ComplexHeatmap学习笔记和总结

ComplexHeatmap学习笔记和总结

声明：
本文档是在学习ComplexHeatmap和测试例子过程中，相关方法的小结，方便回顾查看，快速实现数据可视化，理解错误之处，欢迎批评指正

1. ComplexHeatmap总览

ComplexHeatmap软件包主要用来展现热图，常说的热图包括，body和components(title， dendrograms, matrix names 和热图注释（放在热图旁，该可根据注释组合不同的复杂图形）。
heatmap body部分可以根据行和列分割。
整体上讲，这个包类似于ggplot，layout操作，不同track和使用+组合，此外可以将注释图形直接加入主图中。

2. 单个热图

2.1 input 格式

单个热图和pheatmap功能较一致，且二者可以兼容使用

输入是一个数值矩阵，行和列名可以使用rownames和colnames分别命名
library(circlize)
col_fun = colorRamp2(c(-2, 0, 2), c(“green”, “white”, “red”)) # 矩阵的值会映射到-2和2之间，需要根据实际数据进行调整
colorRamp2() 如果使用同一个mapping color， 允许比较，可以清晰的看出不同处理的数据差异
对于连续型数据，可以提供颜色向量，eg: colorRamp2(seq(min(mat), max(mat), length = 10), rev(rainbow(10))), 会自动mapping，但是会受到离群值影响较大，discrete values（数字或字符）需要提供一个颜色向量

Heatmap(mat, name = "mat", col = col_fun)
Heatmap(mat, name = "mat", col = col_fun, column_title = "mat")
Heatmap(mat/4, name = "mat", col = col_fun, column_title = "mat/4")
Heatmap(abs(mat), name = "mat", col = col_fun, column_title = "abs(mat)")

**discrete values：**
discrete_mat = matrix(sample(letters[1:4], 100, replace = TRUE), 10, 10)
colors = structure(1:4, names = letters[1:4])
Heatmap(discrete_mat, name = "mat", col = colors,
    column_title = "a discrete character matrix")

pheatmap 介绍：
pheatmap(test, legend = FALSE)，如下常用参数和作用
border_color： cells 格子的颜色，颜色值
border：逻辑值，是否显示边框
show_rownames和show_colnames： 逻辑值，是否显示名字
display_numbers：逻辑，是否显示数字
number_format：格式调整
cellwidth 和cellheight： cell的宽和高调整，调整cells的大小
**annotation_col：添加列注释
annotation_row： 添加行注释，注意row和col要和矩阵一致，需要加个对应的rownames和colnames
annotation_colors：各个注释水平的颜色对应列表, 列表每个元素 var=c(注释水平="")
gaps_row 和gaps_col：数值向量，提供需要gap的索引，cluster_rows 和cluster_cols需要设置False
fontsize：数值，字体大小
scale：对row进行归一化
**labels_row：**重点显示几个基因名字，类似Mark annotaion, #labels_row = c("", “”, “”, “”, “”, “”, “”, “”, “”, “”, “”, “”, “”, “”, “”,
“”, “”, “Il10”, “Il15”, “Il1b”)
angle_col = 45
fontsize_row和fontsize_col：row和col字体的大小
pheatmap

2.2 颜色选取和相关参数介绍

默认linearly interpolated in LAB color space，但是可以根据数据使用colorRamp2()调整。

f1 = colorRamp2(seq(min(mat), max(mat), length = 3), c("blue", "#EEEEEE", "red"))
f2 = colorRamp2(seq(min(mat), max(mat), length = 3), c("blue", "#EEEEEE", "red"), space = "RGB")
Heatmap(mat, name = "mat1", col = f1, column_title = "LAB color space")
Heatmap(mat, name = "mat2", col = f2, column_title = "RGB color space")

border/border_gp 和 rect_gp 分别控制heatmap body 和 cell区域
border可以是逻辑值T或颜色向量，border_gp是一个gpar对象，个人理解这个gpar（grid::gpar()）类似于R 的par 和 html style 可以设置相关的属性

