感谢您分享数据。您发布的数据有点长，因此可能无法完全复制和粘贴

您的数据具有广泛的格式，您需要查找相似组之间的每个测量值的平均值（按天，水平，治疗定义）。因此，我们可以采用多种格式进行处理：

tmp <- Data %>% group_by(Trmt,Level) %>% summarise_all(mean)
    > head(tmp)
# A tibble: 6 x 19
# Groups:   Trmt [2]
  Trmt  Level  Day1  Day2  Day4  Day7 Day10 Day13 Day18 Day23 Day28 Day35 Day42
  <fct> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 A01nF X       3.5   8    12    19    23    29.5  32.5  36.5  42      50  53  
2 A01nF Y       4.5   9.5  13    17.5  21    28    32.5  36    43.5    48  54.5
3 A01nF Z       1     8.5  13.5  18.5  22.5  28.5  33    37.5  43      49  51.5
4 A01yT X       2.5   8.5  11    19.5  22.5  28    31.5  38    43      50  52.5
5 A01yT Y       2.5   7.5  13.5  17    22    29.5  31    38.5  43.5    49  52.5
6 A01yT Z       3     7    14.5  18    23    28    33    38    43.5    48  54 

这将为您提供每个Trmt，Level的平均值，并且每一列（天）分别是平均值。下一步是在Trmt下定义2个子组（对于A01，A02 ..为nF和yT），为此，我们可以引入一个名为“站点”的子组，该子组是没有nF，yT的Trmt。一旦使用此“站点”和级别对data.frame进行分组，第一行将始终为nF，第二行将始终为yT，因此，以该分组中所有“天”列的差异为准。所以我们这样做：

    # need to ungroup Trmt to remove it later
    tmp <- tmp%>% ungroup(Trmt) %>% 
    mutate(site = sub("[yn][TF]","",Trmt)) %>% 
    select(-Trmt) %>% 
    group_by(site,Level) %>% 
    summarize_all(diff)

现在您有了每种治疗，每种水平和每天的nF-yT值

> head(tmp)
# A tibble: 6 x 19
# Groups:   site [2]
  site  Level  Day1  Day2  Day4  Day7 Day10 Day13 Day18 Day23 Day28 Day35 Day42
  <chr> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 A01   X      -1     0.5  -1     0.5  -0.5  -1.5  -1     1.5   1     0    -0.5
2 A01   Y      -2    -2     0.5  -0.5   1     1.5  -1.5   2.5   0     1    -2  
3 A01   Z       2    -1.5   1    -0.5   0.5  -0.5   0     0.5   0.5  -1     2.5
4 A02   X       1.5   1     1.5   1    -1    -1.5   2    -1.5  -1.5  -1     2  
5 A02   Y       0.5   0    -1.5  -1     0.5   1.5  -0.5  -3    -1.5   0     1  
6 A02   Z       4     2     1     0.5   1.5   0     2.5   0.5   0.5   1.5   0 

最后一部分是绘图。我们将其转换为long，然后也将“ Day”（天的数字形式）制成。

plotdf <- gather(tmp,day,Diff,Day1:Day102,factor_key=TRUE) %>%
mutate(Day=as.numeric(sub("Day",day)))
# and plot

ggplot(plotdf,aes(x=Day,y=Diff,col=Level,shape=Level)) + geom_line() + geom_point() + facet_wrap(~site) + scale_color_manual(values=c("grey10","grey40","grey80"))

上图显示了每个站点的差异。对于差异是所有网站的平均值：

meandf <- plotdf %>% group_by(Level,Day) %>% summarize(Diff=mean(Diff))
ggplot(meandf,shape=Level)) + geom_line() + geom_point() + scale_color_manual(values=c("grey10","grey80"))

示例数据集，为Day1，Day2和Day4子集

Data <- structure(list(Trmt = structure(c(2L,2L,1L,4L,3L,6L,5L,8L,7L,10L,9L,12L,11L,14L,13L,16L,15L,18L,17L,20L,19L,22L,21L,21L
),.Label = c("A01nF","A01yT","A02nF","A02yT","A03nF","A03yT","A04nF","A04yT","A05nF","A05yT","A06nF","A06yT","A07nF","A07yT","A08nF","A08yT","A10nF","A10yT","A11nF","A11yT","A13nF","A13yT"),class = "factor"),Level = structure(c(1L,3L),.Label = c("X","Y","Z"),Day1 = c(3L,4L),Day2 = c(10L,7L),Day4 = c(11L,14L)),class = "data.frame",row.names = c(NA,-132L))