update doc

function/EnhancedVolcano.R

@@ -3,6 +3,51 @@
 # Github: https://github.com/kevinblighe/EnhancedVolcano
 # Modified by Franck Soubès (add 74-85;  modify 98-107, add 3 parameters displayLab, findfamily, topgenes)
 
+#' Volcano plots represent a useful way to visualise the results of differential expression analyses. Here, we present a highly-configurable function that
+#'produces publication-ready volcano plots. EnhancedVolcano will attempt to fit as many transcript names in the plot window as possible,
+#' thus avoiding 'clogging' up the plot with labels that could not otherwise have been read.
+#'
+#' @param toptable  Requires at least the following: column for transcript names (can be rownames); a column for log2 fold changes; a column for nominal or adjusted p-value. REQUIRED.
+#' @param lab A column name in toptable containing transcript names. Can be rownames(toptable). REQUIRED.
+#' @param x A column name in toptable containing log2 fold changes. REQUIRED.
+#' @param y A column name in toptable containing nominal or adjusted p-values. REQUIRED.
+#' @param selectLab  A vector containing a subset of lab. DEFAULT = NULL. OPTIONAL.
+#' @param displaylab A comma-separated values corresponding to the displayed genes. DEFAULT = NULL. OPTIONAL.
+#' @param findfamily A character to parse familiy of genes. DEFAULT = NULL. OPTIONAL.
+#' @param topgenes A numeric value to display the top n genes based on the regulation.DEFAULT = NULL. OPTIONAL.
+#' @param regulationvolc A character for the regulation ("up", "down", "both") .DEFAULT = NULL. OPTIONAL.
+#' @param xlim Limits of the x-axis. DEFAULT = c(min(toptable[,x], na.rm=TRUE),max(toptable[,x], na.rm=TRUE)). OPTIONAL.
+#' @param ylim Limits of the y-axis. DEFAULT = c(0, max(-log10(toptable[,y]), na.rm=TRUE) + 5). OPTIONAL.
+#' @param xlab Label for x-axis. DEFAULT = bquote(Log[2]~ "fold change"). OPTIONAL.
+#' @param ylab Label for y-axis. DEFAULT = bquote(-Log[10]~ P)).OPTIONAL.
+#' @param axisLabSize Size of x- and y-axis labels. DEFAULT = 18. OPTIONAL.
+#' @param pCutoff Cut-off for statistical significance. A horizontal line will be drawn at -log10(pCutoff). DEFAULT = 10e-6. OPTIONAL.
+#' @param pLabellingCutoff Labelling cut-off for statistical significance. DEFAULT = pCutoff. OPTIONAL
+#' @param FCcutoff Cut-off for absolute log2 fold-change. Vertical lines will be drawn at the negative and positive values of log2FCcutoff. DEFAULT =1.0. OPTIONAL.
+#' @param title Plot title. DEFAULT = 'Volcano plot'. OPTIONAL.
+#' @param titleLabSize Plot subtitle. DEFAULT = 'Bioconductor package, EnhancedVolcano'. OPTIONAL.
+#' @param transcriptPointSize Size of plotted points for each transcript. DEFAULT = 0.8. OPTIONAL.
+#' @param transcriptLabSize Size of labels for each transcript. DEFAULT = 3.0. OPTIONAL.
+#' @param col Colour shading for plotted points, corresponding to< abs(FCcutoff) && > pCutoff, > abs(FCcutoff), < pCutoff,> abs(FCcutoff) && < pCutoff. DEFAULT = c("grey30", "forestgreen","royalblue", "red2"). OPTIONAL.
+#' @param colAlpha Alpha for purposes of controlling colour transparency oftranscript points. DEFAULT = 1/2. OPTIONAL.
+#' @param legend Plot legend text. DEFAULT = c("NS", "Log2 FC", "P","P & Log2 FC"). OPTIONAL.
+#' @param legendPosition Position of legend ("top", "bottom", "left","right"). DEFAULT = "top". OPTIONAL.
