In latlio/tidyde: Tidy Differential Expression

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%",
  fig.height = 8,
  dpi = 320
)

tidyde

Perform differential expression analysis in a tidy way. Currently in early development. Only limma is supported.

Installation

You can install the required packages for this vignette with the following code:

install.packages(c("devtools", "tidyverse", "rlang"))
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
BiocManager::install("edgeR")
library(devtools)
install_github("latlio/tidyde")

Tidy Differential Expression analysis

tidyde is a toolkit for working with gene differential expression analysis in R. Currently, it includes a sample count data set, called GSE60450_Lactation-GenewiseCounts, and sample metadata, called SampleInfo_Corrected.

library(tidyde)
library(readr)
library(purrr)
library(dplyr)
library(stringr)
library(edgeR)
library(ggplot2)
counts <- read_delim("data/GSE60450_Lactation-GenewiseCounts.txt",
                     delim = "\t")
meta <- read_delim("data/SampleInfo_Corrected.txt",
                   delim = "\t") %>%
  mutate(FileName = str_replace(FileName, "\\.", "-"))

tidyde is composed of wrapper functions of the limma package that perform various steps of a standard differential expression analysis in a tidy way.

# Set up objects
meta <- read_delim("data/SampleInfo_Corrected.txt",
                   delim = "\t") %>%
  mutate(FileName = str_replace(FileName, "\\.", "-"))

my_design <- check_sample_names(counts, c(1,2), meta, FileName) %>%
  purrr::pluck("meta") %>%
  make_design_matrix(., c("Status"))

id <- as.character(counts$EntrezGeneID)

# Analysis
res <- check_sample_names(counts, c(1,2), meta, FileName) %>%
  purrr::pluck("mod_count") %>%
  filter_genes(., id, "edgeR") %>%
  make_voom(., my_design) %>%
  model_limma() %>%
  make_contrasts(design_matrix = my_design, Statuspregnant, Statusvirgin) %>%
  model_bayes()

tidyde includes broom-like methods for cleaning differential expression fit results, although it currently is not officially supported by broom. I'm actively working on that.

tidy.marray.lm(res)

Visualization for Meta-Analysis

tidyde includes functionality for working with results in ggplot2, including quick plots for common visualizations.

With tidy data, most data visualizations are easy to build from the ground up. Nevertheless, tidyde has several quick plot functions to make the process easier. For example, volcano_plot() takes a tidied DE analysis and plots a volcano plot.

tidy.marray.lm(res) %>%
  volcano_plot()

Because the results are still ggplot2 objects, it's easy to make changes to the plot.

tidy.marray.lm(res) %>%
  volcano_plot() + 
  scale_color_manual(values = c("#DA70D6", "black", "#00CC00"))

tidyde currently supports boxplots of counts, PCA plots, and volcano plots.

latlio/tidyde documentation built on Dec. 21, 2021, 9:40 a.m.