Geom_Line分组错误：如何实现多组共享起点的点连接？

问题背景 #

需要绘制分组折线图，核心需求：

• Sham和mTBI两组共享3.5月龄的起始数据点
• 后续分别连接各自组对应7.5、9.5月龄的点
  但当前ggplot代码生成的折线错误连接了所有数据点。

数据集结构 #

FA_MRI_APOE

Condition Treatment Age (Months) Region avg_FA sd_FA
<fctr> <chr> <fctr><chr> <dbl>    <dbl>
APOE2   Sham    3.5 CC  0.1990432   NA
APOE2   Sham    3.5 EC  0.2269353   NA
APOE2   Sham    3.5 HP  0.2253147   NA
APOE2   Sham    3.5 TH  0.3257256   NA
APOE2   Sham    7.5 CC  0.3093073   0.08619885
APOE2   Sham    7.5 EC  0.2255272   0.07652789
APOE2   Sham    7.5 HP  0.2897462   0.02708867
APOE2   Sham    7.5 TH  0.2866555   0.03557000
APOE2   Sham    9.5 CC  0.1840524   NA
APOE2   Sham    9.5 EC  0.3347699   NA

数据集dput输出 #

structure(list(Condition = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L), levels = c("APOE2", "APOE3", "APOE4", 
"PS19"), class = "factor"), Treatment = c("Sham", "Sham", "Sham", 
"Sham", "Sham", "Sham", "Sham", "Sham", "Sham", "Sham", "Sham", "Sham", 
"mTBI", "mTBI", "mTBI", "mTBI", "Sham", "Sham", "Sham", "Sham", "Sham", 
"Sham", "Sham", "Sham", "mTBI", "mTBI", "mTBI", "mTBI"), `Age (Months)` = structure(c(1L, 
1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 2L, 2L, 2L, 2L, 
1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), levels = c("3.5", 
"7.5", "9.5"), class = "factor"), 
    Region = c("CC", "EC", "HP", "TH", "CC", "EC", "HP", "TH", 
    "CC", "EC", "HP", "TH", "CC", "EC", "HP", "TH", "CC", "EC", 
    "HP", "TH", "CC", "EC", "HP", "TH", "CC", "EC", "HP", "TH"
    ), avg_FA = c(0.199043221771717, 0.226935303, 0.225314745679497, 
    0.325725596398115, 0.309307302666667, 0.225527225666667, 
    0.289746216333333, 0.286655532, 0.184052395, 0.334769852, 
    0.34609792, 0.237951324, 0.303560921, 0.233204448, 0.30656011, 
    0.296695315, 0.256386488908901, 0.2787716635, 0.270610670559108, 
    0.265520500484854, 0.289559764, 0.337927988666667, 0.370170086833333, 
    0.352682695666667, 0.391898785, 0.112882524, 0.29461883, 
    0.258586437), sd_FA = c(NA, NA, NA, NA, 0.0861988548059451, 
    0.0765278945487883, 0.0270886655892862, 0.0355699995442251, 
    NA, NA, NA, NA, NA, NA, NA, NA, 0.0332061193275388, 0.0121768772823131, 
    0.0441863595350275, 0.063138972154108, 0.11418350071109, 
    0.0829486022039601, 0.0831420050073176, 0.0689368863615038, 
    NA, NA, NA, NA)), class = c("grouped_df", "tbl_df", "tbl", 
"data.frame"), row.names = c(NA, -28L), groups = structure(list(
    Condition = structure(c(1L, 1L, 1L, 1L, 3L, 3L, 3L), levels = c("APOE2", 
    "APOE3", "APOE4", "PS19"), class = "factor"), Treatment = c("Sham", 
    "Sham", "Sham", "mTBI", "Sham", "Sham", "mTBI"), `Age (Months)` = structure(c(1L, 
    2L, 3L, 2L, 1L, 2L, 2L), levels = c("3.5", "7.5", "9.5"), class = "factor"), 
    .rows = structure(list(1:4, 5:8, 9:12, 13:16, 17:20, 21:24, 
        25:28), ptype = integer(0), class = c("vctrs_list_of", 
    "vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -7L), .drop = TRUE))

