Creates a figure using kernel density estimation to find mutational hotspots
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
example_data
README.md
mutation_kde.R

README.md

KDEMutationPlot

Creates a figure of a protein's mutational landspace using kernel density estimation.

You might need to install a few packages for the script to work: Biostrings, ggplot2, zoo, and gridExtra.

install.packages(c("ggplot2","zoo","gridExtra"))
source("https://bioconductor.org/biocLite.R")
biocLite("Biostrings")

You can run the script via RStudio after updating the hardcoded arguments, or you can run it from the command line using:

Rscript mutation_kde.R example_data/mutations.txt example_data/aa.txt example_data/domain.txt example_data/P29597.fasta example_data/test.pdf

Typically, I would get the FASTA file and domain information from Uniprot.org.

Example output (before manually adjusting label positions):

![Image of example output] (https://github.com/UNC-Major-Lab/KDEMutationPlot/blob/master/example_data/TYK2_KDE.png)