assessGrouping {nearBynding} | R Documentation |
Assess grouping of samples assigned to the same category relative to random.
assessGrouping( distances, annotations, measurement = "mean", output = "KS.pvalue", ctrl_iterations = 10000 )
distances |
Data frame object with at least three columns where the first three columns are sample 1 name, sample 2 name, and the distance between them. |
annotations |
Data frame object with at least two columns where the first two columns are sample name and the category of the sample for grouping. Sample names must match sample 1 and sample 2 names in distances data frame. |
measurement |
The measurement for comparison between cases and controls and statistical analysis ("mean", "max", or "min). Default "mean" |
output |
A string denoting what information will be returned: either a list of test and control measurement distances ("measurements"), the p-value of the Kolmogorov-Smirnov test comparing test and control distributions ("KS.pvalue"), or a ggplot object plotting the test and control distributions ("plot"). Default "KS.pvalue" |
ctrl_iterations |
The number of iterations to test for the control distribution; an integer. Default 10000. |
output = "KS.pvalue" |
the p-value of the Kolmogorov-Smirnov test comparing test and control distributions |
output = "plot" |
a ggplot object plotting the test and control distributions |
output = "measurements" |
a list or test and control measurement distances |
## create random distance data frame dist<-expand.grid(letters, letters) dist$distance<-rnorm(nrow(dist)) annot<-data.frame(sample<-letters, category<- rep(1:13, 2)) ## get KS p-value assessGrouping(dist, annot) ## get plot of test vs control distributions assessGrouping(dist, annot, output = "plot")