This tool was developed by the Linköping iGEM team of 2020. Down below you can find all of the dependencies needed to run the developer version of the app.
Video tutorials on how to use ClusteRsy can be found on our youtube
Read more about our project on our Wiki
Adam Lång - Teamleader and full stack developer
Jake P - Full stack developer
Alexander Johansson - R shiny developer
Erika Mattsson - R shiny developer
Lucas Porcile - R shiny developer
Ronja Höglund - HTML/CSS developer
We have prepared a docker image, this is easy to set up and the software will be up and running in no time at all! Please follow the instructions below.
-
First you need to download Docker to your local machine.
-
Open up your terminal then run the following command
docker pull liuigem/clustersy_app
- Once the download is complete you can simply run the app by running the following command in the terminal
docker run -d --rm -p 3838:3838 liuigem/clustersy_app
The docker image is now running locally and it can be found either in the docker desktop application or visit your preferred web browser and then type:
http://localhost:3838/
ClusteRsy should now be up and running!
It's also possible to run the developers version of the app using R. To do this please follow the instructions below.
Before the software can be used there are a couple of dependencies that needs to be installed. Please follow the steps in Installation
MODifieR
We have included MODifieR, a R package for disease module identification. Enrichment analysis such as disease analysis, gene ontology analysis and pathway analysis using the Clusterprofiler package. We also provide visualization of the results as well as a database to store all the input and output data.
To run the developer version of this app there are a few dependencies that needs to added.
MODifieR requires Python 3 with additional Python libraries scipy, sqlite3, numpy and networkx. They are all included within Anaconda
In addition to this some R packages are required as well.
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install(c("AnnotationDbi",
"MODA",
"STRINGdb",
"limma",
"org.Hs.eg.db",
"foreach",
"doParallel",
"Rcpp",
"dynamicTreeCut",
"flashClust",
"reticulate",
"plyr",
"parallel",
"igraph",
"WGCNA",
"RSQLite",
"devtools",
"stackoverflow",
"preprocessCore",
"DESeq2",
"edgeR",
"openxlsx",
"ggplot2",
"ggdendro",
"ggrepel"),
version = "3.8")
Once these has been installed you can install MODifieR from GitHub:
devtools::install_git(url = "https://gitlab.com/Gustafsson-lab/MODifieR.git")
For the database to work an extended version of MODifieR needs to be installed from GitHub:
devtools::install_git(url = "https://github.com/ddeweerd/MODifieRDB.git")
Enrichment analysis is provided with the Clusterprofiler package and can be installed from BiocManager:
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("clusterProfiler")
It's now time to install the actual software. It's provided as a R-package and it can easily be installed from GitHub within
devtools::install_git(url = "https://github.com/igemsoftwareadmin/ClusteRsy-Linkoping.git")
How to run the app Once everything has been installed you can simply run the following using the console in R.
ClusteRsy::run_app()
We provide a SQL database. You can either create a new one or you can use our database that we used during modeling. A PPI network (STRING v.11 filtered to only contain human genes and a score > 700) is included by default.
1. How to set up our database
Here you can download our database. Once this has been downloaded, go to the ClusteRsy library folder and place it in the database folder.
2. How to set up a new database
If you want to use a new database only containing the default PPI network then simply skip step 1 and use in the R console.
ClusteRsy::run_app()
Please note that the first time you start the app it will load for a while, this is because a clique database i.e a database containing all of the possible cliques for the PPI network is being built. (approx. ~5mins)