IMPORTANT - THIS REPOSITORY IS NO LONGER MAINTAINED

Cancer Systems Biology, Section of Bioinformatics, Department of Health and Technology, Technical University of Denmark, 2800, Lyngby, Copenhagen, Denmark

Cancer Structural Biology Group, Danish Cancer Society Research Center, Strandboulevarden 49, 2100, Copenhagen, Denmark

IMPORTANT NOTE: We have recreated this repository because of issues while preparing it for the submission to BioConductor. The original CAMPP2 repository, including commits, PRs and contribution, is available here

CAncer-bioMarker-Prediction-Pipeline - CAMPP2

Installation instructions

Installation from BioConductor

To install the CAMPP2 package from Bioconductor, you need to follow these steps:

1. Install the Bioconductor package manager, BiocManager, if you don't have it already. You can do this by running the following command in your R console:

if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")

2. Install the CAMPP2 package using BiocManager by running the following command:

BiocManager::install("CAMPP2")

Create and activate conda environment with R and devtools in your project directory

If you don't have access to conda please see the Miniconda installer page (https://docs.conda.io/en/latest/miniconda.html) on instructions on how to install Miniconda.

Remember to have installed BiocManager upfront before proceeding with the installation of CAMPP2. install.packages("BiocManager")

Once you have installed it:

conda create --prefix env -c conda-forge r-base=4.1.3 r-devtools
conda activate ./env

Install CAMPP2 from Github

open R console:
R

load devtools library
library(devtools)

install CAMPP2 from private Github repository (using your personal token and commit ID) devtools::install_github(repo = "ELELAB/CAMPP2",auth_token="<your personal token to github>")

CAMPP2 development environment

The CAMPP2 development environment is made available as Docker image, defined in a Dockerfile in this repository. It uses the Bioconductor devel Docker image as a base, and further downloads and installs the requirements for CAMPP2. In order to use it, please install Docker if you don't have it already, and follow these instructions:

1. Clone the CAMPP2 GitHub repository in a local folder:

git clone https://www.github.com/ELELAB/CAMPP2.git
cd CAMPP2

2. Build the container image:

docker build --pull -t campp2:devel-20230206 ./docker

We recommend tagging the container with a date consistent with the day it's been created, since it might contain different package versions depending on the date, as done in this example.

Notice that this needs to be done only once, as well as every time you intend to upgrade your image to account for the latest Bioconductor devel docker image (see below)

3. Run a container from the image you just created:

In order to access the development environment via RStudio on browser, run:

docker run \
    --rm \
    -v /path/to/my/CAMPP2:/home/rstudio/CAMPP2 \
    -p 8787:8787 \
    -e PASSWORD=campp2 \
    campp2:devel-20230206

Open your web browser and head to http://localhost:8787. Log in using rstudio as username and campp2 as password. You should be able to access the CAMPP2 development folder from the Rstudio interface.

If you prefer running the R command line prompt (e.g. using R, it works in the same exacy way with Rscript):

docker run \
    --rm \
    -it \
    -v /path/to/my/CAMPP2:/home/rstudio/CAMPP2 \
    campp2:devel-20230206 \
    R

if you want to access the bash command line prompt of the container you can instead run:

docker run \
    --rm \
    -it \
    -v /path/to/my/CAMPP2:/home/rstudio/CAMPP2 \
    campp2:devel-20230206 \
    bash

Notice that in these commands:

• the /path/to/my/CAMPP2 should be replaced by the actual absolute path to the CAMPP2 development folder, depending on where it is located in your computer
• the container name should be changed to the actual name of the container that you built in step 2. If you're unsure what the container name or tag is, run docker image list, you should be able to find both (a container is specified as NAME:TAG)

We recommend to rebuild your image every few weeks to account for changes in the original BioConductor development image and in the CRAN/Bioconductor repositories. To do so, just repeat step 2 and adjust step 3 according to your new tag.

Running on an Apple Silicon Macs

If you are running the container on Apple Silicon (i.e. M1 and M2 at the time of writing) we currently recommend running the container under x86 emulation using Rosetta2. In order to do so,

1. Install Rosetta2 by running on your terminal:

/usr/sbin/softwareupdate --install-rosetta

2. Turn on the "Use Rosetta for x86/amd64 emulation on Apple Silicon" option in Docker, currently located in Settings, Features under development

Further, we recommend adding the --platform linux/amd64 option to the docker run command lines above, for example:

docker run \
    --rm \
    -it \
    -v /path/to/my/CAMPP2:/home/rstudio/CAMPP2 \
    --platform linux/amd64 \
    campp2:devel-20230206 \
    bash

Example data

The data for testing the functions and workflow includes 2 BRCA datasets (campp2_brca_1, campp2_brca_2) and the associated metadata (campp2_brca_1_meta, campp2_brca_2_meta). Each dataset is represented by raw read counts (10000 genes) for 30 samples: 20 tumours which are divided into 4 subtypes (each subtype has 5 samples), and 10 normals. Metadata includes information about diagnosis, age, vital status, days to death, outcome time, tumor stage, subtype, outcome and survival.

Both, raw read counts and metadata were extracted from TCGA BReast CAncer dataset (TCGA-BRCA) Level 3 data.

Test data (gene counts and metadata for 2 data sets) are integrated into CAMPP2 package and accessible as campp2_brca_1, campp2_brca_2, campp2_brca_1_meta, campp2_brca_2_meta variables once the package is installed. Results from intermediate steps (not described here) are also integrated and used as an input for running the examples of the functions. R data object files (.rda) are available on https://github.com/ELELAB/CAMPP2/tree/main/data.

Example run

Default settings can be executed in R using:

library(CAMPP2)

Test data are already part of the CAMPP2 package so user doesn't need to download them. In case you want to load .rda objects manually from the cloned repository, you can use this code:

load("./data/campp2_brca_1.rda")
load("./data/campp2_brca_1_meta.rda")
load("./data/campp2_brca_2.rda")
load("./data/campp2_brca_2_meta.rda")

Default workflow could be run using this command:
runCampp2(batches=c("tumor_stage","tumor_stage"),prefix="test_CAMPP2", data1=campp2_brca_1, data2=campp2_brca_2, metadata1=campp2_brca_1_meta,metadata2=campp2_brca_2_meta, groups=c("IDs", "diagnosis","IDs", "diagnosis"), technology=c("seq","seq"))

For testing the functions, you can consider the code present in campp2_example_Run.R (git repository).

For more details, see help page of the main function runCampp2.