R/recolor.R

In xgaoo/ClusterMap:

####################### Recolor #######################

#' recolor_s
#'
#' Recolor a single sample based on the matching results from cluster_map_by_marker.
#'
#' @import Seurat
#'
#' @param mapRes_sub
#' A vector of the column named by the sample in the output of cluster_map_by_marker, with the regroup column as the vector name.
#' @param obj
#' A Seurat object for the sample.
#' @param output
#' The output directory to save the plot.
#' @param color
#' A vector of colors used to recolor the new groups. DEFAULT is NULL. Pre defined internal color will be used.
#' @param reduction
#' Select the reduction of "tsne", "umap", or "pca" that used for the recolor image.
#' @return A vector of new group labels with the cell name as the vector name.
#' @export


recolor_s <- function(mapRes_sub, obj, output, color = NULL, reduction="tsne")
{
	## recolor_s will call function gg_colr_hue.
	message(paste0("recolor ", output))

	if (is.null(names(mapRes_sub))) stop("There is no name of mapRes_sub.")
    l <- lapply(strsplit(mapRes_sub, ';'), sub, pattern = '.*_', replacement = '')
    new_match <- setNames(unlist(l, use.names = F), rep(names(l), lengths(l)))
    if(obj@version > 3){
	print("Using Seurat v3")
        new_group <- Idents(object = obj)
        levels(new_group) <- names(new_match)[match(levels(Idents(object=obj)), new_match)]
    }
    else{
        new_group <- obj@ident
        levels(new_group) <- names(new_match)[match(levels(obj@ident), new_match)]
    }
    new_group <- factor(new_group, levels = names(mapRes_sub))
	## t-SNE plot
    obj@meta.data$regroup <- new_group
	
	if(obj@version < 3){	
	if (is.null(color)) color <- gg_color_hue(length(levels(new_group)))
    png(paste0(output, '.recolor.tsne.png'))
		TSNEPlot(obj, do.label = T, label.size = 8, group.by = 'regroup',
			colors.use = color[sort(as.numeric(unique(new_group)))], plot.title = toupper(output))
    dev.off()
	pdf(paste0(output, '.recolor.tsne.pdf'))
		TSNEPlot(obj, do.label = T, label.size = 8, group.by = 'regroup',
			colors.use = color[sort(as.numeric(unique(new_group)))], plot.title = toupper(output))
    dev.off()
    return(new_group)
}

   else{	
	   print("Using Seurat v3")
	   if (is.null(color)) color <- gg_color_hue(length(levels(new_group)))

	p3 <- DimPlot(obj, label = T, label.size = 8, group.by = 'regroup',
			reduction = reduction,
			cols = color[sort(as.numeric(unique(new_group)))])
    		ggtitle(paste(toupper(output))) 
		ggsave(plot = p3, filename = paste0(output, '.recolor.', reduction, '.png'))
		ggsave(plot = p3, filename = paste0(output, '.recolor.', reduction, '.pdf'))

    return(new_group)
	}	   
}
#' recolor_comb
#'
#' Recolor the combined sample based on the matching results from recolor_s.
#'
#' @import Seurat
#'
#' @param comb_obj
#' A Seurat object for the combined sample. Cells in different samples are labelled by the sample names with the comb_delim.
#' @param new_group_list
#' A list of vectors of new group assignment outputted from recolor_s.
#' @param output
#' The output directory to save the plot.
#' @param comb_delim
#' The delimiter used in the cell names in the combined object to connect sample name and cell name in individual sample. DEFAULT is '-'.
#' @param color
#' A vector of colors used to recolor the new groups. DEFAULT is NULL. Pre defined internal color will be used.
#' @param reduction
#' Select the reduction of "tsne", "umap", or "pca" that used for the recolor image.
#' @return A vector of new group labels with the cell name as the vector name.
#' @export


recolor_comb <- function(comb_obj, new_group_list, output, comb_delim = '-', color = NULL, reduction="tsne")
{
	## Change comb_delim if v3 Seurat
	if(comb_obj@version > 3){
		comb_delim = '_'
		print("Changed comb_delim to '_'")
	}

