predict_precision_recall: Predict precision-recall

In neurogenomics/EpiCompare: Comparison, Benchmarking & QC of Epigenomic Datasets

Predict precision-recall

Description

Predict specific values of precision or recall by fitting a model to a precision-recall curve. Predictions that are <0 will automatically be set to 0. Predictions that are >100 will automatically be set to 100.

Usage

predict_precision_recall(
  pr_df,
  fun = stats::loess,
  precision = seq(10, 100, 10),
  recall = seq(10, 100, 10)
)

Arguments

pr_df	Precision-recall data.frame generated by precision_recall.
fun	Function to fit the data with.
precision	Precision values to predict recall from.
recall	Recall values to predict precision from.

Value

A named list of fitted models and predictions.

Source

Fix for producing NAs from loess fun.

Examples

data("CnR_H3K27ac")
data("CnT_H3K27ac")
data("encode_H3K27ac")
peakfiles <- list(CnR_H3K27ac=CnR_H3K27ac, CnT_H3K27ac=CnT_H3K27ac)
reference <- list("encode_H3K27ac" = encode_H3K27ac)
pr_df <- precision_recall(peakfiles = peakfiles,
                          reference = reference)
predictions <- predict_precision_recall(pr_df = pr_df)           

neurogenomics/EpiCompare documentation built on April 30, 2024, 3:58 p.m.