原错误绘图代码 #

ggplot(FA_MRI_APOE, aes(x = `Age (Months)`,
                        y = avg_FA,
                        color = factor(Treatment, levels = c("Sham", "mTBI")),
                        group = factor(Treatment, levels = c("Sham", "mTBI")))) +
  geom_errorbar(aes(ymin = avg_FA - sd_FA,
                    ymax = avg_FA + sd_FA),
                width = 0.1) +
  geom_point(size = 2.5) +
  geom_line(group = 1) +
  geom_jitter(aes(x = `Age (Months)`,
                  y = `Mean`),
              position = position_jitter(0.1),
              alpha = 0.5,
              size = 1.5,
              data = FA_MRI_APOE_raw) +
  scale_color_manual(name = "Treatment", values = c("#0074C1", "#F7530B")) +
  facet_grid(Region ~ Condition) +
  theme(legend.position = "bottom")

错误原因 #

1. geom_line(group = 1)强制将所有数据归为同一组，直接覆盖了全局aes中按Treatment分组的规则，导致所有点被错误连接。
2. 原数据集中mTBI组没有3.5月龄的数据，无法直接实现两组共享起始点的需求。

修正步骤与代码 #

1. 预处理数据集 #

复制Sham组的3.5月龄数据，标记为mTBI组，合并到原数据中，确保两组有相同的起始点：

library(dplyr)

# 提取3.5月龄的Sham数据并转换为mTBI组
shared_start <- FA_MRI_APOE %>%
  filter(`Age (Months)` == "3.5", Treatment == "Sham") %>%
  mutate(Treatment = "mTBI")

# 合并原数据与共享起始点数据，并按分组排序确保折线连接顺序正确
FA_MRI_APOE_adj <- bind_rows(FA_MRI_APOE, shared_start) %>%
  arrange(Condition, Region, Treatment, `Age (Months)`)

2. 修正绘图代码 #

移除geom_line中手动设置的group = 1，沿用全局分组规则，同时处理NA值避免绘图错误：

ggplot(FA_MRI_APOE_adj, aes(x = `Age (Months)`,
                            y = avg_FA,
                            color = factor(Treatment, levels = c("Sham", "mTBI")),
                            group = factor(Treatment, levels = c("Sham", "mTBI")))) +
  geom_errorbar(aes(ymin = avg_FA - sd_FA,
                    ymax = avg_FA + sd_FA),
                width = 0.1,
                na.rm = TRUE) + # 忽略sd_FA中的NA值，避免误差棒绘制失败
  geom_point(size = 2.5) +
  geom_line() + # 使用全局group参数按Treatment分组绘制折线
  geom_jitter(aes(x = `Age (Months)`,
                  y = `Mean`),
              position = position_jitter(0.1),
              alpha = 0.5,
              size = 1.5,
              data = FA_MRI_APOE_raw,
              inherit.aes = FALSE) + # 避免继承主图的color和group设置
  scale_color_manual(name = "Treatment", values = c("#0074C1", "#F7530B")) +
  facet_grid(Region ~ Condition) +
  theme(legend.position = "bottom")

关键说明 #

• 数据集预处理：通过复制Sham组的3.5月龄数据并标记为mTBI，实现两组共享起始点的核心需求。
• 移除group=1：该设置会破坏分组规则，去掉后ggplot会自动按全局aes中的Treatment分组绘制折线。
• na.rm=TRUE：处理sd_FA中的NA值，避免误差棒因缺失值无法绘制。
• inherit.aes=FALSE：确保散点图不继承主图的颜色和分组设置，避免干扰。

内容的提问来源于stack exchange，提问作者Ta-Ty