Fig. 4: Enhancing circadian clock signal in the mammalian liver by filtering spatial zonation signals.
From: SiFT: uncovering hidden biological processes by probabilistic filtering of single-cell data
a Single-cell RNA-seq of hepatocytes isolated at four different times within a day26. Figure is created with BioRender.com. b A UMAP representation of the original data (top), SiFT filtered data based on zonated genes (middle), and novoSpaRc zonated mapping (bottom). Plots are colored by the lobular layer (left) and time point (right). c Reconstructed temporal (top) and spatial (bottom) expression patterns of zonated and rhythmic genes (hlf and elovl3). Solid line depicts mean expression and bandwidth represents the 95% confidence interval over \(n=14364\) cells. d Pearson correlation between reconstructed and original rhythmic signals as a function of the number of temporal reference genes used for the reconstruction (Pearson correlation is evaluated over \(n=20\) rhythmic informative genes, results present the mean value over \(n=10\) independent repetitions of novoSpaRc). In the box plot middle line, median; box boundary, interquartile range (IQR); whiskers, 1.5*IQR; minimum and maximum, not indicated in the box plot; gray dots, points beyond the minimum or maximum whisker. Source data of (c, d) are provided as a Source Data file.
