seurat单细胞数据分析实现DimHeatmap函数

上游分析：https://www.jianshu.com/p/4f7aeae81ef1

001、

cell <- pbmc[["pca"]]@cell.embeddings
cell <- cell[order(cell[,1], decreasing = T),]
cell <- rownames(cell)[c(1:10, (length(rownames(cell)) - 9):length(rownames(cell)))]
cell ## 提取细胞
feature <- pbmc[["pca"]]@feature.loadings
feature <- feature[order(feature[,1], decreasing = T),]
feature <- rownames(feature)[c(1:15, (length(rownames(feature)) - 14):length(rownames(feature)))]
feature ## 提取基因
dat <- pbmc[["RNA"]]@scale.data
dat <- t(dat)[cell, feature] ## 提取绘图数据
par(mar = c(1, 1, 3, 5))
plot.new()
image(dat, axes = FALSE,add = TRUE, ## 绘图
col = PurpleAndYellow()
)
axis(side = 4, at = seq(0, 1, length = ncol(dat)), labels = colnames(dat),
las = 1, tick = FALSE, mgp = c(0, -0.5, 0), cex.axis = 0.75
)
title(main = "PC1")

002、标准答案

DimHeatmap(pbmc, dims = 1, cells = 20, balanced = TRUE)

003、使用默认颜色

par(mar = c(1, 1, 3, 5))
plot.new()
image(dat, axes = FALSE,add = TRUE,
#col = PurpleAndYellow()
)
axis(side = 4, at = seq(0, 1, length = ncol(dat)), labels = colnames(dat),
las = 1, tick = FALSE, mgp = c(0, -0.5, 0), cex.axis = 0.75
)
title(main = "PC1")

004、增加细胞名称

par(mar = c(9, 1, 3, 5))
plot.new()
image(dat, axes = FALSE,add = TRUE,
#col = PurpleAndYellow()
)
axis(side = 4, at = seq(0, 1, length = ncol(dat)), labels = colnames(dat),
las = 1, tick = FALSE, mgp = c(0, -0.5, 0), cex.axis = 0.75
)
axis(side = 1, at = seq(0, 1, length = nrow(dat)), labels = rownames(dat), ## 增加细胞名称
las = 2, tick = FALSE, mgp = c(0, -0.5, 0), cex.axis = 0.75
)
title(main = "PC1")