mcc: Compute a Mathews Correlation Coefficient
In PharmacoGx: Analysis of Large-Scale Pharmacogenomic Data

Description Usage Arguments Details Value Examples

The function computes a Matthews correlation coefficient for two factors provided to the function. It assumes each factor is a factor of class labels, and the enteries are paired in order of the vectors.

1	mcc(x, y, nperm = 1000, nthread = 1)

`x`	`factor` of the same length with the same number of levels
`y`	`factor` of the same length with the same number of levels
`nperm`	`numeric` number of permutations for significance estimation. If 0, no permutation testing is done
`nthread`	`numeric` can parallelize permutation texting using BiocParallels bplapply

Please note: we recommend you call set.seed() before using this function to ensure the reproducibility of your results. Write down the seed number or save it in a script if you intend to use the results in a publication.

A list with the MCC as the $estimate, and p value as $p.value

1
2
3

x <- factor(c(1,2,1,2,3,1))
y <- factor(c(2,1,1,1,2,2))
mcc(x,y)

Warning message:
no DISPLAY variable so Tk is not available 
$estimate
[1] -0.452267

$p.value
[1] 0.8

Warning message:
In mcc(x, y) :
  The number of levels x and y was different. Taking the union of all class labels.