缘由
统计是生信的基石之一
碎碎念
要爱一个具体的人，不要爱一个抽象的人； 要爱生活，不要爱生活的意义。 ——《卡拉马佐夫兄弟》
Ref
GGPUBR: How to Add P-Values Generated Elsewhere to a GGPLOT - Datanovia
解读
1. 基础p值标注
手动创建统计结果
library(ggpubr)
df <- ToothGrowth
df$dose <- factor(df$dose)
# 手动创建统计检验结果
stat.test <- tibble::tribble(
~group1, ~group2, ~p.adj,
"0.5", "1", 2.54e-07,
"0.5", "2", 1.32e-13,
"1", "2", 1.91e-05
)
stat.test
# 基础箱线图添加p值
ggboxplot(df, x = "dose", y = "len") +
stat_pvalue_manual(
stat.test,
y.position = 35, step.increase = 0.1,
label = "p.adj"
)stat_pvalue_manual(): 手动添加p值标注y.position: 设置p值标注的垂直位置step.increase: 多个比较间的垂直间距label: 指定显示p值还是显著性星号
自动统计检验与标注
# 自动进行t检验
stat.test <- df %>% t_test(len ~ dose)
stat.test
# 自动计算标注位置并添加显著性星号
stat.test <- stat.test %>% add_xy_position(x = "dose")
bxp + stat_pvalue_manual(stat.test, label = "p.adj.signif", tip.length = 0.01)2. 分组分析与分面标注
以参考组为基础的多重比较
# 在每个supp组内，以0.5剂量为参考进行比较
stat.test <- df %>%
group_by(supp) %>%
t_test(len ~ dose, ref.group = "0.5")
stat.test
# 分面显示并添加p值
stat.test <- stat.test %>% add_xy_position(x = "dose")
ggboxplot(df, x = "dose", y = "len", facet.by = "supp") +
stat_pvalue_manual(stat.test, label = "p.adj.signif", tip.length = 0.01)复杂分组布局
# 分组t检验
stat.test <- df %>%
group_by(supp) %>%
t_test(len ~ dose, ref.group = "0.5")
# 分组箱线图与p值标注
stat.test <- stat.test %>%
add_xy_position(x = "supp", dodge = 0.8) # dodge参数控制位置偏移
bxp <- ggboxplot(df, x = "supp", y = "len", color = "dose")
bxp + stat_pvalue_manual(stat.test, label = "p.adj", tip.length = 0.01)ref.group: 指定参考组进行多重比较dodge: 在分组图中调整p值标注的水平位置
3. 手动位置控制与高级定制
精确控制y轴位置
# 手动指定每个比较的y轴位置
stat.test <- ToothGrowth %>%
t_test(len ~ dose) %>%
mutate(y.position = c(29, 35, 39)) # 分别为三个比较指定位置
stat.test分面图中的组内比较
# 在每个剂量水平比较supp
stat.test <- df %>%
group_by(dose) %>%
t_test(len ~ supp) %>%
adjust_pvalue() %>%
mutate(y.position = 35) # 统一y轴位置
p <- ggboxplot(
df, x = "supp", y = "len",
color = "supp", palette = "jco",
facet.by = "dose", ylim = c(0, 40)
)
p + stat_pvalue_manual(stat.test, label = "p.adj")4. 水平连接线与高级定位
使用xmin参数创建水平连接
stat.test <- df %>%
group_by(dose) %>%
t_test(len ~ supp) %>%
adjust_pvalue() %>%
mutate(y.position = 35)
p <- ggboxplot(
ToothGrowth, x = "dose", y = "len",
color = "supp", palette = "jco",
ylim = c(0, 40)
)
# 使用xmin创建水平方向的p值标注
p + stat_pvalue_manual(stat.test, xmin = "dose", xmax = NULL)5. geom_bracket函数：更灵活的标注
基础括号标注
# 使用geom_bracket手动添加括号和标注
ggboxplot(df, x = "dose", y = "len") +
geom_bracket(
xmin = "0.5", xmax = "1", y.position = 30,
label = "t-test, p < 0.05"
)自定义外观
# 调整尖端长度和标签位置
ggboxplot(df, x = "dose", y = "len") +
geom_bracket(
xmin = "0.5", xmax = "1", y.position = 30,
label = "t-test, p < 0.05",
tip.length = c(0.2, 0.02), # 左右尖端长度
vjust = -1 # 垂直调整标签位置
)数学表达式标注
# 使用plotmath表达式创建学术风格的p值标注
ggboxplot(df, x = "dose", y = "len") +
geom_bracket(
xmin = "0.5", xmax = "1", y.position = 30,
label = "list(~italic(p)<=0.001)",
type = "expression", # 指定为表达式类型
tip.length = c(0.2, 0.02)
)多重括号标注
# 一次性添加多个比较的括号
ggboxplot(df, x = "dose", y = "len") +
geom_bracket(
xmin = c("0.5", "1"), xmax = c("1", "2"),
y.position = c(30, 35), label = c("***", "**"),
tip.length = 0.01
)6. 与compare_means集成
自动化工作流
# 使用compare_means自动完成检验和标注
stat.test <- compare_means(len ~ dose, ToothGrowth, method = "t.test")
ggboxplot(df, x = "dose", y = "len") +
geom_bracket(
aes(xmin = group1, xmax = group2, label = signif(p, 2)),
data = stat.test, y.position = 35
)高级布局控制
# 增加括号间距
ggboxplot(df, x = "dose", y = "len") +
geom_bracket(
aes(xmin = group1, xmax = group2, label = signif(p, 2)),
data = stat.test, y.position = 35, step.increase = 0.1
)
# 或精确指定每个比较的位置
ggboxplot(df, x = "dose", y = "len") +
geom_bracket(
aes(xmin = group1, xmax = group2, label = signif(p, 2)),
data = stat.test, y.position = c(32, 35, 38)
)