ComplexHeatmap|根据excel表绘制突变景观图（oncoplot）

如果有maf格式的文件，可以直接oncoplot包绘制瀑布图，有多种展示和统计maftools | 从头开始绘制发表级oncoplot（瀑布图）和maftools|TCGA肿瘤突变数据的汇总，分析和可视化，如果只有多个样本的基因突变与否的excel，不用担心，也可以用complexheatmap包绘制。

这个包功能很强大，本次只简单的介绍如何绘制基因组景观图（瀑布图）。

一载入R包，数据

#if (!requireNamespace("BiocManager", quietly = TRUE))
#  install.packages("BiocManager")
#BiocManager::install("ComplexHeatmap")
#install.packages("openxlsx")
#install.packages("circlize")

#后面直接加载即可
library(openxlsx)
library(ComplexHeatmap)
library(circlize)
#读入数据
mut <- read.xlsx("TCGA_data.xlsx",sheet = "突变信息")
cli <- read.xlsx("TCGA_data.xlsx",sheet = "临床信息")

查看变异数据

rownames(mut) <- mut$sample
mat <- mut[,-1]
mat[is.na(mat)]<-""
mat[1:6,1:6]

二绘制oncoplot图

2.0 绘制“初始”瀑布图

oncoPrint(mat)

可以展示结果，但是为了paper，还需要一些调整！

2.1 指定变异类型的颜色和形状大小

#指定颜色, 调整颜色代码即可
col <- c( "mutation" = "blue" , "indel" = "green")
#指定变异的样子，x,y,w,h代表变异的位置(x,y)和宽度(w)，高度(h)
alter_fun <- list(
  background = function(x, y, w, h) {
    grid.rect(x, y, w-unit(0.5, "mm"), h-unit(0.5, "mm"),
              gp = gpar(fill = "#CCCCCC", col = NA))
  },
  mutation = function(x, y, w, h) {
    grid.rect(x, y, w-unit(0.5, "mm"), h-unit(0.5, "mm"),
              gp = gpar(fill = col["mutation"], col = NA))
  },
  indel = function(x, y, w, h) {
    grid.rect(x, y, w-unit(0.5, "mm"), h*0.33,  
              gp = gpar(fill = col["indel"], col = NA))
  }
)

#指定变异类型的标签，和数据中的类型对应
heatmap_legend_param <- list(title = "Alternations",
                             at = c("mutation","indel"),
                             labels = c( "mutation","indel"))

绘制景观图

#设定标题
column_title <- "This is Oncoplot "  
#画图并去除无突变的样本和基因
oncoPrint(mat,
          alter_fun = alter_fun, col = col,
          column_title = column_title,
          heatmap_legend_param = heatmap_legend_param)

2.2 简单的调整

oncoPrint(mat,
          alter_fun = alter_fun, col = col,
          column_title = column_title,
          remove_empty_columns = TRUE, #去掉空列
          remove_empty_rows = TRUE, #去掉空行
          row_names_side = "left", #基因在左
          pct_side = "right",
          heatmap_legend_param = heatmap_legend_param)

三添加注释

3.1 添加临床注释信息

pdata <- cli
head(pdata)

#对应患者
pdata <- subset(pdata,pdata$sampleID %in% colnames(mat))
mat <- mat[, pdata$sampleID]
#定义注释信息
ha<-HeatmapAnnotation(Age=pdata$age,
                     Gender=pdata$gender, 
                     GeneExp_Subtype  = pdata$GeneExp_Subtype ,
                     censor = pdata$censor,
                     os = pdata$os,
                     show_annotation_name = TRUE,
                     annotation_name_gp = gpar(fontsize = 7))

3.2 瀑布图 + 临床注释

oncoPrint(mat,
          bottom_annotation = ha, #注释信息在底部
          alter_fun = alter_fun, col = col,            remove_empty_columns = TRUE, #去掉空列          remove_empty_rows = TRUE, #去掉空行          column_title = column_title, heatmap_legend_param = heatmap_legend_param )

此处使用默认颜色注释，有时候会比较接近，且“变动”

3.3 自定义注释颜色以及样本顺序

#自定义样本顺序
s <- pdata[order(pdata$censor,pdata$GeneExp_Subtype),]
sample_order <- as.character(s$sampleID)

#自定义颜色
#连续性变量设置颜色（外）
col_os = colorRamp2(c(0, 4000), c("white", "red"))
ha<-HeatmapAnnotation(Age=pdata$age,
                      Gender=pdata$gender,
                      GeneExp_Subtype  = pdata$GeneExp_Subtype ,
                      censor = pdata$censor,
                      os = pdata$os,
                      #指定颜色
                      col = list(censor = c("death" =  "red", "alive" = "blue"),
                                 GeneExp_Subtype = c("Classical" = "orange","Mesenchymal" = "green","Neural" = "skyblue" ),
                                 os = col_os),
                      show_annotation_name = TRUE,
                      annotation_name_gp = gpar(fontsize = 7))

绘制瀑布图

oncoplot_anno = oncoPrint(mat,bottom_annotation = ha,
              alter_fun = alter_fun, col = col,
              column_order = sample_order,
              remove_empty_columns = TRUE, #去掉空列
              remove_empty_rows = TRUE, #去掉空行
              column_title = column_title, heatmap_legend_param = heatmap_legend_param) 
oncoplot_anno

注：颜色不一定好看，只是为了当默认的颜色比较接近时，或者有要求时候，可以自定义。

3.4 调整注释的位置

draw(oncoplot_anno ,annotation_legend_side = "bottom")

更改注释的位置，方便后续拼图需求。

更多参数：

https://github.com/jokergoo/ComplexHeatmap