Heatmap(mat, name = "mat", border_gp = gpar(col = "black", lty = 2),
    column_title = "set heatmap borders")
Heatmap(mat, name = "mat", column_title = "I am a column title at the bottom", column_title_side = "bottom")
Heatmap(mat, name = "mat", column_title = "I am a column title", 
    column_title_gp = gpar(fill = "red", col = "white", border = "blue"))
    

2.3 聚类track

Heatmap(mat, name = "mat", clustering_distance_rows = "pearson",
    column_title = "pre-defined distance method (1 - pearson)")   #pearson
Heatmap(mat, name = "mat", clustering_distance_rows = function(m) dist(m), column_title = "a function that calculates distance matrix")  #距离
Heatmap(mat, name = "mat", clustering_distance_rows = function(x, y) 1 - cor(x, y), column_title = "a function that calculates pairwise distance")  #pairwise distance
#pairwise distance 去除离群值，（0.1，0.9)之间过滤筛选
at_with_outliers = mat
for(i in  1:10) mat_with_outliers[i, i] = 1000
robust_dist = function(x, y) {
    qx = quantile(x, c(0.1, 0.9))
    qy = quantile(y, c(0.1, 0.9))
    l = x > qx[1] & x < qx[2] & y > qy[1] & y < qy[2]
    x = x[l]
    y = y[l]
    sqrt(sum((x - y)^2))
}
Heatmap(mat_with_outliers, name = "mat", 
    col = colorRamp2(c(-2, 0, 2), c("green", "white", "red")),
    clustering_distance_rows = robust_dist,
    clustering_distance_columns = robust_dist,
    column_title = "robust_dist")

cell_fun： 控制cell显示，聚类方法通过clustering_method_rows 和 clustering_method_columns设置，与**hclust()**方法类似

library(cluster)
Heatmap(mat, name = "mat", cluster_rows = diana(mat),
   cluster_columns = agnes(t(mat)), column_title = "clustering objects")

如果想修改旁边的系统树的 style，可以先得到dendrogram对象，通过nodePar 和 edgePar来设置边和顶点，这个用的不多，

ibrary(dendextend)
row_dend = as.dendrogram(hclust(dist(mat)))
row_dend = color_branches(row_dend, k = 2) # `color_branches()` returns a dendrogram object
Heatmap(mat, name = "mat", cluster_rows = row_dend)

同样的，row_dend_gp 和 column_dend_gp控制系统树的设置

Heatmap(mat, name = "mat", cluster_rows = row_dend, row_dend_gp = gpar(col = "red"))

控制行和列的显示顺序，使用row_order和column_order，使用factors也可以，自然，行和列聚类关闭才生效。

Heatmap(mat, name = "mat", row_order = order(as.numeric(gsub("row", "", rownames(mat)))), column_order = order(as.numeric(gsub("column", "", colnames(mat)))), column_title = "reorder matrix")

列和行名显示位置参数
row_names_side # rowname 显示
row_dend_side #行进化树
column_names_side # 列名
column_dend_side # 列进化树

Heatmap(mat, name = "mat", row_names_side = "left", row_dend_side = "right", column_names_side = "top", column_dend_side = "bottom")

2.4 热图分割 split
控制分割的参数： row_km, row_split, column_km, column_split
row_km and column_km按照均值分割，另外可以设置row_km_repeats和column_km_repeats分别跑多次，最后取个一致性的分割值，比默认的要小。

Heatmap(mat, name = "mat",  row_km = 2, row_km_repeats = 100,
 column_km = 3, column_km_repeats = 100)

可根据字符向量分割，比较常用，row_split or column_split字符向量或数据框，需要和矩阵的维度一致。

Heatmap(mat, name = "mat", 
    row_split = rep(c("A", "B"), 9), column_split = rep(c("C", "D"), 12))
    #字符型矩阵
    # split by the first column in `discrete_mat`
Heatmap(discrete_mat, name = "mat", col = 1:4, row_split = discrete_mat[, 1])

slices（subgroups）顺序问题，默认是排序的，可以设置cluster_row_slices or cluster_column_slices为False, 这样顺序就按照column_split分割的顺序了

