In DeLorean: Estimates Pseudotimes for Single Cell Expression Data

library(ggplot2)
library(dplyr)

# Attach the DeLorean data frame to access members
dl <- analyse.variance(dl, adjust.cell.sizes=FALSE)
attach(dl)

Expression

This data set contains expression data, $x_{g,c}$, for $G=r nrow(dl$expr)$ genes and $C=r ncol(dl$expr)$ cells. The expression values are distributed as

stopifnot(ncol(expr) > 0)
stopifnot(nrow(expr) > 0)
qplot(as.vector(expr))

Cells

Meta data exists for r nrow(dl$cell.meta) cells. These were captured at r length(levels(cell.meta$capture)) distinct time points.

ggplot(cell.meta, aes(capture)) + geom_bar()

We have the following meta data for the cells:

summary(cell.meta)

Cell sizes

Estimates of cell sizes by capture time are:

ggplot(cell.sizes %>% left_join(cell.meta), aes(x=capture, y=S.hat)) +
    geom_boxplot()

Genes

Boxplots of expression over capture time of some randomly sampled genes. The red lines show the mean expression at each capture time.

plot(dl, type="expr.data")

We have the following meta data for the genes:

summary(gene.meta)

# Detach the previously attached DeLorean data frame
detach(dl)

DeLorean documentation built on May 2, 2019, 9:24 a.m.