mrcschwering / rbdpredict

R package for unbiased enigm RNA-binding protein (RBP) prediction based on shrinkage discriminant analysis. Enigm RBPs have been found to make up for one half of all RBPs.

unbiased enigm RNA-binding protein prediction

This package provides functions for identifying RNA-binding proteins (RBPs) from their amino acid. In contrast to common Hidden-Markov-Model-based procedures RBDpredict does not use prior knowledge such as known RNA-binding domains. RBDpredict uses supervised learning to identify sequence structures which are abundant in RBPs but not in non-RBPs. Thus, it is capable of identifying RBPs which bind to RNA due to unstructured amino acid sequences -- enigm RBPs. These enigm RBPs have been found to map to disordered regions and are thought to make one half of all RBPs Castello et al.

Getting started

The package can be installed from R with devtools::install_github("mRcSchwering/RBDpredict", build_vignettes = TRUE). This will also create vignettes which explain the package functions browseVignettes("RBDpredict"). Building the vignette can take a while. If no vignette is needed use devtools::install_github("mRcSchwering/RBDpredict") to install the package without vignettes.