简介

本文记录了：

• 1、箱型图的绘制，并进行显著性差异检验，一是两两比较指定组，二是两两比较所有组；
• 2、以及比较“某个变量”在两个分组间是否存在差异。

加载所需要的包

library(magrittr)

library(Seurat) %>% suppressMessages()
paste("Seurat", packageVersion("Seurat")) %>% print()

library(dplyr) %>% suppressMessages()
paste("dplyr", packageVersion("dplyr")) %>% print()

library(survival) %>% suppressMessages()
paste("survival", packageVersion("survival")) %>% print()

library(dittoSeq) %>% suppressMessages()
paste("dittoSeq", packageVersion("dittoSeq")) %>% print()

library(ggplot2) %>% suppressMessages()
paste("ggplot2", packageVersion("ggplot2")) %>% print()

library(ggpubr) %>% suppressMessages()
paste("ggpubr", packageVersion("ggpubr")) %>% print()

[1] "dplyr 1.1.4"
[1] "survival 3.8.3"
[1] "dittoSeq 1.20.0"
[1] "ggplot2 3.5.2"
[1] "ggpubr 0.6.0"

构建数据的格式

data <- data.frame("hcc_type" = mye$HCC_type,
                   "cell_clusters" = mye$identify_myeloid,
                   "risk_score" = mye$risk_score)
head(data)
           hcc_type cell_clusters risk_score
P01_T_0014        R   Macro AZGP1 0.34462416
P01_T_0121        R   Macro AZGP1 0.82306410
P01_T_0159        R   Macro AZGP1 1.37164070
P01_T_0167        R    Macro MT1X 0.12383450
P01_T_0172        R   Macro CXCL3 0.13059335
P01_T_0187        R  Macro HAVCR1 0.03061694

一、两两比较指定组

my_comparisons <- list(c("Macro AZGP1", "Macro MT1X"), c("Macro AZGP1", "Macro CXCL3"))

ggboxplot(data, x = "cell_clusters", y = "risk_score", fill = "cell_clusters",
          ylab = "log(TPM + 1)",
          xlab = "",
          legend.title = "",
          palette = dittoColors(),
          width = 0.6
)+
  NoLegend()+
  theme(axis.text.x = element_text(angle = 45, hjust = 1))+
  ggtitle("risk scores")+theme(plot.title = element_text(hjust = 0.5))+
  stat_compare_means(comparisons = my_comparisons,
                     method = "wilcox.test",
                     symnum.args = list(cutpoints = c(0, 0.001, 0.01, 0.05, 1), 
                                        symbols = c("***", "**", "*", "ns")
                     ),
                     label = "p.signif",
                     hide.ns = TRUE
                     )

注：my_comparisons是设置要比较的两组。

二、两两比较所有组

data$cell_clusters <- as.factor(data$cell_clusters)
all_pairs <- combn(levels(data$cell_clusters), 2, simplify = FALSE)

ggboxplot(data, x = "cell_clusters", y = "risk_score", fill = "cell_clusters",
          ylab = "log(TPM + 1)",
          xlab = "",
          legend.title = "",
          palette = dittoColors(),
          width = 0.6)+
  NoLegend()+
  theme(axis.text.x = element_text(angle = 45, hjust = 1))+
  ggtitle("risk scores")+theme(plot.title = element_text(hjust = 0.5))+
  stat_compare_means(comparisons = all_pairs,
                     method = "wilcox.test",
                     symnum.args = list(cutpoints = c(0, 0.001, 0.01, 0.05, 1), 
                                        symbols = c("***", "**", "*", "ns")
                     ),
                     label = "p.signif",
                     hide.ns = TRUE
                     )

注：要将要比较的那个变量设置为因子。

三、比较“某个变量”在两个分组间是否存在差异。

ggboxplot(data, x="cell_clusters", y="risk_score", fill = "hcc_type", 
            ylab="log(TPM + 1)",
            xlab="",
            legend.title="HCC type",
            palette = c("#3C4A93","#C7263B"),
            width = 0.6
          )+
  theme(axis.text.x = element_text(angle = 45, hjust = 1))+
  ggtitle("risk score") + theme(plot.title = element_text(hjust = 0.5))+
  stat_compare_means(aes(group = hcc_type),
                     method = "wilcox.test",
                     symnum.args = list(cutpoints = c(0, 0.001, 0.01, 0.05, 1), 
                                        symbols = c("***", "**", "*", "ns")
                                        ),
                     label = "p.signif"
                     )

附录

代码：
https://github.com/wPencil/MyNotes/blob/9b354894c20a59a9e94d64d28337020b01fe3e75/R/BoxPlot