pipelines

pipelines is a R package for performing different sorts of exploratory analyses on omics data sets.

Installation

1. Install devtools from CRAN with install.packages("devtools").

2. Install/Update pipelines with devtools::install_github("erdto/TPDT")

After that you load it as usual with library(pipelines)

Demo

To give a short example, we will simulate some metabolite data, 3 groups and 2 covariates that should be controlled for.

library(pipelines)
metdata <- matrix(rnorm(20 * 100, sd = 5), nrow = 100)
groups <- factor(sample(c("N", "FL", "DLBCL"), size = 100, replace = TRUE))
covariates <- data.frame(C_Sex = factor(sample(0:1, size = 100, replace = TRUE)),
                         C_Age_at_diagnosis = as.numeric(sample(20:60, 100, replace = TRUE)))

# run it
pipeline_pairwise(data = metdata, groups = groups, 
                  compname = "demo_pairwise",
                  vargroups = vargroups,
                  covars = covariates)

# The output is saved in an excel file and plots in pdfs in the working directory.

For further help, look into the vignette or the help pages with ?pipeline_pairwise.

Volcano Plot

As a visual summary, among the saved pdfs are volcano plots for each combination of groups. The observations beyond the cutoff lines in x and y direction can be considered worth to further investigate.