# extractNegLogPval: Find the negative log p-value of a pair of vectors. In jamesdalg/CNVScope: A Versatile Toolkit for Copy Number Variation Relationship Data Analysis and Visualization

## Description

Finds the negative log p-value of a matrix, if it exists. Checks first to see if there is a p-value to return.

## Usage

 `1` ```extractNegLogPval(x, y, repval = 300, lowrepval = 0, signed = F) ```

## Arguments

 `x` a vector that is regressed in the fashion y~x. `y` a vector that is regressed in the fashion y~x. `repval` the replacement value if the regression cannot be performed, default 300 (the vectors are identical if this is used). `lowrepval` The low replacement value in the case that a regression p-value is undefined. `signed` change the sign of the negative log p-value based on the sign of beta? e.g. if the line has a negative slope, so will the returned value. If there is a positive slope, there will be a positive negative log p-value. if this option is disabled, then no sign changes will happen based on the sign of the slope.

## Value

The negative log p-value or replacement value.

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13``` ```#small example xval<-c(1,1,1,1,1) yval<-c(1,2,3,4,5) a<-c(3,4,5,6,7) extractNegLogPval(x=xval,y=yval) #no possible p-value if one vector is constant. #Some edge cases this may not be correct (if the data lies near a constant), # but the indiviual sample data should reveal true trends. suppressWarnings(cor(xval,yval)) #you can't get a correlation value either. cor(a,a) #gives correlation of 1. extractNegLogPval(a,a) #gives replacement value. suppressWarnings(extractNegLogPval(x=a,y=yval)) #gives 107.3909 and warns about a nearly perfect fit. ```

jamesdalg/CNVScope documentation built on Aug. 4, 2019, 9:24 p.m.