ARMADA.heatmap {armada} | R Documentation |
Heatmap of the selected covariates.
Description
Heatmap of the selected covariates.
Usage
ARMADA.heatmap(X, Y, res.ARMADA.summary, threshold = 5)
Arguments
X |
the matrix (or data.frame) of covariates, dimension n*p (n is the sample size, p the number of covariates). X must have rownames, which are the names of the n subjects (i.e. the user ID of the n subjects). X must have colnames, which are the names of the p covariates. |
Y |
the vector of the response, length n. |
res.ARMADA.summary |
the result of the function ARMADA, or output of the function ARMADA.summary. |
threshold |
an integer between 0 and 8: the selected covariates are those which have a score greater or equal to "threshold." |
Details
This function plots the heatmap of the covariates which have a score higher than some threshold chosen by the user, with respect to the values of Y.
Value
the plot of the heatmap, and a data.frame of the selected covariables.
Examples
library(ClustOfVar)
library(impute)
library(FAMT)
library(VSURF)
library(glmnet)
library(anapuce)
library(qvalue)
library(ComplexHeatmap)
library(circlize)
set.seed(1)
p <- 40
n <- 30
indexRow <- paste0("patient",1:n)
indexCol <- paste0("G",1:p)
X <- matrix(rnorm(p*n),ncol=p)
colnames(X) <- indexCol
rownames(X) <- indexRow
Y <- c(rep(-1,n/2), rep(1,n/2))
X[,1:4] <- X[,1:4] + matrix(rnorm(n*4, mean=2*Y, sd=1), ncol=4)
Y<-as.factor(Y)
resultat <- ARMADA(X,Y, nclust=1)
tracer <- ARMADA.heatmap(X, Y, resultat[[3]], threshold=5)
## Not run:
X<-toys.data$x
Y<-toys.data$Y
result<-ARMADA(X,Y, nclust=2)
select<-ARMADA.heatmap(X, Y, result[[3]], threshold=5)
## End(Not run)