transcriptR (version 1.0.2)

predictTssOverlap: predictTssOverlap

Description

The function classifies ChIP-seq peaks on gene associated and background using classification model based on a logistic regression.

Usage

predictTssOverlap(object, feature, p = 0.8)
## S3 method for class 'ChipDataSet':
predictTssOverlap(object, feature, p = 0.8)

Arguments

object

A ChipDataSet object.

feature

Character. Specify feature(s) that are be used as predictors in the classification model. By default, all the features stored in the ChipDataSet object will be used.

p

Numeric. The proportion of data that is used to train the model. Default: 0.8.

Value

• The slot tssOverlapPrediction of the provided TranscriptionDataSet object will be updated by the the following elements: 'predicted.tssOverlap', 'confusionMatrix', 'logFitSummary' and 'roc'.

Details

In order to improve the accuracy of the model the data is internally partitioned into a training and testing data sets. A repeated 10-Fold Cross-Validation is used to calculate performance measures on the training data set and to prevent over-fitting.

The model fit and validation is internally acomplised by the functions implemented in the http://topepo.github.io/caret/index.html{caret} package.

See Also

ChipDataSet constructCDS http://topepo.github.io/caret/index.html{caret}

Examples

### Load ChipDataSet object
data(cds)

predictTssOverlap(object = cds, feature = "pileup", p = 0.75)

### View a short summary of the gene associated peaks prediction
cds

### View peaks and associated prediction
getPeaks(cds)