README.md

In enriquea/feseR: Feature Selection workflow in R

feseR: Feature Selection in R

Introduction

We provide here a R package which combine multiple Feature Selection (FS) methods in a workflow for analizing high-dimentional omics data. The different feature selection steps can be classificated in: i) Univariate (Correlation filter and Gain Information), ii) Multivariate (Principal Component Analysis and Matrix Correlation based) and iii) Recursive Feature Elimination (wrapped up with a Machine Learning algorithm, e.g. Random Forest). The goal is to essemble the different steps in an efficient workflow to perform feature selection in the contex of classification and regression problems.

How to install

The first step is to install devtools:

install.packages("devtools")
library(devtools)

Then, we can install the package using:

install_github("enriquea/feseR")
library(feseR)

How to use

We provide here some examples for illustrating how to use the package.

This library has been used in:

Enrique Audain, Yassel Ramos, Henning Hermjakob, Darren R. Flower, Yasset Perez-Riverol; Accurate estimation of isoelectric point of protein and peptide based on amino acid sequences, Bioinformatics, Volume 32, Issue 6, 15 March 2016, Pages 821–827, article

How to cite

If you find useful this tool in your work, you could want citing us: Perez-Riverol Y, Kuhn M, Vizcaíno JA, Hitz M-P, Audain E (2017) Accurate and fast feature selection workflow for high-dimensional omics data. PLoS ONE 12(12): e0189875. https://doi.org/10.1371/journal.pone.0189875

Mainteiner

Enrique Audain (enriquea)

Dmitry Rychkov (drychkov)

Yasset Perez-Riverol (ypriverol)

enriquea/feseR documentation built on March 30, 2021, 4:12 p.m.