	## recolor_comb will call function gg_color_hue.
	message(paste0("recolor ", output))

    sample_label <- as.factor(sub(paste0(comb_delim, '.*'), '', colnames(GetAssayData(object = comb_obj))))
	message("levels(sample_label):")
	print(levels(sample_label))
	message("names(new_group_list):")
	print(names(new_group_list))
    if (all(levels(sample_label) == names(new_group_list)) == FALSE)
		stop("Sample label in comb_obj doesn't match names(new_group_list) or names(single_obj_list).")
	names(sample_label) <- colnames(GetAssayData(object = comb_obj))
	## color by samples
	comb_obj$samples <- sample_label

	if (comb_obj@version <3) {
		print("Seurat v2")
	png(paste0(output, '.color.by.sample.tsne.png'))
		TSNEPlot(comb_obj, do.label = F, label.size = 8, group.by = 'samples', plot.title = 'Colored by sample')
    dev.off()
	pdf(paste0(output, '.color.by.sample.tsne.pdf'))
		TSNEPlot(comb_obj, do.label = F, label.size = 8, group.by = 'samples', plot.title = 'Colored by sample')
    dev.off()
	## assign new group
    new_group <- unlist(new_group_list)
	names(new_group) <- sub('\\.', comb_delim, names(new_group))
    new_group <- factor(new_group, levels = levels(new_group_list[[1]]))
    new_group <- new_group[match(colnames(GetAssayData(object = comb_obj)), as.vector(names(new_group)))] ## some cells may be filtered out in combined sample.
	if (is.na(new_group[1]))
		stop("Cell names in comb_obj don't match cell names in new_group_list or single_obj_list. Cell names in comb_obj should be sample name and cell name in individual sample connected by comb_delim.")
	names(new_group) <- colnames(GetAssayData(object = comb_obj))
	## color by new group
    comb_obj <- AddMetaData(object = comb_obj, metadata = new_group, col.name = "regroup")
	if (is.null(color)) color  <-  gg_color_hue(length(levels(new_group)))
    png(paste0(output, '.recolor.tsne.png'))
		TSNEPlot(comb_obj, do.label = T, label.size = 8, group.by = 'regroup',
			colors.use = color[sort(as.numeric(unique(new_group)))], plot.title = 'Combined')
    dev.off()
    pdf(paste0(output, '.recolor.tsne.pdf'))
		TSNEPlot(comb_obj, do.label = T, label.size = 8, group.by = 'regroup',
			colors.use = color[sort(as.numeric(unique(new_group)))], plot.title = 'Combined')
    dev.off()
    return(new_group)
		}
else if(comb_obj@version > 3){
	print("Seurat v3 comb_obj")
		p5 <- DimPlot(comb_obj, label = F, label.size = 8, group.by = 'samples', 
			reduction = reduction) + 
			ggtitle('Colored by sample') 
    			ggsave(plot = p5, filename = paste0(output, '.color.by.sample.', reduction, '.png'))
		p4 <- DimPlot(comb_obj, label = F, label.size = 8, group.by = 'samples',
			reduction = reduction) +
			ggtitle('Colored by sample')  
    			ggsave(plot = p4, filename = paste0(output, '.color.by.sample.', reduction, '.pdf'))
	## assign new group
    new_group <- unlist(new_group_list)
	names(new_group) <- sub('\\.', comb_delim, names(new_group))
    new_group <- factor(new_group, levels = levels(new_group_list[[1]]))
    new_group <- new_group[match(colnames(GetAssayData(object = comb_obj)), 
				 as.vector(names(new_group)))] ## some cells may be filtered out in combined sample.
	if (is.na(new_group[1]))
		stop("Cell names in comb_obj don't match cell names in new_group_list or single_obj_list. Cell names in comb_obj should be sample name and cell name in individual sample connected by comb_delim.")
	names(new_group) <- colnames(GetAssayData(object = comb_obj))
	## color by new group
    comb_obj <- AddMetaData(object = comb_obj, metadata = new_group, col.name = "regroup")
	if (is.null(color)) color  <-  gg_color_hue(length(levels(new_group)))
		plot1 <- DimPlot(comb_obj, label = T, label.size = 8, 
			reduction = reduction, group.by = 'regroup',
			cols = color[sort(as.numeric(unique(new_group)))]) +
			ggtitle('Combined') 
    			ggsave(plot = plot1, filename = paste0(output, '.recolor.', reduction, '.png'))
		plot2 <- DimPlot(comb_obj, label = T, label.size = 8,
			reduction = reduction, group.by = 'regroup',
			cols = color[sort(as.numeric(unique(new_group)))]) + 
			ggtitle('Combined') 
    			ggsave(plot = plot2, filename = paste0(output, '.recolor.', reduction, '.pdf'))
    return(new_group)
}
		
}