R/plotNullDistribution.R

Defines functions plotNullDistribution

Documented in plotNullDistribution

#' @title Plot permutation p-values distribution
#' @description Create a plot of permutation pvalues with corresponding specified critical vectors.
#' @usage plotNullDistribution(P=NULL,family="simes",alpha = 0.05, 
#' path = getwd(), name = "plot", delta = 0,
#' copes=NULL,mask=NULL, alternative = "two.sided", rand = FALSE, B = 1000)
#' @param P matrix of pvalues with dimensions \eqn{m \times B} used instead of the data matrix \code{X}. Default to @NULL.
#' @param family string character. Choose a family of confidence envelopes to compute the critical vector. 
#' from \code{"simes"}, \code{"aorc"}, \code{"beta"} and \code{"higher.criticism"}. If more than one critical vector are considered, it must be a vector. Default \code{"simes"}.
#' @param alpha numeric value in `[0,1]`. It expresses the alpha level to control the family-wise error rate. Default 0.1.
#' @param path character string. Path to save the plot. The path does not must end with \code{/}. Default to \code{getwd()}.
#' @param name character string. The name of file that will be used to save the plot. Default to name.
#' @param delta numeric value. It expresses the delta value, please see the references. 
#' Default to 0. If more than one critical vector are considered, \code{delta} must be a vector having length equals to the length of critical vectors specified in \code{family}.
#' @param copes list of NIfTI file if \code{P = NULL}. The list of copes, i.e., constrasts maps, one for each subject used to compute the statistical tests.
#' @param mask NIfTI file or character string. 3D array of logical values (i.e. \code{TRUE/FALSE} in/out of the brain). 
#' Alternatively it may be a (character) NIfTI file name. If \code{mask=NULL}, it is assumed that non of the voxels have to be excluded.
#' @param alternative character string. It refers to the alternative hypothesis, must be one of \code{"two.sided"} (default), \code{"greater"} or \code{"lower"}.
#' @param rand Boolean value. Default @FALSE. If \code{rand = TRUE}, the p-values are computed by \code{\link{rowRanks}}.
#' @param B numeric value. Number of permutations, default to 1000. 
#' @author Angela Andreella
#' @return Save a plot in \code{path} with name specified in \code{name} describing the p-values null distribution with critical value curve and observed pvalues in red.
#' @export
#' @importFrom grDevices png
#' @importFrom grDevices dev.off
#' @importFrom graphics plot
#' @importFrom graphics lines
#' @importFrom grDevices rainbow
#' @importFrom graphics legend
#' @examples 
#' \dontrun{
#'db <- simulateData(pi0 = 0.8, m = 100, n = 20, rho = 0)
#'out <- signTest(X = db)
#'pv <- cbind(out$pv, out$pv_H0)
#'plotNullDistribution(P = pv)
#' }
#' 
plotNullDistribution <- function(P=NULL,family="simes",alpha = 0.05, path = getwd(), name = "plot", delta = 0,copes=NULL,mask=NULL, alternative = "two.sided", rand = FALSE, B = 1000){
  
  family_set <- c("simes", "aorc", "beta", "higher.criticism")
  fam_match <- function(x) {match.arg(tolower(x), family_set)}
  alternative_set <- c("two.sided", "greater", "lower")
  alternative <- match.arg(tolower(alternative), alternative_set)
  if(!is.null(family)){family <- unlist(lapply(family, fam_match))}
  if(is.null(copes) & is.null(P)){stop('Please insert pvalues matrix or copes images')}
  
  if(!is.null(copes)){
    
    if(is.null(mask)){stop('please insert the mask as character path or Nifti image')}
    if(is.character(mask)){mask = readNifti(mask)}
    if(!is.list(copes)){stop("Please insert the list of copes as list class object")}
    
    img_dims <- c(91,  109 , 91)
    img <- array(NA, c(img_dims, length(copes)))
    
    for (sid in 1:length(copes)) {  
      img[,,,sid] <- copes[[sid]]
      
    }
    
    scores <- matrix(img,nrow=(91*109*91),ncol=length(copes))
    scores <- scores[which(mask==1),]
    res <- signTest(X=scores, B = B,alternative = alternative, rand = rand) #variables times number of permutation
    
    P <- cbind(res$pv,res$pv_H0)
    rm(res)
    rm(scores)
    rm(copes)
    rm(img)
  
    
  }

  if(is.null(family)){
    if(is.unsorted(P[,1])){pvalues_ord <- colSortC(P)}else{pvalues_ord <- P}
    
    png(paste0(path,"/", name, ".png")) 
    plot(pvalues_ord[,1], type = 'l', col = ' red', xlab = expression(i), ylab = expression(p[(i)]))
    for(i in 2:ncol(pvalues_ord)){
      
      lines(pvalues_ord[,i],col='black',type="l")
      
    }
    lines(pvalues_ord[,1], lwd =2, col= 'red')
    dev.off()
  }else{
    if(is.unsorted(P[,1])){pvalues_ord <- colSortC(P)}else{pvalues_ord <- P}
    
    lcv <- function(family,delta=NULL, cols = "blue"){
      lambdaO <- lambdaOpt(pvalues = P,family=family,alpha=alpha, delta = delta)
      cvO<- criticalVector(pvalues = P, family = family, alpha = alpha, lambda = lambdaO, delta = delta)
      lines(cvO, lwd =2, col= cols)
    }
    firstup <- function(x) {
      substr(x, 1, 1) <- toupper(substr(x, 1, 1))
      x
    }
    lcvV <- Vectorize(lcv,vectorize.args = c("family", "delta", "cols"))
    cols = rainbow(length(family))
    png(paste0(path,"/", name, ".png")) 
    plot(pvalues_ord[,1], type = 'l', col = ' red', lty = "dashed", xlab = expression(i), ylab = expression(p[(i)]))
    for(i in 2:ncol(pvalues_ord)){
      
      lines(pvalues_ord[,i],col='black',type="l")
      
    }
    lines(pvalues_ord[,1], lwd =2, col= 'red', lty = "dashed")
    #lines(cvO, col= 'blue', lwd =2)
    mapply(lcv, family, delta, cols)
    family <- firstup(family)
    family <- ifelse(family == "Aorc", "AORC", family)
    legend('top',legend=c(sapply(c(1:length(family)), 
                                 function(x) as.expression(bquote(~ .(family[x]) ~ delta == .(delta[x]) ))), 
                          " Observed Pvalues"), col= c(cols, "red"),lwd =2, lty =c(rep("solid", length(family)), "dashed"))
    
    dev.off()
  }
  
  
}
angeella/ARIpermutation documentation built on Jan. 12, 2022, 1:01 p.m.