kenny, Ingmar Schoen
Abstract
A recent article in MBoC (Goedhart, 2021) presented a web interface for the creation of \n‘SuperPlots’. SuperPlots were introduced by Lord and colleagues last year (Lord et al., 2020) to \nvisualise both cell-level variability within replicates as well as the experimental reproducibility \nbetween replicates in one single plot. Simple bar charts or boxplots of mean or median values \nfrom experimental replicates mask the contribution of underlying cell-to-cell variations in \nindividual experiments, whereas pooling cell-level data across replicates overemphasises \nstatistical differences. The SuperPlot put forward by Lord et al. uses a beeswarm plot to display \nthe cell-level data color-coded according to the individual replicates, and overlays the mean (or \nmedian) and error bars (standard deviation or confidence intervals) of each replicate (Figure\n1a). The new web interface (Goedhart, 2021) offers an online option for researchers to generate \nbeeswarm SuperPlots, as well as RainCloud plots (Allen et al., 2021), using their own data. We \nwelcome the transparency brought by SuperPlots and would like to introduce an augmentation, \nthe Violin SuperPlot, to further simplify visual inspection of raw data containing large sample \nsizes.