bias_histogram: Bias histogram
In d93espinoza/barcodetrackR: Functions for Analyzing Cellular Barcoding Data

Description Usage Arguments Value Examples

Given a summarized experiment, gives histogram of log biases for 2 cell types. Each stacked bar in the histogram represents a clone binned by log bias defined as the log2 of the percentage abundance in the sample specified in "bias_1" divided by the percentage abundance in "bias_2."

bias_histogram(
  your_SE,
  split_bias_on,
  bias_1,
  bias_2,
  split_bias_over,
  bias_over = NULL,
  remove_unique = FALSE,
  breaks = c(10, 2, 1, 0.5),
  text_size = 10,
  linesize = 0.4,
  ncols = 1,
  scale_all_y = TRUE,
  return_table = FALSE
)

`your_SE`	Your SummarizedExperiment of barcode data and associated metadata
`split_bias_on`	The column in 'colData(your_SE)' from which 'bias_1' and 'bias_2' will be chosen
`bias_1`	The first cell type (or other factor) to be compared. Must be a possible value of the split_bias_on column of your metadata. Will be on the RIGHT side of the histogram.
`bias_2`	The second cell type (or other factor) to be compared. Must be a possible value of the split_bias_on column of your metadata. Will be on the LEFT side of the ridge plot
`split_bias_over`	The column in 'colData(your_SE)' that you wish to observe the bias split for. The output will contain a faceted plot: one facet for each value of 'split_bias_over' comparing the samples matching 'bias_1' and 'bias_2' from the 'split_bias_on' argument.
`bias_over`	Choice(s) from the column designated in 'split_bias_over' that will be used for plotting. Defaults to all.
`remove_unique`	If set to true, only clones present in both samples will be considered.
`breaks`	Numeric. The breaks specified for bins on the x-axis (how biased the clones are towards one factor or the other).
`text_size`	The size of the text in the plot.
`linesize`	The linewidth of the stacked bars which represent individual barcodes
`ncols`	Numeric. Number of columns to plot on using plot_grid from cowplot.
`scale_all_y`	Logical. Whether or not to plot all plots on the same y axis limits.
`return_table`	Logical. If set to TRUE, instead of a plot, tbe function will return a list containing a dataframe for each sample-sample log bias combination containing each barcode sequence and its bias between the samples.

Histogram of log bias for two factors faceted over another set of factors. Or, if return_table is set to TRUE, a list of dataframes containing the log bias data for each bias comparison passed to the function.

data(wu_subset)
bias_histogram(
    your_SE = wu_subset, split_bias_on = "celltype",
    bias_1 = "B", bias_2 = "T",
    split_bias_over = "months", ncols = 2
)