# PAC: PAC score In jokergoo/cola: A Framework for Consensus and Hierarchical Partition

PAC score

## Usage

 ```1 2``` ```PAC(consensus_mat, x1 = seq(0.1, 0.3, by = 0.02), x2 = seq(0.7, 0.9, by = 0.02), trim = 0.2) ```

## Arguments

 `consensus_mat` a consensus matrix. `x1` lower bound to define "ambiguous clustering". The value can be a vector. `x2` upper bound to define "ambihuous clustering". The value can be a vector. `trim` percent of extreme values to trim if combinations of `x1` and `x2` are more than 10.

## Details

This a variant of the orignial PAC (proportion of ambiguous clustering) method.

For each `x_1i` in `x1` and `x_2j` in `x2`, `PAC_k = F(x_2j) - F(x_1i)` where `F(x)` is the ecdf of the consensus matrix (the lower triangle matrix without diagnals). The final PAC is the mean of all `PAC_k` by removing top `trim/2` percent and bottom `trim/2` percent of all values.

## Value

A single numeric score.

## See

See https://www.nature.com/articles/srep06207 for explanation of PAC score.

## Author(s)

Zuguang Gu <[email protected]>

## Examples

 ```1 2 3 4 5 6``` ```data(cola_rl) PAC(get_consensus(cola_rl[1, 1], k = 2)) PAC(get_consensus(cola_rl[1, 1], k = 3)) PAC(get_consensus(cola_rl[1, 1], k = 4)) PAC(get_consensus(cola_rl[1, 1], k = 5)) PAC(get_consensus(cola_rl[1, 1], k = 6)) ```

jokergoo/cola documentation built on Nov. 13, 2018, 1:22 p.m.