plot_clonealign: Plot gene expression and copy number

In kieranrcampbell/clonealign: Assigning scRNA-seq to clone-of-origin using copy number from ultra-low-depth scDNA-seq

Description

Plot gene expression and copy number as a function of genomic coordinate.

Usage

plot_clonealign(
  sce,
  clones,
  cnv_data,
  chromosome = "1",
  chr_str = "chr",
  start_str = "start_position",
  end_str = "end_position",
  jitter_cnv = TRUE,
  ggplot_palette = "Set1",
  expression_ylim = c(-0.15, 0.15),
  cnv_dodge_sd = 0.1
)

Arguments

sce	A SingleCellExperiment. Note this must have the fields chr_str, start_str, and end_str in rowData. See details.
clones	The clone assignments of each cell - must be of the same set as the colnames of L
cnv_data	The gene by clone copy number matrix used as input to clonealign_fit. The genes (rows) of this must match to the genes in sce
chromosome	The chromosome to plot
chr_str	The column of rowData(sce) that refers to the chromosome of each feature
start_str	The column of rowData(sce) that refers to the start position of each feature
end_str	The column of rowData(sce) that refers to the end position of each feature
jitter_cnv	Logical - if true random noise is added to the copy number profiles to help display all clones in case two lie on top of each other
ggplot_palette	The palette to be passed to scale_colour_brewer
expression_ylim	The y axis limits on the smoothed expression plot
cnv_dodge_sd	The standard deviation of the (0 mean) noise to add to the CNV plot

Details

This function requires the chromosome, start, and end positions of each features to plot. These are encoded as the chr_str, start_str, and end_str features in rowData(sce) respectively. If we have ensembl ids or similar we can get add the required fields using the getBMFeatureAnnos function in the scater package -
sce <- getBMFeatureAnnos(sce, filters = "ensembl_gene_id", attributes = c("chromosome_name", "start_position", "end_position"), feature_symbol = "hgnc_symbol", feature_id = "ensembl_gene_id", dataset = "hsapiens_gene_ensembl")
Then we would call plot_clonealign using chr_str == "chromosome_name", start_str == "start_position", and end_str == "end_position".

Value

A ggplot2 object displaying a CNV and gene expression track for each (inferred) clone

Examples

library(SingleCellExperiment)
library(scater)
library(dplyr)
data(example_clonealign_fit)
cnv_data <- rowData(example_sce)[,c("A", "B", "C")]
gene_position <- as_data_frame(cnv_data) %>%
mutate(gene = seq_len(nrow(cnv_data))) %>%
  arrange(A, B, C) %>%
 mutate(position = seq_len(nrow(cnv_data))) %>%
  arrange(gene) %>%
  .$position

rowData(example_sce)$chromosome <- "1"
rowData(example_sce)$start_pos <- gene_position
rowData(example_sce)$end_pos <- gene_position
example_sce <- normalize(example_sce)
plot_clonealign(example_sce, example_clonealign_fit$clone, cnv_data,
chromosome = "1",
chr_str = "chromosome",
start_str = "start_pos",
end_str = "end_pos")

kieranrcampbell/clonealign documentation built on Dec. 18, 2020, 3:49 a.m.