Heatmap(mat, name = "mat", 
  row_split = rep(LETTERS[1:3], 6),
    column_split = rep(letters[1:6], 4))
Heatmap(mat, name = "mat", row_split = factor(rep(LETTERS[1:3], 6), levels = LETTERS[3:1]),column_split=factor(rep(letters[1:6], 4), levels = letters[6:1]), cluster_row_slices = FALSE, cluster_column_slices = FALSE)

slices的其他属性，graphic parameters需要和slices的个数一致

ht_opt$TITLE_PADDING = unit(c(4, 4), "points")
Heatmap(mat, name = "mat", 
    row_km = 2, row_title_gp = gpar(col = c("red", "blue"), font = 1:2),
    row_names_gp = gpar(col = c("green", "orange"), fontsize = c(10, 14)),
    column_km = 3, column_title_gp = gpar(fill = c("red", "blue", "green"), font = 1:3),
    column_names_gp = gpar(col = c("green", "orange", "purple"), fontsize = c(10, 14, 8)))

slices之间的距离：row_gap = unit(5, “mm”)
row_gap = unit(5, “mm”)

边框： border = TRUE

另外分割时，注释图形一起分割，cell_fun分别画一个cells，layer_fun垂直版本，

small_mat = mat[1:9, 1:9]
col_fun = colorRamp2(c(-2, 0, 2), c("green", "white", "red"))
Heatmap(small_mat, name = "mat", col = col_fun,
    cell_fun = function(j, i, x, y, width, height, fill) {
        grid.text(sprintf("%.1f", small_mat[i, j]), x, y, gp = gpar(fontsize = 10))
})

3. 热图注释

热图注释赋予图形丰富的内容，可以展示轴相关的row和columns额外的信息，op_annotation, bottom_annotation, left_annotation 和 right_annotation控制位置参数。参数的值需要是HeatmapAnnotation 类，有HeatmapAnnotation和rowAnnotation() [注释row] 构建产生，官方文档说明，rowAnnotation只是HeatmapAnnotation(…, which = “row”)的特殊情况。

set.seed(123)
mat = matrix(rnorm(100), 10)
rownames(mat) = paste0("R", 1:10)
colnames(mat) = paste0("C", 1:10)
column_ha = HeatmapAnnotation(foo1 = runif(10), bar1 = anno_barplot(runif(10)))
row_ha = rowAnnotation(foo2 = runif(10), bar2 = anno_barplot(runif(10)))
Heatmap(mat, name = "mat", top_annotation = column_ha, right_annotation = row_ha)

注意注释列，行维度需要和热图中数据分别一致，注释图形函数以anno_*形式，可以注释block，图片，points， lines，Barplot, Boxplot, histogram, Density， Text， Mark，分别格式如下：

# blocks
Heatmap(matrix(rnorm(100), 10), name = "mat",
    top_annotation = HeatmapAnnotation(foo = anno_block(gp = gpar(fill = 2:4))),
    column_km = 3)
#images
image_png = sample(dir("IcoMoon-Free-master/PNG/64px", full.names = TRUE), 10)
ha = HeatmapAnnotation(foo = anno_image(image_png))
ha = HeatmapAnnotation(foo = anno_image(image_png, space = unit(3, "mm")))

#points
ha = HeatmapAnnotation(foo = anno_points(matrix(runif(20), nc = 2), 
    pch = 1:2, gp = gpar(col = 2:3)))   #  foo 只是个名字，可以修改
  # lines
  ha = HeatmapAnnotation(foo = anno_lines(cbind(c(1:5, 1:5), c(5:1, 5:1)), 
    gp = gpar(col = 2:3), add_points = TRUE, pt_gp = gpar(col = 5:6), pch = c(1, 16)))   # 数据可以是一个向量或者矩阵
 # barplot
 ha = HeatmapAnnotation(foo = anno_barplot(1:10, gp = gpar(fill = 1:10)))  #  向量是普通的bar图
 ha = HeatmapAnnotation(foo = anno_barplot(cbind(1:10, 10:1),                     #矩阵堆叠在一起，stack
    gp = gpar(fill = 2:3, col = 2:3)))
    # boxplot
    ha = HeatmapAnnotation(foo = anno_boxplot(m, height = unit(4, "cm"), 
    gp = gpar(fill = 1:10)))
#Histogram,对于row注释比较合适，输入数据类似 anno_boxplot()  [***a matrix or a list***]
‘m = matrix(rnorm(1000), nc = 100)
ha = rowAnnotation(foo = anno_histogram(m))
#Density, 类似Histogram，拟合曲线分布，type参数比较重要，可以取heatmap, violin，数据形式需要数值矩阵

