biosigner-package: Molecular signature discovery from omics data

In biosigner: Signature discovery from omics data

Description

Feature selection is critical in omics data analysis to extract restricted and meaningful molecular signatures from complex and high-dimension data, and to build robust classifiers. This package implements a new method to assess the relevance of the variables for the prediction performances of the classifier. The approach can be run in parallel with the PLS-DA, Random Forest, and SVM binary classifiers. The signatures and the corresponding 'restricted' models are returned, enabling future predictions on new datasets. A Galaxy implementation of the package is available within the Workflow4metabolomics.org online infrastructure for computational metabolomics.

Author(s)

Philippe Rinaudo <[e-mail]> and Etienne Thevenot <[e-mail]>.

Maintainer: Philippe Rinaudo <[e-mail]>

Examples

## loading the diaplasma dataset

data(diaplasma)
attach(diaplasma)

## restricting to a smaller dataset for this example

featureSelVl <- variableMetadata[, "mzmed"] >= 490 & variableMetadata[, "mzmed"] < 500
dataMatrix <- dataMatrix[, featureSelVl]
variableMetadata <- variableMetadata[featureSelVl, ]

## signature selection for all 3 classifiers
## a bootI = 5 number of bootstraps is used for this example
## we recommend to keep the default bootI = 50 value for your analyzes

set.seed(123)
diaSign <- biosign(dataMatrix, sampleMetadata[, "type"], bootI = 5)

detach(diaplasma)

biosigner documentation built on Nov. 24, 2020, 2 a.m.