graph_metrics: Make lots of graphs!

In elsayed-lab/hpgltools: A pile of (hopefully) useful R functions

Make lots of graphs!

Description

Plot out a set of metrics describing the state of an experiment including library sizes, # non-zero genes, heatmaps, boxplots, density plots, pca plots, standard median distance/correlation, and qq plots.

Usage

graph_metrics(
  expt,
  cormethod = "pearson",
  distmethod = "euclidean",
  title_suffix = NULL,
  qq = NULL,
  ma = NULL,
  cv = NULL,
  gene_heat = NULL,
  ...
)

Arguments

expt	an expt to process
cormethod	The correlation test for heatmaps.
distmethod	define the distance metric for heatmaps.
title_suffix	Text to add to the titles of the plots.
qq	Include qq plots?
ma	Include pairwise ma plots?
cv	Include coefficient of variance plots? (they are slow)
gene_heat	Include a heatmap of the gene expression data?
...	Extra parameters optionally fed to the various plots

Value

a loooong list of plots including the following:

1. nonzero = a ggplot2 plot of the non-zero genes vs library size

2. libsize = a ggplot2 bar plot of the library sizes

3. boxplot = a ggplot2 boxplot of the raw data

4. corheat = a recordPlot()ed pairwise correlation heatmap of the raw data

5. smc = a recordPlot()ed view of the standard median pairwise correlation of the raw data

6. disheat = a recordPlot()ed pairwise euclidean distance heatmap of the raw data

7. smd = a recordPlot()ed view of the standard median pairwise distance of the raw data

8. pcaplot = a recordPlot()ed PCA plot of the raw samples

9. pcatable = a table describing the relative contribution of condition/batch of the raw data

10. pcares = a table describing the relative contribution of condition/batch of the raw data

11. pcavar = a table describing the variance of the raw data

12. qq = a recordPlotted() view comparing the quantile/quantiles between the mean of all data and every raw sample

13. density = a ggplot2 view of the density of each raw sample (this is complementary but more fun than a boxplot)

See Also

[plot_nonzero()] [plot_legend()] [plot_libsize()] [plot_disheat()] [plot_corheat()] [plot_topn()] [plot_pca()] [plot_sm()] [plot_boxplot()]

Examples

## Not run: 
 toomany_plots <- graph_metrics(expt)
 toomany_plots$pcaplot
 norm <- normalize_expt(expt, convert = "cpm", batch = TRUE, filter_low = TRUE,
                        transform = "log2", norm = "rle")
 holy_asscrackers <- graph_metrics(norm, qq = TRUE, ma = TRUE)

## End(Not run)

elsayed-lab/hpgltools documentation built on May 9, 2024, 5:02 a.m.