Spanning tree Progression of Density normalized Events (SPADE) is an algorithmic visualization tool for high-dimensional flow and mass cytometry data. SPADE takes in FCS files and settings, and outputs a network graph file (GML), PDFs of the graphs, and FCS files with the “cluster” column appended. SPADE is implemented as a program in the R statistical computing environment. Therefore, to use SPADE, you must install R and the required R packages onto your computer. Once installed, you will no longer need to open R to access SPADE in FlowJo, In FlowJo 10.1r7 SPADE is run as a Plugin, which takes a gated population from FlowJo and runs SPADE within R to generate the associated derivative files. For more information on SPADE, please see the references below.
- Please see the Installing Plugins page for detailed information on how to obtain and setup FlowJo plugins.
- This plugin requires the R statistical computing environment and the packages “devtools” and “spade”.
- Download R from the Comprehensive R Archive Network (CRAN) website. Follow the links and installation directions for your operating system.
- If installing R on a Mac: Install R in your Applications folder. FlowJo will automatically look for R within Applications. Do not specify the path to your R installation within the Diagnostics Preferences.
- If installing R on a PC: R can be installed in any folder, but you must tell FlowJo the location of your R installation by selecting it in the Diagnostics panel of your FlowJo preferences.
Installing SPADE within R
The Nolan Lab group who developed SPADE no longer supports distribution of SPADE via CRAN or Bioconductor. Instructions and updates can be found on the at the Nolan Lab Github SPADE site.
- open R and enter:
install.packages("devtools") library(devtools) devtools::install_github("nolanlab/Rclusterpp") source("http://bioconductor.org/biocLite.R") devtools::install_github("nolanlab/spade")
- test the installation by typing at the R prompt:
- you should receive a response that includes:
Loading required package: igraph Loading required package: Rclusterpp Loading required package: Rcpp Loading required package: RcppEigen
- Select/highlight a population within the gating tree of a sample you wish to run SPADE on.
- Click on Spade within the Plugins menu. A Spade window will appear.
- Select/Highlight the parameters you wish to use for the SPADE calculation.
- Enter a valid number of clusters (3-1000).
- Click OK to run.
- A node called Spade of ‘population name’ will be created on the selected population, and will display a (calculating) message.
- Once the calculation completes within R and results are returned to FlowJo, the (calculating) message will disappear, and cluster populations will be created as child gates of the original SPADE input population.
- Double Click on the Spade of ‘population name’ will bring you back to the GUI where you can choose to:
- Show Trees – displays the PDF output tree diagram files produced by SPADE in your default web browser.
- Get Zip File – opens up the zip directory containing all files associated with the SPADE, including input and output FCS, the network GML and PDF files.
- Show Output – displays information entered into the R computing session to execute the SPADE program.
For more information on installing and running specific Plugins:
Questions about plugins or FlowJo? Send us an email at TechSupport [at] FlowJo [dot] com