ComplexHeatmap 绘制肿瘤突变分布图

来源：https://jokergoo.github.io/ComplexHeatmap-reference/book/

安装

library(devtools)
install_github("jokergoo/ComplexHeatmap")

7.2 Apply to cBioPortal dataset

通过使用实际数据展示oncoPring()的优势。数据来源于cBioPortal。执行步骤如下：

登入 http://www.cbioportal.org，
查找研究 “Lung Adenocarcinoma Carcinoma” 并选择: “Lung Adenocarcinoma Carcinoma (TCGA, Provisinal)”,
在Enter Gene Set处选择 “General: Ras-Raf-MEK-Erk/JNK signaling (26 genes)”,
提交
在结果页面，
点击Download键，在“Type of Genetic alterations across all cases”处下载text。
样本也可以从结果页面下载，
点击“OncoPrint”，移动鼠标到图上，点击“download” 图标，点击“Sample order”.

这组数据已经在ComplexHeatmap包中。首先读入数据，进行前期处理。

mat = read.table(system.file("extdata", package = "ComplexHeatmap", 
    "tcga_lung_adenocarcinoma_provisional_ras_raf_mek_jnk_signalling.txt"), 
    header = TRUE, stringsAsFactors = FALSE, sep = "\t")
mat[is.na(mat)] = ""
rownames(mat) = mat[, 1]
mat = mat[, -1]
mat=  mat[, -ncol(mat)]
mat = t(as.matrix(mat))
mat[1:3, 1:3]

##      TCGA-05-4384-01 TCGA-05-4390-01 TCGA-05-4425-01
## KRAS "  "            "MUT;"          "  "           
## HRAS "  "            "  "            "  "           
## BRAF "  "            "  "            "  "

在mat中有三种不同的异常（alternation，不讨论翻译是否准确）：HOMDEL，AMP，MUT。首先定义如何添加这些不同的异常。

col = c("HOMDEL" = "blue", "AMP" = "red", "MUT" = "#008000")
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))
    },
    # big blue
    HOMDEL = function(x, y, w, h) {
        grid.rect(x, y, w-unit(0.5, "mm"), h-unit(0.5, "mm"), 
            gp = gpar(fill = col["HOMDEL"], col = NA))
    },
    # bug red
    AMP = function(x, y, w, h) {
        grid.rect(x, y, w-unit(0.5, "mm"), h-unit(0.5, "mm"), 
            gp = gpar(fill = col["AMP"], col = NA))
    },
    # small green
    MUT = function(x, y, w, h) {
        grid.rect(x, y, w-unit(0.5, "mm"), h*0.33, 
            gp = gpar(fill = col["MUT"], col = NA))
    }
)

现在，使用oncoPrint。将column_title和heatmap_legend_param保存成向量，因为随后它们会被用到多次。

column_title = "OncoPrint for TCGA Lung Adenocarcinoma, genes in Ras Raf MEK JNK signalling"
heatmap_legend_param = list(title = "Alternations", at = c("HOMDEL", "AMP", "MUT"), 
        labels = c("Deep deletion", "Amplification", "Mutation"))
oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    column_title = column_title, heatmap_legend_param = heatmap_legend_param)

7.2.1 移除空行空列

默认情况下，即使样本没有alteration，仍在热图中，但可以通过设置remove_empty_columns和remove_empty_rows到TRUE来移除这些没有alteration的样本：

oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    remove_empty_columns = TRUE, remove_empty_rows = TRUE,
    column_title = column_title, heatmap_legend_param = heatmap_legend_param)

7.2.2 重新排序

同Heatmap()函数已有，row_order或者column_order也能够排序（数字或者字符都行）。以下示例中，样本的顺序来自于cBio。可以看到memo sort和使用cBio的差异。

sample_order = scan(paste0(system.file("extdata", package = "ComplexHeatmap"), 
    "/sample_order.txt"), what = "character")
oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    row_order = 1:nrow(mat), column_order = sample_order,
    remove_empty_columns = TRUE, remove_empty_rows = TRUE,
    column_title = column_title, heatmap_legend_param = heatmap_legend_param)

