In elisabettasciacca/DEGGs: Differentially Expressed Gene-Gene pairs

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

The DEGGs package enables the exploration of differential gene-gene correlations in different phenotipical groups.
This vignette uses breast cancer RNA-seq data from The Cancer Genome Atlas program (TCGA) to show the functionalities of the package.
In this example we will compare two different breast cancer subgroups: HER2-positive and luminal A.

Installing from Github

devtools::install_github("elisabettasciacca/DEGGs", build_vignettes = TRUE)

Overview

Let's start by loading the package and sample data

library(DEGGs)
data("BRCA_metadata")
data("BRCA_normCounts")

Always make sure clinical and counts data are aligned

identical(colnames(BRCA_normCounts), rownames(BRCA_metadata))

We can then proceed generating specific gene networks associated with the two subgroups.
We will specify the groups of interest using the subgroups parameter. If not specified, all breast cancer subtypes will be compared (when more than two groups are compared, a one-way ANOVA is used in place of the robust linear model).
The provided count matrix uses entrez IDs, therefore we will set entrezIDs = TRUE and convert_to_gene_symbols = TRUE to visualise the matching gene symbols in our output.

subnetworks_object <- generate_subnetworks(
  normalised_counts = BRCA_normCounts,
  metadata = BRCA_metadata,
  subgroup_variable = "SUBTYPE",
  subgroups = c(
    "BRCA_Her2",
    "BRCA_LumA"
  ),
  entrezIDs = TRUE,
  convert_to_gene_symbols = TRUE,
  cores = 2
)

Visualisation

To inspect the network in a shiny-style, interactive way, we use the View_interactive_subnetwork function that only requires an object of class DEGGs generated from the generate_subnetworks.

View_interactive_subnetwork(subnetworks_object)

This will let you navigate the networks associated to each subgroup of patients (1), filter by gene-gene link significance (2), and search for specific genes (3).

 

{width=60%}

 

When clicking on a link, the related gene-gene regression plot will appear on the left.

 

{width=50%}

 

When clicking on a node, a box plot will visualise the expression of the single node across the subgroups.

 

{width=60%}

 

List significant gene-gene interactions

To get a table listing all the significant gene-gene pairs found in all subgroups, you will need to use the extract_sig_deggs with subnetworks_object as input:

head(extract_sig_deggs(subnetworks_object))

The listed genes can be used as features of machine learning models (see Sciacca, E., Surace, A.E.A., Alaimo, S. et al. Network analysis of synovial RNA sequencing identifies gene-gene interactions predictive of response in rheumatoid arthritis. Arthritis Res Ther 24, 166 (2022). https://doi.org/10.1186/s13075-022-02803-z).

Single regression plots

To plot the differential regression fits for a single gene-gene interaction the print_regressions function can be used by specifying the names of the two genes:

print_regressions(
  gene_A = "NOTCH2", gene_B = "DTX4",
  deggs_object = subnetworks_object,
  legend_position = "bottomright"
)

Run View_interactive_subnetwork in a non-blocking way

Shiny typically locks the R console until the user presses escape to close the shiny app. However this behaviour can be bypassed by calling the app in background as explained by Will Landau in his (post).
To do so, the shiny function must be wrapped in a second function and run at a given IP address and TCP port.

run_app <- function(host = "127.0.0.1", port = 49152, deggs_object = subnetworks_object) (
  print(View_interactive_subnetwork(subnetworks_object, host = host, port = port))
)

process <- callr::r_bg(
  func = run_app,
  args = list(
    host = "127.0.0.1",
    port = 49152,
    subnetworks_object = subnetworks_object
  ),
  supervise = TRUE
)

To know when the app is up and running the pingr package can be used:

while (!pingr::is_up(destination = "127.0.0.1", port = 49152)) {
  if (!process$is_alive()) stop(process$read_all_error())
  Sys.sleep(0.01)
}

After the loop completes, the app can be open in a web browser:

browseURL("http://127.0.0.1:49152")

Once the user completes his analysis, the app can be closed:

process$kill()

Session info

sessionInfo()

elisabettasciacca/DEGGs documentation built on April 27, 2024, 12:51 a.m.