m2 = matrix(rnorm(50*10), nrow = 50)
m = matrix(rnorm(1000), nc = 100)

ha = rowAnnotation(foo = anno_density(m, joyplot_scale = 2, 
    gp = gpar(fill = "#CCCCCC80")))
  ha = rowAnnotation(foo = anno_density(m2, type = "heatmap", width = unit(6, "cm")))
# Mark 注释，行和列有多个时，需要mark一部分，使用anno_mark()，至少需要两个参数，at  原始矩阵的索引，labels相应的text
m = matrix(rnorm(1000), nrow = 100)
rownames(m) = 1:100
ha = rowAnnotation(foo = anno_mark(at = c(1:4, 20, 60, 97:100), labels = month.name[1:10]))
Heatmap(m, name = "mat", cluster_rows = FALSE, right_annotation = ha,
    row_names_side = "left", row_names_gp = gpar(fontsize = 4))
Heatmap(m, name = "mat", cluster_rows = FALSE, right_annotation = ha,
    row_names_side = "left", row_names_gp = gpar(fontsize = 4), row_km = 4)

多个注释的，只需要name-value写到HeatmapAnnotation即可

ha = HeatmapAnnotation(foo = 1:10, 
    bar = cbind(1:10, 10:1),
    pt = anno_points(1:10),
    show_legend = c("bar" = FALSE)
)
Heatmap(matrix(rnorm(100), 10), name = "mat", top_annotation = ha)

小结：
注释同通过anno_*获取对象，通过 [right|left|bottom|top]_annotation显示注释的位置，所有注释对象可以使用+连接
hitlist<-anno1+anno1 draw(hitlist)，和多个下面介绍的多个热图组合一样

4. 热图列表操作

热图列表操作类似于，上面的注释，多个热图可以组合成一个list，使用draw进行显示，可以水平黏贴和垂直黏贴，水平黏贴比较常用，便于比较，所有热图和对应注释的的row数量 需要相同，即行数需要相同，

ht1 = Heatmap(mat1, name = "rnorm")
ht2 = Heatmap(mat2, name = "runif")
ht3 = Heatmap(le, name = "letters")
draw(ht_list, row_title = "Three heatmaps, row title", row_title_gp = gpar(col = "red"),
    column_title = "Three heatmaps, column title", column_title_gp = gpar(fontsize = 16))

ht_list =ht1 + ht2 + ht3 #返回heatmap list，会面可以任意追加 （+）热图
组合图常用调整参数：

1. size: width = unit(5, “mm”)
2. 热图间Gap: draw(ht_list, ht_gap = unit(1, “cm”))
   3）Row annotations，水平heatmaplist 可以连接注释

ha1 = rowAnnotation(foo = 1:12, bar = anno_barplot(1:12, width = unit(4, "cm")))
ht1 = Heatmap(mat1, name = "rnorm", col = col_rnorm, row_km = 2)
ht1 + ha1

Heatmap(mat1, name = "rnorm", col = col_rnorm, row_km = 2) + 
    rowAnnotation(foo = 1:12) +
    rowAnnotation(bar = anno_barplot(1:12, width = unit(4, "cm")))

4）adding a text annotation row names

ht1 + ha1 + rowAnnotation(rn = anno_text(rownames(mat1), 
    location = unit(0, "npc"), just = "left"))

5)adjust_annotation_extension,是否考虑轴和names之间的自动调整，heatmap name 和heatbody之间的空白区域

其他参数参考官方文档：
https://jokergoo.github.io/ComplexHeatmap-reference/book/upset-plot.html