[R] How to add error bars to a line plot with ggplot2?

[Rui Barradas]

Às 17:33 de 14/07/2023, Luigi Marongiu escreveu:
> Hello,
> I am measuring a certain variable at given time intervals and
> different concentrations of a reagent. I would like to make a scatter
> plot of the values, joined by a line to highlight the temporal
> measure.
> I can plot this all right. Now, since I have more than one replicate,
> I would like to add he error bars.
> I prepared  a dataframe with the mean measures and a column with the
> standard deviations, but when I run the code, I get the error:
> ```
> Error in `check_aesthetics()`:
> ! Aesthetics must be either length 1 or the same as the data (20): colour
> Run `rlang::last_trace()` to see where the error occurred.
> ```
> I am missing something, but what?
> Thank you
> 
> 
> WORKING EXAMPLE
> ```
> measTime    = c(1    ,2    ,4    ,24    ,48    ,1    ,2    ,4    ,24
>   ,48    ,1    ,2    ,4    ,24    ,48    ,1    ,2    ,4    ,24    ,48)
> conc    = c(0.25    ,0.25    ,0.25    ,0.25    ,0.25    ,1.12    ,1.12
>     ,1.12    ,1.12    ,1.12    ,2.5    ,2.5    ,2.5    ,2.5    ,2.5
> ,25    ,25    ,25    ,25    ,25)
> varbl    = c(0.0329999999999999    ,0.27    ,0.0785    ,0.1015
> ,-0.193    ,0.048    ,0.113    ,0.1695    ,-0.775    ,0.464    ,-0.257
>     ,-0.154    ,-0.3835    ,-1.23    ,-0.513    ,1.3465    ,1.276
> ,1.128    ,-2.56    ,-1.813)
> stdDev    =c(0.646632301492381    ,0    ,1.77997087991162
> ,0.247683265482349    ,0    ,0.282901631902917    ,0
> ,0.273086677326693    ,1.03807578400295    ,0    ,0.912213425319609
> ,0    ,1.64371621638287    ,2.23203614068709    ,0    ,0.2615396719429
>     ,0    ,0.54039985196149    ,2.15236180353893    ,0)
> df = data.frame(Time=measTime, mM=conc, ddC=varbl, SD=stdDev)
> library(ggplot2)
> COLS = c("green", "red", "blue", "yellow")
>    ggplot(df,
>           aes(x=Time, y=ddC, colour=mM, group=mM)) +
>    geom_line(aes(x=Time, y=ddC, colour=mM, group=mM)) +
>    geom_errorbar(aes(x=Time, ymin=ddC-SD, ymax=ddC+SD, colour=mM, group=mM),
>                  width=.1, colour=COLS) +
>    geom_point(size=6) +
>    scale_colour_manual(values = COLS) +
>    ggtitle("Exposure") +
>    xlab(expression(bold("Time (h)"))) +
>    ylab(expression(bold("Value"))) +
>    geom_hline(aes(yintercept=0)) +
>    theme_classic()
>    ```
> 
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
Hello,

Two notes:

1. If you want to use a discrete colours vector, your 'colour' aesthetic 
must be mapped to a discrete variabe. The most frequent cases are 
character or factor columns.

2. If you start the plot with certain aesthetics set you don't have to 
repeat them in subsequent layers, geom_line can be called with no aes() 
and gem_errorbar doesn't need x=measTime again.


As for the main error, the colors vector COLS should be removed from 
geom_errorbar.



df <- data.frame(Time = measTime,
                  mM = factor(conc),  # this must be a factor
                  ddC = varbl,
                  SD = stdDev)

library(ggplot2)

COLS = c("green", "red", "blue", "yellow")

ggplot(df, aes(x = Time, y = ddC, colour = mM, group = mM)) +
   geom_line() +
   geom_errorbar(aes(ymin = ddC - SD, ymax = ddC + SD), width = 0.1) +
   geom_point(size = 6) +
   geom_hline(aes(yintercept = 0)) +
   scale_colour_manual(values = COLS) +
   ggtitle("Exposure") +
   xlab(expression(bold("Time (h)"))) +
   ylab(expression(bold("Value"))) +
   theme_classic()


Hope this helps,

Rui Barradas