• 💾 Installation
• 🏀 Example
  • Sample group assignment json file
  • Rendering the analysis document
• Shiny
• ⏳ History

This analysis largely follows the DESeq2 vigniette.

You can install the development version of dereportr using devtools with:

devtools::install_github("paulklemm/dereportr")

You'll need

1 A json file containing the group assignments for each sample 2 Raw as well as library-size normalized counts

In order to run the DE pipeline, you need to specify a json file with group assignments for each sample, e.g.:

{
  "groups": {
    "BAT_W": ["BAT_W_2", "BAT_W_3", "BAT_W_4", "BAT_W_5"],
    "BAT_C": ["BAT_C_1", "BAT_C_2", "BAT_C_3", "BAT_C_4", "BAT_C_5"],
    "BAT_D": ["BAT_D_1", "BAT_D_2", "BAT_D_3", "BAT_D_4", "BAT_D_5"],
    "BAT_CD": ["BAT_CD_1", "BAT_CD_2", "BAT_CD_3", "BAT_CD_4", "BAT_CD_5"]
  },
  "comparisons": {
    "BAT_W vs BAT_C": {
      "group_a": "BAT_W",
      "group_b": "BAT_C"
    },
    "BAT_W vs BAT_D": {
      "group_a": "BAT_W",
      "group_b": "BAT_D"
    }
  }
}

Note that the comparisons section is optional. When there are no comparisons specified, the tool will automatically compare all groups pairwise.

You can use the built-in render function for the DESeq2 RMarkdown document.

count_data <- readr::read_csv("nf-rnaseq/results/salmon/salmon_merged_gene_counts.csv")
count_data_normalized <- readr::read_csv("nf-rnaseq/results/salmon/salmon_merged_gene_tpm.csv")
dereportr::run_differential_expression(
  path_config_json = "philipp_config.json",
  count_data = count_data,
  count_data_normalized = count_data_normalized,
  out_path = getwd()
)

You can also use the rmarkdown render function directly if you want to customize the rendering call.

count_data <- readr::read_csv("nf-rnaseq/results/salmon/salmon_merged_gene_counts.csv")
count_data_normalized <- readr::read_csv("nf-rnaseq/results/salmon/salmon_merged_gene_tpm.csv")
# Render command utilizing the default parameters
rmarkdown::render(
  system.file("rmd/differential_expression.Rmd", package = "dereportr"),
  params = list(
    path_config_json = "philipp_config.json",
    count_data = count_data,
    count_data_normalized = count_data_normalized,
    out_path = output_path
  ),
  # Change the intermediate path to the output to avoid write access errors
  intermediates_dir = output_path,
  knit_root_dir = output_path,
  # clean: TRUE to clean intermediate files created during rendering.
  clean = TRUE,
  output_dir = output_path,
  output_options = list(
    self_contained = TRUE
  )
)

Dereportr also provides a shiny app that allows for interactively analyzing the dereportr output. Find the app with inst/rmd/dereportr_shiny/app.R. You can send the path via the URL, e.g. http://127.0.0.1:6252/?dereportrpath=/beegfs/scratch/bruening_scratch/pklemm/2022-08-ecem-rnaseq/release/DESeq2/cortex/deseq_diff

• 2023-10-09
  • Fix error in updated tidyr::pivot_longer
• 2022-09-06
  • Add Shiny App for interactively analyzing the dereportr output
• 2021-03-22
  • Add full DESeq2 result table to "Differentially Expressed (DE) Genes" tab
  • Bump version to 0.3.1
• 2021-02-05
  • goterm_analysis_of_all_comparisons can now run up- and down-regulated genes separately
• 2021-02-04
  • Drop support for providing flat files, require to provide data frames
  • Change name do dereportr
  • Bump version to 0.3.0
• 2020-11-20
  • Add count_normalized and path_salmon_tpm variables that allow for proper filtering of minimum expressed genes based on counts normalized on library size
  • Bump version to 0.2.0
• 2020-10-21
  • Add minimum_padj parameter setting the minimum threshold for padj for a gene to be differentially expressed
  • Bump version to 0.1.0
• 2020-09-07
  • Add minimum_count parameter where for each gene, at least one sample has to be equal or larger than this count
  • Bump version to 0.0.6
• 2020-05-08
  • Improve heat map output and add table of DE genes. Bump version to 0.0.5
• 2020-03-23
  • Improve description and layout of analysis doc. Bump version to 0.0.4
• 2020-03-14
  • Put deseq2 diff files into a separate folder. Bump version to 0.0.3
• 2020-02-28
  • Added ability to input count_data directly instead of Salmon counts
• 2020-01-23
  • Add goterm analysis function using the mygo package goterm_analysis_of_all_comparisons
• 2020-01-08
  • Close Use TPM over counts file #5
• 2020-01-07
  • Remove old debug mode, add reference to Xaringan
  • Add Use TPM over counts file #5
• 2019-11-12
  • Add support for multiple conditions