binarize {biclust} R Documentation

## Binarize

### Description

Methods to convert a real matrix to a binary matrix.

### Usage

```binarize(x, threshold=NA)
binarizeByPercentage(x,percentage, error=0.2, gap=0.1)
densityOnes(x)
```

### Arguments

 `x` The data matrix to be binarized. `threshold` Threshold used to binarize. Values over threshold will be set to 1, the rest to 0. If threshold is NA, median is used as threshold. Default NA. `percentage` Percentage of ones against zeros desired in the binary matrix. `error` Percentage of ones against zeros in the final matrix will be in [percentage-error, percentage+error]. Default 0.2 `gap` Value used for incremental search of threshold. Default 0.1

### Details

The `binarize` function returns a matrix binarized by input threshold, or by the median if no threshold is given.

The `binarizeByPercentage` function returns a matrix binarize by input percentage, given as desired density of ones against zeros.

The `densityOnes` function returns the percentage of ones against zeros in a logical matrix

### Author(s)

Rodrigo Santamaria rodri@usal.es

### Examples

```  data(BicatYeast)
m1=binarize(BicatYeast)
m2=binarize(BicatYeast, 0.2)
m3=binarizeByPercentage(BicatYeast, 5)
densityOnes(m3)
densityOnes(m2)
densityOnes(m1)
drawHeatmap(BicatYeast)
drawHeatmap(m1)
drawHeatmap(m2)
drawHeatmap(m3)
```

[Package biclust version 2.0.3 Index]