7.2.3 OncoPrint注释

在顶部和右侧，有barplots展示每个基因或者样本的数目，在左侧，展示了每个基因发生突变所在样本的百分比。
barplot注释通过anno_oncoprint_barplot()实现。Barplots默认绘制所有的alteration类型，但是可以选择相应的子集添加到anno_oncoprint_barplot展示。以下是示例：

oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    top_annotation = HeatmapAnnotation(
        column_barplot = anno_oncoprint_barplot("MUT", border = TRUE, # only MUT
            height = unit(4, "cm"))),
    right_annotation = rowAnnotation(
        row_barplot = anno_oncoprint_barplot(c("AMP", "HOMDEL"),  # only AMP and HOMDEL
            border = TRUE, height = unit(4, "cm"), 
            axis_param = list(side = "bottom", labels_rot = 90))),
    remove_empty_columns = TRUE, remove_empty_rows = TRUE,
    column_title = column_title, heatmap_legend_param = heatmap_legend_param)

可以通过show_pct和show_row_names打开和关闭百分数及行名称。

oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    remove_empty_columns = TRUE, remove_empty_rows = TRUE,
    pct_side = "right", row_names_side = "left",
    column_title = column_title, heatmap_legend_param = heatmap_legend_param)

barplot注释属于常规注释，还可以添在HeatmapAnnotation()或者rowAnnotation中加更多的注释：

oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    remove_empty_columns = TRUE, remove_empty_rows = TRUE,
    top_annotation = HeatmapAnnotation(cbar = anno_oncoprint_barplot(),
        foo1 = 1:172,
        bar1 = anno_points(1:172)),
    left_annotation = rowAnnotation(foo2 = 1:26),
    right_annotation = rowAnnotation(bar2 = anno_barplot(1:26)),
    column_title = column_title, heatmap_legend_param = heatmap_legend_param)

7.2.4 oncoPrint作为热图

oncoPrint()实际上返回一个Heatmap()对象，因此可以水平或者垂直的添加更多的热图和注释，从而时间更加的复杂的展示。
以下示例添加了一个水平方向的热图。

ht_list = oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    column_title = column_title, heatmap_legend_param = heatmap_legend_param) +
Heatmap(matrix(rnorm(nrow(mat)*10), ncol = 10), name = "expr", width = unit(4, "cm"))
draw(ht_list)

亦可垂直添加：

ht_list = oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    column_title = column_title, heatmap_legend_param = heatmap_legend_param) %v%
Heatmap(matrix(rnorm(ncol(mat)*10), nrow = 10), name = "expr", height = unit(4, "cm"))
draw(ht_list)

像一般的热图已有，也可以分割热图：

ht_list = oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    column_title = column_title, heatmap_legend_param = heatmap_legend_param) +
Heatmap(matrix(rnorm(nrow(mat)*10), ncol = 10), name = "expr", width = unit(4, "cm"))
draw(ht_list, row_split = sample(c("a", "b"), nrow(mat), replace = TRUE))

当remove_empty_columns或者remove_empty_rows是设置为TRUE时，基因的数目和样本的数目可能不是原始的。如果原始的矩阵有行名和列名，可通过下面的方式获得：

