Extended Data Fig. 7: Complex set of molecular spike-ins.
From: Molecular spikes: a gold standard for single-cell RNA counting
(a,b) Number of unique spike-in molecules determined in the ground-truth complexity sequencing experiment as a function of sequencing depth for (a) each of the distinct spike-in sequences or (b) overall. (c) Scatter plot showing the Smart-seq3 molecule counts (y-axis) versus spUMI molecule counts (x-axis) in a randomly drawn HEK293FT cell. Dots are colored by spike-in transcript sequence. Spearman rank correlation r = 0.99. (d) Read mapping statistics for wells with HEK293FT cells and spike-ins only. Left panel: percentage of reads mapping to human exons (x-axis) against percentage of reads mapping to molecular spikes. Middle panel: Percentage of reads mapping to human exons (y-axis) against sequenced reads per well. Right panel: Percentage of reads mapping to molecular spikes (y-axis) against sequenced reads per well. (e) Scatter plot showing observed cellular RNA counts (y-axis) against the number of sequenced reads (x-axis), per cell, with linear regression shown as line and 95% confidence interval as gray shaded area. (d) The percent of reads aligning to molecular spike-ins as a function of sequence depth, with each colored line showing a unique cell.
