Fig. 3: Single cell genomic analysis using DISCO. | Nature Communications

Fig. 3: Single cell genomic analysis using DISCO.

From: Digital microfluidic isolation of single cells for -Omics

Fig. 3

U87 cell monocultures were loaded onto DISCO devices, where they were fixed, targeted (0 or 1 cells per droplet), lysed, amplified, and analyzed according to the genome sequencing pipeline (Supplementary Fig. S5). For comparison, bulk samples of monocultures (~1000 cells) were also evaluated (foregoing the pipeline, with no amplification). a Lorenz curve of the average cumulative fraction of mapped reads as a function of the cumulative fraction of the U87 genome for single cells (“1 Cell,” blue, n = 10), and bulk samples (“Bulk,” gray, n = 3). The dashed red line represents a theoretical perfect match between reads and genome. b Plot of normalized number of reads as a function of chromosome for DNA collected from DISCO experiments with (blue, n = 10) and without (orange, n = 5) lysis of a single cell, and bulk samples (gray, n = 3). Error bars represent ±1 SEM, gray dots represent individual replicates, and the dashed red lines illustrate the length of each chromosome and the theoretical read distribution across the genome. c Representative plot of copy numbers measured for DNA extracted a single cell (top) and from bulk DNA (bottom). Data are plotted as a function of chromosome number, with dark blue, lt. blue, dark gray, and lt. gray circles representing the estimated copy numbers as determined by read depth analysis of bins containing 500 kb of uniquely mappable sequences and black line-segments representing the segmented medians as determined by circular binary segmentation. d Heatmap (with green-high and blue-low) of DNA copy numbers measured from single cells (n = 10) and bulk samples (n = 3). Each column represents a replicate, which are plotted as a function of location in each chromosome (rows). The color bar (bottom) illustrates the estimated DNA copy number. e Log2 correlation plot of single-cell and bulk data from d. Each circle represents the estimated copy number. A linear least squares regression fit (dashed red line) and coefficient of determination (R2 = 0.843) are shown.

Back to article page