Skip to contents

Plot the total number of barcodes detected in a sample

Usage

plotDetectedBarcodes(
  counts,
  percentile = 0.95,
  plot = T,
  sample.order = NULL,
  group = NULL,
  title = NULL
)

Arguments

counts

DGEList or dataframe containing raw or normalised barcode counts

percentile

desired percentile value. 95th percentile by default

plot

Logical. plot data instead of returning counts table.

sample.order

desired ordering of the samples on the plot

group

grouping field in dgelist$samples to color samples by

title

desired plot title

Value

Returns a histogram plot of the number of detected barcodes per sample.

Examples

data(test.dge)
plotDetectedBarcodes(test.dge, percentile = .95)

plotDetectedBarcodes(test.dge, plot = FALSE)
#>    Sample Barcodes
#> 1    T0-1      593
#> 2    T0-2      590
#> 3   S10-1        3
#> 4   S10-2        4
#> 5   S11-1       19
#> 6   S11-2       20
#> 7   S12-1       14
#> 8   S12-2       14
#> 9   S13-1       10
#> 10  S13-2       11
#> 11  S14-1        6
#> 12  S14-2        6
#> 13  S15-1        7
#> 14  S15-2        7
#> 15  S16-1       19
#> 16  S16-2       20
#> 17  S17-1        5
#> 18  S17-2        6
#> 19  S18-1       27
#> 20  S18-2       27
#> 21   S1-1       14
#> 22   S1-2       15
#> 23   S2-1       12
#> 24   S2-2       11
#> 25   S3-1       10
#> 26   S3-2       10
#> 27   S4-1       25
#> 28   S4-2       26
#> 29   S5-1       11
#> 30   S5-2       12
#> 31   S6-1        9
#> 32   S6-2       10
#> 33   S7-1       16
#> 34   S7-2       16
#> 35   S8-1       20
#> 36   S8-2       21
#> 37   S9-1       19
#> 38   S9-2       19