ht = oncoPrint(mat,
    alter_fun = alter_fun, col = col, 
    remove_empty_columns = TRUE, remove_empty_rows = TRUE,
    column_title = column_title, heatmap_legend_param = heatmap_legend_param)
rownames(ht@matrix)
##  [1] "KRAS"   "HRAS"   "BRAF"   "RAF1"   "MAP3K1" "MAP3K2" "MAP3K3"
##  [8] "MAP3K4" "MAP3K5" "MAP2K1" "MAP2K2" "MAP2K3" "MAP2K4" "MAPK3" 
## [15] "MAPK4"  "MAPK7"  "MAPK8"  "MAPK9"  "MAPK12" "MAPK14" "DAB2"  
## [22] "RAB25"
colnames(ht@matrix)
##   [1] "TCGA-05-4390-01" "TCGA-38-4631-01" "TCGA-44-6144-01"
##   [4] "TCGA-44-6145-01" "TCGA-44-6146-01" "TCGA-49-4488-01"
##   [7] "TCGA-50-5930-01" "TCGA-50-5931-01" "TCGA-50-5932-01"
##  [10] "TCGA-50-5933-01" "TCGA-50-5941-01" "TCGA-50-5942-01"
##  [13] "TCGA-50-5944-01" "TCGA-50-5946-01" "TCGA-50-6591-01"
##  [16] "TCGA-50-6592-01" "TCGA-50-6594-01" "TCGA-67-4679-01"
##  [19] "TCGA-67-6215-01" "TCGA-73-4658-01" "TCGA-73-4676-01"
##  [22] "TCGA-75-5122-01" "TCGA-75-5125-01" "TCGA-75-5126-01"
##  [25] "TCGA-75-6206-01" "TCGA-75-6211-01" "TCGA-86-6562-01"
##  [28] "TCGA-05-4396-01" "TCGA-05-4405-01" "TCGA-05-4410-01"
##  [31] "TCGA-05-4415-01" "TCGA-05-4417-01" "TCGA-05-4424-01"
##  [34] "TCGA-05-4427-01" "TCGA-05-4433-01" "TCGA-44-6774-01"
##  [37] "TCGA-44-6775-01" "TCGA-44-6776-01" "TCGA-44-6777-01"
##  [40] "TCGA-44-6778-01" "TCGA-49-4487-01" "TCGA-49-4490-01"
##  [43] "TCGA-49-6744-01" "TCGA-49-6745-01" "TCGA-49-6767-01"
##  [46] "TCGA-50-5044-01" "TCGA-50-5051-01" "TCGA-50-5072-01"
##  [49] "TCGA-50-6590-01" "TCGA-55-6642-01" "TCGA-55-6712-01"
##  [52] "TCGA-71-6725-01" "TCGA-91-6828-01" "TCGA-91-6829-01"
##  [55] "TCGA-91-6835-01" "TCGA-91-6836-01" "TCGA-35-3615-01"
##  [58] "TCGA-44-2655-01" "TCGA-44-2656-01" "TCGA-44-2662-01"
##  [61] "TCGA-44-2666-01" "TCGA-44-2668-01" "TCGA-55-1592-01"
##  [64] "TCGA-55-1594-01" "TCGA-55-1595-01" "TCGA-64-1676-01"
##  [67] "TCGA-64-1677-01" "TCGA-64-1678-01" "TCGA-64-1680-01"
##  [70] "TCGA-67-3771-01" "TCGA-67-3773-01" "TCGA-67-3774-01"
##  [73] "TCGA-05-4244-01" "TCGA-05-4249-01" "TCGA-05-4250-01"
##  [76] "TCGA-35-4122-01" "TCGA-35-4123-01" "TCGA-44-2657-01"
##  [79] "TCGA-44-3398-01" "TCGA-44-3918-01" "TCGA-05-4382-01"
##  [82] "TCGA-05-4389-01" "TCGA-05-4395-01" "TCGA-05-4397-01"
##  [85] "TCGA-05-4398-01" "TCGA-05-4402-01" "TCGA-05-4403-01"
##  [88] "TCGA-05-4418-01" "TCGA-05-4420-01" "TCGA-05-4422-01"
##  [91] "TCGA-05-4426-01" "TCGA-05-4430-01" "TCGA-05-4434-01"
##  [94] "TCGA-38-4625-01" "TCGA-38-4626-01" "TCGA-38-4628-01"
##  [97] "TCGA-38-4630-01" "TCGA-44-3396-01" "TCGA-49-4486-01"
## [100] "TCGA-49-4505-01" "TCGA-49-4506-01" "TCGA-49-4507-01"
## [103] "TCGA-49-4510-01" "TCGA-73-4659-01" "TCGA-73-4662-01"
## [106] "TCGA-73-4668-01" "TCGA-73-4670-01" "TCGA-73-4677-01"
## [109] "TCGA-05-5428-01" "TCGA-05-5715-01" "TCGA-50-5045-01"
## [112] "TCGA-50-5049-01" "TCGA-50-5936-01" "TCGA-55-5899-01"
## [115] "TCGA-64-5774-01" "TCGA-64-5775-01" "TCGA-64-5778-01"
## [118] "TCGA-64-5815-01" "TCGA-75-5146-01" "TCGA-75-5147-01"
## [121] "TCGA-80-5611-01"

posted @ 2019-09-11 10:36 yuwq 阅读(7205) 评论(1) 编辑收藏举报

会员力量，点亮园子希望

刷新页面返回顶部

科研！跑步，羽毛球，骑行

充实的过每一天，把握每一次机会，我肯定是幸运的那个