+#' @param legendLabSize Size of plot legend text. DEFAULT = 14. OPTIONAL.
+#' @param legendIconSize Size of plot legend icons / symbols. DEFAULT = 4.0.OPTIONAL.
+#' @param DrawConnectors Logical, indicating whether or not to connect plotlabels to their corresponding points by line connectors. DEFAULT = FALSE.OPTIONAL.
+#' @param widthConnectors Line width of connectors. DEFAULT = 0.5. OPTIONAL.
+#' @param colConnectors Line colour of connectors. DEFAULT = 'grey10'. OPTIONAL.
+#' @param cutoffLineType Line type for FCcutoff and pCutoff ("blank","solid", "dashed", "dotted", "dotdash", "longdash", "twodash").DEFAULT = "longdash". OPTIONAL.
+#' @param cutoffLineCol Line colour for FCcutoff and pCutoff. DEFAULT ="black". OPTIONAL.
+#' @param cutoffLineWidth Line width for FCcutoff and pCutoff. DEFAULT = 0.4. OPTIONAL.
+#'
+#' @author Kevin Blighe <[email protected]>
+#'
+#' @return A list of two elements
+#' @export
+#'
+#' @examples
+
 EnhancedVolcano <- function(
     toptable,
     lab,
@@ -38,7 +83,7 @@ EnhancedVolcano <- function(
     cutoffLineType = "longdash",
     cutoffLineCol = "black",
     cutoffLineWidth = 0.4)
-    
+
 {
     if(!requireNamespace("ggplot2")) {
         stop("Please install ggplot2 first.", call.=FALSE)
@@ -60,7 +105,7 @@ EnhancedVolcano <- function(
     requireNamespace("ggrepel")
     requireNamespace("dplyr")
     i <- xvals <- yvals <- Sig <- NULL
-    
+
     toptable <- as.data.frame(toptable)
     toptable$GeneName <- sapply(toptable$GeneName, function(v) {
       if (is.character(v)) return(toupper(v))
@@ -74,9 +119,9 @@ EnhancedVolcano <- function(
         (abs(toptable[,x])>FCcutoff)] <- "FC_P"
     toptable$Sig <- factor(toptable$Sig,
         levels=c("NS","FC","P","FC_P"))
-    
-  
-    
+
+
+
     if(is.na(topgenes) && !is.na(displaylab) ){
       selectLab <- as.character(displaylab)
     }
@@ -94,32 +139,32 @@ EnhancedVolcano <- function(
       else{
         toptable$abs <- unlist((toptable[x]))
       }
-        
+
       toptable$X <- rownames(toptable)
-      myval <- toptable %>%   dplyr::filter(Sig =="FC_P") %>% dplyr::select(GeneName,X,abs)  %>%  
+      myval <- toptable %>%   dplyr::filter(Sig =="FC_P") %>% dplyr::select(GeneName,X,abs)  %>%
       {if (regulationvolc == "down") top_n(.,-topgenes) else top_n(.,topgenes)}
       myvalueind <- myval$X
       selectLab <- as.character(myval$GeneName)
-      
+
     }
-  
-      
+
+
     if (min(toptable[,y], na.rm=TRUE) == 0) {
         warning("One or more P values is 0. Converting to minimum possible value...", call. = FALSE)
         toptable[which(toptable[,y] == 0), y] <- .Machine$double.xmin
     }
-    
+
     toptable$lab <-  sapply(toptable$GeneName, function(v) {
       if (is.character(v)) return(toupper(v))
       else return(v)
     })
-    
-    
-    
+
+
+
     toptable$xvals <- toptable[,x]
     toptable$yvals <- toptable[,y]
-  
-    
+
+
    if (!is.null(selectLab)) {
     if(!is.na(topgenes) && is.na(displaylab)&& is.na(findfamily)){
     names.new <- rep("", length(toptable$lab))
@@ -134,15 +179,15 @@ EnhancedVolcano <- function(
         toptable$lab <- names.new
       }
     }
-    
+
     subdata = subset(toptable,
                      toptable[,y]<pLabellingCutoff &
-                       abs(toptable[,x])>FCcutoff) 
+                       abs(toptable[,x])>FCcutoff)
+
+
+
+
 
