可视化—转录组箱线图美化

plot_genes_box_tianfu <- function(exp, Group, related_gene, 
                                  output_file = "figure/genes_box.pdf",
                                   fig_width = 12, fig_height = 8, 
                                   ncol_facet = 4) {
  # -------------------------------------------------------------
  # exp: 表达矩阵，行为基因，列为样本
  # Group: 因子型向量，长度与列数一致，表示样本分组（如 normal / tumor）
  # related_gene: 感兴趣的基因向量
  # file_name: 保存图片的文件名（含路径）
  # fig_width: 保存图片的宽度（单位：英寸）
  # fig_height: 保存图片的高度（单位：英寸）
  # ncol_facet: facet_wrap 每行排列多少个基因
  # -------------------------------------------------------------
  
  library(ggplot2)
  library(ggpubr)
  library(dplyr)
  library(tidyr)
  library(tibble)
  library(rstatix)
  
  # 1. 整理数据 ---------------------------------------------------
  df_long <- exp %>%
    as.data.frame() %>%
    rownames_to_column(var = "Gene") %>%
    filter(Gene %in% related_gene) %>%
    pivot_longer(
      cols = -Gene,
      names_to = "Sample",
      values_to = "Expression"
    )
  
  # 添加分组信息
  df_long$Group <- Group[match(df_long$Sample, colnames(exp))]
  df_long$Group <- factor(df_long$Group, levels = c("normal", "tumor"))
  
  # 2. 计算中位值 --------------------------------------------------
  bb <- df_long %>%
    group_by(Gene, Group) %>%
    summarise(median_value = median(Expression), .groups = "drop")
  
  # 3. 每个基因单独显著性检验 -------------------------------------
  pval_df <- df_long %>%
    group_by(Gene) %>%
    wilcox_test(Expression ~ Group) %>%
    add_significance("p") %>%
    mutate(Group1 = "normal", Group2 = "tumor")
  
  # 确定显著性标签位置（最大值的 105%）
  ypos <- df_long %>%
    group_by(Gene) %>%
    summarise(y.position = max(Expression) * 1.05, .groups = "drop")
  pval_df <- left_join(pval_df, ypos, by = "Gene")
  
  # 4. 自定义配色 --------------------------------------------------
  color1 <- c("#66C2A5", "#FC8D62")
  
  # 5. 绘图 --------------------------------------------------------
  p <- ggplot(df_long, aes(x = Group, y = Expression)) +
    geom_boxplot(aes(fill = Group), outlier.shape = NA, lty = 0) +
    stat_boxplot(geom = "errorbar", linewidth = 0.8, width = 0.3, color = "grey40") +
    # 中位线（关闭继承映射防止 fill 冲突）
    geom_segment(data = bb,
                 aes(x = as.numeric(Group) - 0.25,
                     xend = as.numeric(Group) + 0.25,
                     y = median_value,
                     yend = median_value),
                 inherit.aes = FALSE,
                 color = "blue", linewidth = 0.8) +
    # 平均值
    stat_summary(fun = mean, geom = "point", shape = 8, size = 2, color = "white") +
    # 显著性
    stat_pvalue_manual(pval_df,
                       label = "p.signif",
                       xmin = "Group1",
                       xmax = "Group2",
                       y.position = "y.position",
                       tip.length = 0.01,
                       size = 3) +
    scale_fill_manual(values = color1) +
    facet_wrap(~ Gene, scales = "free_y", ncol = ncol_facet) +
    theme_classic() +
    theme(strip.text = element_text(face = "plain", size = 8),#控制每个面板顶部基因名字的字体粗细（bold）和大小（12）
          strip.background = element_blank(),                      # 去掉顶部基因名字的背景方框
          axis.text.y = element_text(size = 7, color = "grey20"),
          axis.text.x = element_text(size = 7, color = "grey20"),
          axis.title.y = element_text(size = 8, color = "grey20"),
          axis.title.x = element_blank(),#隐藏 x 轴标题
          axis.line = element_line(linewidth = 0.5),
          axis.ticks = element_line(linewidth = 0.5),
          #panel.grid.major = element_line(linewidth = 0.3, color = "#E8E8E8"),
          panel.grid.major = element_blank(),  # 不显示主网格线
          legend.position = "none")#隐藏图例
  
  # 6. 保存图片 ----------------------------------------------------
  ggsave(output_file, plot = p, width = fig_width, height = fig_height)
  
  # 返回绘图对象，方便在 R 里直接查看
  return(p)
}

p <- plot_genes_box_tianfu(exp, Group, related_gene,
                            output_file = "figure/genes_box.pdf",
                            fig_width = 8, fig_height = 6,
                            ncol_facet = 6)
print(p)