MSTherm is an R package to aid in modeling and analyzing mass spectrometry-based protein melting data. Quantitative data is imported and normalized and thermal behavior is modeled at the protein level. Methods exist for normalization, visualization, exploratory analysis, and data export.
The following command in R will installed the latest CRAN release:
install.packages("mstherm")
To install the latest commit to master
, in Linux run:
git clone https://github.com/jvolkening/r-mstherm.git
cd r-mstherm
R CMD INSTALL .
You may need to install some prerequisites first depending on your environment -- the installer will notify you of this.
Once you have produced spectral quantification tables and created the necessary metadata files describing the experimental setup (as described in the vignette), a modeling session can be a simple as:
library(mstherm)
expt <- MSThermExperiment("control.tsv", "annotations.tsv")
expt <- normalize_to_std(expt, "BOVINE_SERUM_ALBUMIN")
res <- model_experiment(expt)
tbl <- as.data.frame(res)
pdf("plots.pdf", 5, 5, pointsize=10)
plot(res)
dev.off()
(Of couse, you will likely want to experiment with the various options available to the modeling and plotting methods).
Further details of the required input files and available methods are available in the package vignette.