-
-
-
-
     plot <- ggplot2::ggplot(toptable,
             ggplot2::aes(x=xvals, y=-log10(yvals))) +
 
@@ -207,9 +252,9 @@ EnhancedVolcano <- function(
             linetype=cutoffLineType,
             colour=cutoffLineCol,
             size=cutoffLineWidth)
-   
-    
-    
+
+
+
     if (DrawConnectors == TRUE) {
         plot <- plot + ggrepel::geom_text_repel(max.iter = 100,
             data=subdata ,
@@ -237,7 +282,7 @@ EnhancedVolcano <- function(
                 check_overlap = F,
                 vjust = 1.0)
     }
-    
+
     if(!is.na(topgenes)) subdata <- filter(subdata, lab != "")
     mylist = list(plot, subdata)
     return(mylist)

function/PCA.R

@@ -8,36 +8,36 @@
 
 #' res.pca is a function that computed a PCA of non-normalized data with the FactoMineR package
 #'
-#' @param workingset a data frame corresponding to the WorkingSet
-#' @param scale a boolean; by default this value is set to False non-normalized data
+#' @param workingset A data frame corresponding to the WorkingSet
+#' @param scale A boolean value; by default this value is set to False for non-normalized data
+#'
+#' @return A data frame with PCA attributes
 #'
-#' @return \PCAres a data frame with PCA attributes
-#' 
 #' @export
 
 res.pca <- function(workingset, scale = F) {
-  
+
   myt = transpose(workingset)
   row.names(myt) = colnames(workingset)
-  
+
   PCAres = PCA(myt,
                scale.unit = F,
                graph = F)
-  
+
   return(PCAres)
 }
 
 
 #' eboulis is a function which aim is to display the eigenvalues of the data with the package factoextra
 #'
-#' @param PCAresa a data frame with PCA attributes
+#' @param PCAres A data frame with PCA attributes
+#'
+#' @return A ggplot object
 #'
-#' @return \p a factoextra object
-#' 
 #' @export
 
 eboulis <- function(PCAres){
-  
+
   p <- fviz_eig(PCAres, addlabels=TRUE, hjust = -0.3, barfill="white", barcolor ="darkblue", linecolor ="red")
   p + theme(panel.grid.major = element_blank(),  panel.grid.minor = element_blank(),panel.border = element_blank(),
             panel.background = element_blank())
@@ -48,45 +48,52 @@ eboulis <- function(PCAres){
 
 #' PCAplot is a function that return a factoextra object of PCA type
 #'
-#' @param PCAres a data frame with PCA attributes
-#' @param myax a numeric vector of length 2 specifying the dimensions to be plotted
-#' @param elips a boolean value to add ellipse to the data distribution  for the different groups; default = False
-#' @param rep a boolean value to avoid overlaps between the label points
-#' @param mylevel a data frame corresponding to the pData
-#' @param mylabsize a numeric value representing the police size to display for the different labels
-#' @param dispelip a numeric value representing the ellipsoid dispersion
-#' @param labeled a character to display labels and/or points
-#' @param pal a color object from the RcolorBrewer package
+#' @param PCAres A data frame with PCA attributes
+#' @param myax A numeric vector of length 2 specifying the dimensions to be plotted
+#' @param elips A boolean value to add ellipse to the data distribution  for the different groups; default = False
+#' @param rep A boolean value to avoid overlaps between the label points
+#' @param mylevel A data frame corresponding to the pData
+#' @param mylabsize A numeric value representing the police size to display for the different labels
+#' @param dispelip A numeric value representing the ellipsoid dispersion
+#' @param labeled A character to display labels and/or points
+#' @param pal A color object from the RcolorBrewer package
+#'
+#' @return A ggplot object
 #'
-#' @return
-#' 
 #' @export
 
 PCAplot <- function(PCAres, myax = c(1,2), elips = T , rep = T , mylevel = groups$Grp,  mylabsize = 4, dispelip = 0.8 , labeled = 'all', pal = brewer.pal(8, "Dark2")){
-  
-  
+
+
   p <- fviz_mca_ind(PCAres, label= labeled , habillage = mylevel, addEllipses= elips ,
                     ellipse.level= 0.8, repel = rep, axes = myax, pointsize = 2 , labelsize = mylabsize)
 
-  
+
   return(p + scale_color_manual(values=pal))
 }
 
 
 # res.pca <- function(workingset, restab ,scale = F, variable = F) {
-#   
+#
 #   myt = transpose(workingset)
 #   row.names(myt) = colnames(workingset)
-#   
+#
 #   if(variable)
 #     colnames(myt) =  make.names(restab$GeneName, unique=TRUE)
-#   
+#
 #   PCAres = PCA(myt,
 #                scale.unit = F,
 #                graph = F)
-#   
+#
 #   return(PCAres)
 # }
 
 
 
