r - 如何在ggplot中绘制虚线95%置信区间

Question

我对 R 相当陌生，所以这可能是一个简单的问题。
对于演示，我需要制作带有置信区域的折线图。

数据代表化疗治疗期间不同时间点的生活质量评分(时间点 0(基线)、时间点 4(治疗 4 个周期后)和时间点 8(8 个周期后)。

这是我的数据:

ID Time Score
146 8   0.0
4   0   16.7
51  0   16.7
94  4   25.0
117 4   25.0
26  0   25.0
149 8   33.3
151 8   33.3
153 8   33.3
93  4   33.3
103 4   33.3
31  0   33.3
57  0   33.6
145 8   41.6
33  0   41.6
79  4   41.7
5   0   41.7
6   0   41.7
15  0   41.7
38  0   41.7
133 8   50.0
177 8   50.0
75  4   50.0
80  4   50.0
87  4   50.0
88  4   50.0
98  4   50.0
100 4   50.0
114 4   50.0
12  0   50.0
28  0   50.0
32  0   50.0
37  0   50.0
44  0   50.0
47  0   50.0
49  0   50.0
50  0   50.0
54  0   50.0
122 8   58.3
135 8   58.3
139 8   58.3
154 8   58.3
158 8   58.3
176 8   58.3
82  4   58.3
104 4   58.3
105 4   58.3
2   0   58.3
3   0   58.3
21  0   58.3
25  0   58.3
42  0   58.3
45  0   58.3
140 8   66.7
147 8   66.7
148 8   66.7
160 8   66.7
162 8   66.7
169 8   66.7
180 8   66.7
62  4   66.7
73  4   66.7
77  4   66.7
85  4   66.7
92  4   66.7
101 4   66.7
106 4   66.7
109 4   66.7
118 4   66.7
10  0   66.7
13  0   66.7
16  0   66.7
17  0   66.7
18  0   66.7
24  0   66.7
27  0   66.7
29  0   66.7
30  0   66.7
34  0   66.7
43  0   66.7
53  0   66.7
55  0   66.7
56  0   66.7
58  0   66.7
1   0   67.0
125 8   75.0
129 8   75.0
137 8   75.0
155 8   75.0
71  4   75.0
74  4   75.0
95  4   75.0
113 4   75.0
22  0   75.0
46  0   75.0
59  0   75.0
121 8   83.3
126 8   83.3
127 8   83.3
128 8   83.3
130 8   83.3
131 8   83.3
134 8   83.3
136 8   83.3
142 8   83.3
143 8   83.3
150 8   83.3
152 8   83.3
170 8   83.3
172 8   83.3
173 8   83.3
178 8   83.3
179 8   83.3
61  4   83.3
65  4   83.3
66  4   83.3
68  4   83.3
69  4   83.3
76  4   83.3
84  4   83.3
90  4   83.3
96  4   83.3
97  4   83.3
99  4   83.3
110 4   83.3
112 4   83.3
116 4   83.3
119 4   83.3
7   0   83.3
8   0   83.3
9   0   83.3
11  0   83.3
14  0   83.3
20  0   83.3
35  0   83.3
36  0   83.3
39  0   83.3
40  0   83.3
41  0   83.3
52  0   83.3
144 8   91.7
156 8   91.7
159 8   91.7
168 8   91.7
120 4   91.7
19  0   91.7
60  0   91.7
166 8   100.0
67  4   100.0
83  4   100.0
23  0   100.0
48  0   100.0
123 8   NA
124 8   NA
132 8   NA
138 8   NA
141 8   NA
157 8   NA
161 8   NA
163 8   NA
164 8   NA
165 8   NA
167 8   NA
171 8   NA
174 8   NA
175 8   NA
63  4   NA
64  4   NA
70  4   NA
72  4   NA
78  4   NA
81  4   NA
86  4   NA
89  4   NA
91  4   NA
102 4   NA
107 4   NA
108 4   NA
111 4   NA
115 4   NA
Showing 1 to 21 of 180 entries, 2 total columns

这是我的数据结构(我已将数据库从宽格式融化为长格式:

str(Qol3)
Classes ‘tbl_df’, ‘tbl’ and 'data.frame':   180 obs. of  2 variables:
 $ Time : Factor w/ 3 levels "0","4","8": 1 1 1 1 1 1 1 1 1 1 ...
 $ Score: num  67 58.3 58.3 16.7 41.7 41.7 83.3 83.3 83.3 66.7 ...

summary(Qol3)
 Time       Score       
 0:60   Min.   :  0.00  
 4:60   1st Qu.: 56.23  
 8:60   Median : 66.70  
        Mean   : 66.94  
        3rd Qu.: 83.30  
        Max.   :100.00  
        NA's   :28      
> 

head(Qol3)
# A tibble: 6 x 2
  Time  Score
  <fct> <dbl>
1 0      67  
2 0      58.3
3 0      58.3
4 0      16.7
5 0      41.7
6 0      41.7

我想做一个看起来像这样的情节:


到目前为止，我得到了这个与以下代码一起工作:

ggplot(data=Qol3, aes(x=Time, y=Score, group=1)) +
  stat_summary(fun.y=mean, geom = "line") +
  stat_summary(fun.y=mean, geom = "point") +
  xlab("Induction cycle") + ylab("Score") +
  ggtitle("EORTC QLQ-C30")+
  stat_summary(fun.data = mean_cl_boot, geom = "errorbar", width = .2)+
  expand_limits(y=c(55, 85)) +
  theme(panel.background = element_rect(fill = "white"),
        panel.grid.minor = element_line(color = "gray90", size = 0.25),
        plot.background = element_rect(fill = "gray90"))

这给了我以下情节:



但我想用虚线更改这些误差条(在 X 轴上每个化疗周期点的平均值之间绘制的线上方和下方)

有人可以为我提供解决方案吗？另外我应该使用什么设置来使 X 轴上的直接变量“0”向左移动(我想去掉行尾的空白(这在视觉上对我更有吸引力)

非常感谢!

Answer 1

您可以使用以下代码

ggplot(Qol3, aes(x=Time , y=Score)) + 
  stat_summary(geom = "line", fun.y = mean) +
  stat_summary(geom="ribbon", fun.data=mean_cl_boot, 
               conf.int=0.95, alpha = 0.0, linetype="dashed", colour="red")

使用基本的情节，你可以这样做

#Calculation of confidence interval
library(Rmisc)

try <- group.CI(Score ~ Time,
         data=Qol3,
         ci = 0.95)

plot(try$Time, try$Score.mean, type = "l", xlab="Induction cycle", ylab= "Score")
lines(try$Time, try$Score.upper, lty = 2)
lines(try$Time, try$Score.lower, lty = 2)