+
+
+
+
+
+
+

function/cutheat.R

@@ -9,23 +9,22 @@
 #' cutHeatmaps if a function that takes as input an heatmap object and depending on the cut height and the cluster
 #' choosen render a ggplot object or an heatmap object
 #'
-#' @param hmp an heatmap object
-#' @param height a numeric value to cut the dendogram
-#' @param exprData a data frame with specific columns depending on the user's choices
-#' @param DEGres a data frame corresponding to the xxx_topTableAll
-#' @param groups a data frame of the choosen groups
-#' @param cexcol a positive numbers, used as cex.axis in for the row or column axis labeling
-#' @param cexrow a positive numbers, used as cex.axis in for the row or column axis labeling
-#' @param labrow a character vectors with row and column labels to use
-#' @param fileType a character to select the plot to display heatmap, boxplot or stripchart
-#' @param scale a character indicating if the values should be centered and scaled in either the row direction or the column direction, or none
-#' @param meanGrp a boolean value to computes the mean for each groups; default = F
-#' @param col.hm a character vector
-#' @param type a character to select the plot to display heatmap, boxplot or stripchart
-#' @param las a numeric value
-#' @param distfun function used to compute the distance (dissimilarity) between both rows and columns.
-#' @param palette.col a character vector
-#' @param num an item of the heatmap object corresponding to a specific cluster choosen by the user
+#' @param hmp An heatmap object
+#' @param height A numeric value to cut the dendogram
+#' @param exprData A data frame with specific columns depending on the user's choices
+#' @param groups A data frame of the choosen groups
+#' @param cexcol A positive numbers, used as cex.axis in for the row or column axis labeling
+#' @param cexrow A positive numbers, used as cex.axis in for the row or column axis labeling
+#' @param labrow A character vectors with row and column labels to use
+#' @param fileType A character to select the plot to display heatmap, boxplot or stripchart
+#' @param meanGrp A boolean value to computes the mean for each groups; default = F
+#' @param type A character to select the plot to display heatmap, boxplot or stripchart
+#' @param las A numeric value
+#' @param distfun Function used to compute the distance (dissimilarity) between both rows and columns.
+#' @param palette.col A character vector of colors
+#' @param num An item of the heatmap object corresponding to a specific cluster choosen by the user
+#' @param genename A character vector of gene symbols
+#' @param scales A character indicating if the values should be centered and scaled in either the row direction or the column direction, or none. The default is "none"
 #' @param ...
 #'
 #' @return a ggplot object or heatmapply object
@@ -92,7 +91,7 @@ cutHeatmaps = function(hmp, height, exprData, groups, cexcol = 1, cexrow = 1, la
   ###=======================
   ## cut the heatmap
   ###=======================
-  
+
   # Cut the dendogram in 2 part with the desired height
 
   cut02 = cut(hmp$rowDendrogram, h = height)
@@ -114,7 +113,7 @@ cutHeatmaps = function(hmp, height, exprData, groups, cexcol = 1, cexrow = 1, la
 
     ## scaling
 #		HCgroupsLabExrsCenterScale <- ifelse(scales=="row",lapply(HCgroupsLabExrs,function(y){t(scale(t(y),center=T,scale=T))}),HCgroupsLabExrs)
-    
+
     if(scales=="row"){
       HCgroupsLabExrsCenterScale <- lapply(HCgroupsLabExrs,function(y){t(scale(t(y),center=T,scale=T))})
     }else HCgroupsLabExrsCenterScale <- HCgroupsLabExrs
@@ -252,7 +251,7 @@ cutHeatmaps = function(hmp, height, exprData, groups, cexcol = 1, cexrow = 1, la
                 sep = "")
 
         if (!probes.boxplot) {
-          
+
           ##=============
           ## plot stripchart
           ##
@@ -293,7 +292,7 @@ cutHeatmaps = function(hmp, height, exprData, groups, cexcol = 1, cexrow = 1, la
               title = paste("Cluster", i),
               y = "Expression Z-score",
               caption = footnote
-            ) + 
+            ) +
             theme(
               plot.title = element_text(size = 20, hjust = 0.5),
               plot.caption = element_text(size = 10, hjust = 0.5),
@@ -344,7 +343,7 @@ cutHeatmaps = function(hmp, height, exprData, groups, cexcol = 1, cexrow = 1, la
               title = paste("Cluster", i),
               y = "Expression Z-score",
               caption = footnote
-            ) + 
+            ) +
             theme(
               plot.title = element_text(size = 24, hjust = 0.5),
               axis.title.x = element_text(size = 10),
@@ -379,7 +378,7 @@ cutHeatmaps = function(hmp, height, exprData, groups, cexcol = 1, cexrow = 1, la
             if(length(labels(cut02$lower[[i]]))>1){
             rowIds=NA;
 
-            
+
             useRasterTF=T;
             hm02gp=heatmap(exprData[labels(cut02$lower[[1]]),], Rowv=str(cut02$lower[[1]]),
             Colv=hmp$colDendrogram,